Original Research

The incidence of skin cancer continues to increase, and it is by far the most common malignancy in the United States. Based on the sheer incidence and prevalence of skin cancer, early detection and treatment are critical. Looking at melanoma alone, the 5-year survival rate is greater than 99% when detected early but falls to 71% when the disease reaches the lymph nodes and 32% with metastasis to distant organs.¹ Furthermore, a 2018 study found stage I melanoma patients who were treated 4 months after biopsy had a 41% increased risk of death compared with those treated within the first month.² However, many patients are not seen by a dermatologist first for examination of suspicious skin lesions and instead are referred by a general practitioner or primary care mid-level provider. Therefore, many patients experience a longer time to diagnosis or treatment, which directly correlates with survival rate.

Dermoscopy is a noninvasive diagnostic tool for skin lesions, including melanoma. Using a handheld dermoscope (or dermatoscope), a transilluminating light source magnifies skin lesions and allows for the visualization of subsurface skin structures within the epidermis, dermoepidermal junction, and papillary dermis.³ Dermoscopy has been shown to improve a dermatologist’s accuracy in diagnosing malignant melanoma vs clinical evaluation with the unaided eye.^4,5 More recently, dermoscopy has been digitized, allowing for the collection and documentation of case photographs. Dermoscopy also has expanded past the scope of dermatologists and has become increasingly useful in primary care.⁶ Among family physicians, dermoscopy also has been shown to have a higher sensitivity for melanoma detection compared to gross examination.⁷ Therefore, both the increased diagnostic performance of malignant melanoma using a dermoscope and the expanded use of dermoscopy in medical care validate the evaluation of an artificial intelligence (AI) algorithm in diagnosing malignant melanoma using dermoscopic images.

Triage (Triage Technologies Inc) is an AI application that uses a web interface and combines a pretrained convolutional neural network (CNN) with a reinforcement learning agent as a question-answering model. The CNN algorithm can classify 133 different skin diseases, 7 of which it is able to classify using dermoscopic images. This study sought to evaluate the performance of Triage’s dermoscopic classifier in identifying lesions as benign or malignant to determine whether AI could assist in the triage of skin cancer cases to shorten time to diagnosis.

Materials and Methods

The MClass-D test set from the International Skin Imaging Collaboration was assessed by both AI and practicing medical providers. The set was composed of 80 benign nevi and 20 biopsy-verified malignant melanomas. Board-certified US dermatologists (n=23), family physicians (n=7), and primary care mid-level providers (n=12)(ie, nurse practitioners, physician assistants) were asked to label the images as benign or malignant. The results from the medical providers were then compared to the performance of the AI application by looking at the sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV). Statistical significance was determined with a 1 sample t test run through RStudio (Posit Software, PBC), and P<.05 was considered significant.

Results

The AI application performed extremely well in differentiating between benign nevi and malignant melanomas, with a sensitivity of 80%, specificity of 95%, accuracy of 92%, PPV of 80%, and NPV of 95% (Table 1). When compared with practicing medical providers, the AI performed significantly better in almost all categories (P<.05)(Figure 1). With all medical providers combined, the AI had significantly higher accuracy, sensitivity, and specificity (P<.05). The accuracy of the individual medical providers ranged from 32% to 78%.

Compared with dermatologists, the AI was significantly more specific and accurate and demonstrated a higher PPV and NPV (P<.05). There was no significant difference between the AI and dermatologists in sensitivity or labeling the true malignant lesions as malignant. The dermatologists who participated had been practicing from 1.5 years to 44 years, with an average of 16 years of dermatologic experience. There was no correlation between years practicing and performance in determining the malignancy of lesions. Of 14 dermatologists, dermoscopy was used daily by 10 and occasionally by 3, but only 6 dermatologists had any formal training. Dermatologists who used dermoscopy averaged 11 years of use.

The AI also performed significantly better than the primary care providers, including both family physicians and mid-level providers (P<.05). With the family physicians and mid-level provider scores combined, the AI showed a statistically significantly better performance in all categories examined, including sensitivity, specificity, accuracy, PPV, and NPV (P<.05). However, when compared with family physicians alone, the AI did not demonstrate a statistically significant difference in sensitivity.

Automatic Visual Recognition Development—The AI application we studied was developed by dermatologists as a tool to assist in the screening of skin lesions suspicious for melanoma or a benign neoplasm.⁸ Developing AI applications that can reliably recognize objects in photographs has been the subject of considerable research. Notable progress in automatic visual recognition was shown in 2012 when a deep learning model won the ImageNet object recognition challenge and outperformed competing approaches by a large margin.^9,10 The ImageNet competition, which has been held annually since 2010, required participants to build a visual classification system that distinguished among 1000 object categories using 1.2 million labeled images as training data. In 2017, participants developed automated visual systems that surpassed the estimated human performance.¹¹ Given this success, the organization decided to deliver a more challenging competition involving 3D imaging—Medical ImageNet, a petabyte-scale, cloud-based, open repository project—with goals including image classification and annotation.¹²

Convolutional Neural Networks—Convolutional neural networks are computer system architectures commonly employed for making predictions from images.¹³ Convolutional neural networks are based on a set of layers of learned filters that perform convolution, a mathematical operation that reflects the relationship between the 2 functions. The main algorithm that makes the learning possible is called backpropagation, wherein an error is computed at the output and distributed backward through the neural network’s layers.¹⁴ Although CNNs and backpropagation methods have existed since 1989, recent technologic advances have allowed for deep learning–based algorithms to be widely integrated with everyday applications.¹⁵ Advances in computational power in the form of graphics processing units and parallelization, the existence of large data sets such as the ImageNet database, and the rise of software frameworks have allowed for quick prototyping and deployment of deep learning models.^16,17

Convolutional neural networks have demonstrated potential to excel at a wide range of visual tasks. In dermatology, visual recognition methods often rely on using either a pretrained CNN as a feature extractor for further classification or fine-tuning a pretrained network on dermoscopic images.^18-20 In 2017, a model was trained on 130,000 clinical images of benign and malignant skin lesions. Its performance was found to be in line with that of 21 US board-certified dermatology experts when diagnosing skin cancers from clinical images confirmed by biopsy.²¹

Triage—The AI application Triage is composed of several components contained in a web interface (Figure 2). To use the interface, the user must sign up and upload a photograph to the website. The image first passes through a gated-logic visual classifier that rejects any images that do not contain a visible skin condition. If the image contains a skin condition, the image is passed to a skin classifier that predicts the probability of the image containing 1 of 133 classes of skin conditions, 7 of which the application can diagnose with a dermoscopic image.

Image courtesy of Triage Technologies Inc and Izhaar Tejani, BA (Toronto, Ontario, Canada).

The AI application uses several techniques when training a CNN model. To address skin condition class imbalances (when more examples exist for 1 class than the others) in the training data, additional weights are applied to mistakes made on underrepresented classes, which encourages the model to better detect cases with low prevalence in the data set. Data augmentation techniques such as rotating, zooming, and flipping the training images are applied to allow the model to become more familiar with variability in the input images. Convolutional neural networks are trained using a well-known neural network optimization method called Stochastic gradient descent with momentum.²²

The final predictions are refined by a question-and-answer system that encodes dermatology knowledge and is currently under active development. Finally, the top k most probable conditions are displayed to the user, where k≤5. An initial prototype of the system was described in a published research paper in the 2019 medical imaging workshop of the Neural Information Systems conference.²³

The prototype demonstrated that combining a pretrained CNN with a reinforcement learning agent as a question-answering model increased the classification confidence and accuracy of its visual symptom checker and decreased the average number of questions asked to narrow down the differential diagnosis. The reinforcement learning approach increases the accuracy more than 20% compared with the CNN-only approach, which only uses visual information to predict the condition.²³

This application’s current visual question-answering system is trained on a diverse set of data that includes more than 20 years of clinical encounters and user-uploaded cases submitted by more than 150,000 patients and 10,000 clinicians in more than 150 countries. All crowdsourced images used for training the dermoscopy classifier are biopsy-verified images contributed by dermatologists. These data are made up of case photographs that are tagged with metadata around the patient’s age, sex, symptoms, and diagnoses. The CNN algorithm used covers 133 skin disease classes, representing 588 clinical conditions. It also can automatically detect 7 malignant, premalignant, and benign dermoscopic categories, which is the focus of this study (Table 2). Diagnoses are verified by patient response to treatment, biopsy results, and dermatologist consensus.

In addition to having improved performance, supporting more than 130 disease classes, and having a diverse data set, the application used has beat competing technologies.^20,24 The application currently is available on the internet in more than 30 countries after it received Health Canada Class I medical device approval and the CE mark in Europe.

Can AI Reliably Detect Melanoma?—In our study, of the lesions labeled benign, the higher PPV and NPV of the AI algorithm means that the lesions were more reliably true benign lesions, and the lesions labeled as malignant were more likely to be true malignant lesions. Therefore, the diagnosis given by the AI compared with the medical provider was significantly more likely to be correct. These findings demonstrate that this AI application can reliably detect malignant melanoma using dermoscopic images. However, this study was limited by the small sample size of medical providers. Further studies are necessary to assess whether the high diagnostic accuracy of the application translates to expedited referrals and a decrease in unnecessary biopsies.

Dermoscopy Training—This study looked at dermoscopic images instead of gross examination, as is often done in clinic, which draws into question the dermoscopic training dermatologists receive. The diagnostic accuracy using dermoscopic images has been shown to be higher than evaluation with the naked eye.^5,6 However, there currently is no standard for dermoscopic training in dermatology residencies, and education varies widely.²⁵ These data suggest that there may be a lack of dermoscopic training among dermatologists, which could accentuate the difference in performance between dermatologists and AI. Most primary care providers also lack formal dermoscopy training. Although dermoscopy has been shown to increase the diagnostic efficacy of primary care providers, this increase does not become apparent until the medical provider has had years of formal training in addition to clinical experience, which is not commonly provided in the medical training that primary care providers receive.^8,26

Conclusion

It is anticipated that AI will shape the future of medicine and become incorporated into daily practice.²⁷ Artificial intelligence will not replace physicians but rather assist clinicians and help to streamline medical care. Clinicians will take on the role of interpreting AI output and integrate it into patient care. With this advancement, it is important to highlight that for AI to improve the quality, efficiency, and accessibility of health care, clinicians must be equipped with the right training.^27-29

The incidence of skin cancer continues to increase, and it is by far the most common malignancy in the United States. Based on the sheer incidence and prevalence of skin cancer, early detection and treatment are critical. Looking at melanoma alone, the 5-year survival rate is greater than 99% when detected early but falls to 71% when the disease reaches the lymph nodes and 32% with metastasis to distant organs.¹ Furthermore, a 2018 study found stage I melanoma patients who were treated 4 months after biopsy had a 41% increased risk of death compared with those treated within the first month.² However, many patients are not seen by a dermatologist first for examination of suspicious skin lesions and instead are referred by a general practitioner or primary care mid-level provider. Therefore, many patients experience a longer time to diagnosis or treatment, which directly correlates with survival rate.

Dermoscopy is a noninvasive diagnostic tool for skin lesions, including melanoma. Using a handheld dermoscope (or dermatoscope), a transilluminating light source magnifies skin lesions and allows for the visualization of subsurface skin structures within the epidermis, dermoepidermal junction, and papillary dermis.³ Dermoscopy has been shown to improve a dermatologist’s accuracy in diagnosing malignant melanoma vs clinical evaluation with the unaided eye.^4,5 More recently, dermoscopy has been digitized, allowing for the collection and documentation of case photographs. Dermoscopy also has expanded past the scope of dermatologists and has become increasingly useful in primary care.⁶ Among family physicians, dermoscopy also has been shown to have a higher sensitivity for melanoma detection compared to gross examination.⁷ Therefore, both the increased diagnostic performance of malignant melanoma using a dermoscope and the expanded use of dermoscopy in medical care validate the evaluation of an artificial intelligence (AI) algorithm in diagnosing malignant melanoma using dermoscopic images.

Triage (Triage Technologies Inc) is an AI application that uses a web interface and combines a pretrained convolutional neural network (CNN) with a reinforcement learning agent as a question-answering model. The CNN algorithm can classify 133 different skin diseases, 7 of which it is able to classify using dermoscopic images. This study sought to evaluate the performance of Triage’s dermoscopic classifier in identifying lesions as benign or malignant to determine whether AI could assist in the triage of skin cancer cases to shorten time to diagnosis.

Materials and Methods

The MClass-D test set from the International Skin Imaging Collaboration was assessed by both AI and practicing medical providers. The set was composed of 80 benign nevi and 20 biopsy-verified malignant melanomas. Board-certified US dermatologists (n=23), family physicians (n=7), and primary care mid-level providers (n=12)(ie, nurse practitioners, physician assistants) were asked to label the images as benign or malignant. The results from the medical providers were then compared to the performance of the AI application by looking at the sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV). Statistical significance was determined with a 1 sample t test run through RStudio (Posit Software, PBC), and P<.05 was considered significant.

Results

The AI application performed extremely well in differentiating between benign nevi and malignant melanomas, with a sensitivity of 80%, specificity of 95%, accuracy of 92%, PPV of 80%, and NPV of 95% (Table 1). When compared with practicing medical providers, the AI performed significantly better in almost all categories (P<.05)(Figure 1). With all medical providers combined, the AI had significantly higher accuracy, sensitivity, and specificity (P<.05). The accuracy of the individual medical providers ranged from 32% to 78%.

Compared with dermatologists, the AI was significantly more specific and accurate and demonstrated a higher PPV and NPV (P<.05). There was no significant difference between the AI and dermatologists in sensitivity or labeling the true malignant lesions as malignant. The dermatologists who participated had been practicing from 1.5 years to 44 years, with an average of 16 years of dermatologic experience. There was no correlation between years practicing and performance in determining the malignancy of lesions. Of 14 dermatologists, dermoscopy was used daily by 10 and occasionally by 3, but only 6 dermatologists had any formal training. Dermatologists who used dermoscopy averaged 11 years of use.

The AI also performed significantly better than the primary care providers, including both family physicians and mid-level providers (P<.05). With the family physicians and mid-level provider scores combined, the AI showed a statistically significantly better performance in all categories examined, including sensitivity, specificity, accuracy, PPV, and NPV (P<.05). However, when compared with family physicians alone, the AI did not demonstrate a statistically significant difference in sensitivity.

Automatic Visual Recognition Development—The AI application we studied was developed by dermatologists as a tool to assist in the screening of skin lesions suspicious for melanoma or a benign neoplasm.⁸ Developing AI applications that can reliably recognize objects in photographs has been the subject of considerable research. Notable progress in automatic visual recognition was shown in 2012 when a deep learning model won the ImageNet object recognition challenge and outperformed competing approaches by a large margin.^9,10 The ImageNet competition, which has been held annually since 2010, required participants to build a visual classification system that distinguished among 1000 object categories using 1.2 million labeled images as training data. In 2017, participants developed automated visual systems that surpassed the estimated human performance.¹¹ Given this success, the organization decided to deliver a more challenging competition involving 3D imaging—Medical ImageNet, a petabyte-scale, cloud-based, open repository project—with goals including image classification and annotation.¹²

Convolutional Neural Networks—Convolutional neural networks are computer system architectures commonly employed for making predictions from images.¹³ Convolutional neural networks are based on a set of layers of learned filters that perform convolution, a mathematical operation that reflects the relationship between the 2 functions. The main algorithm that makes the learning possible is called backpropagation, wherein an error is computed at the output and distributed backward through the neural network’s layers.¹⁴ Although CNNs and backpropagation methods have existed since 1989, recent technologic advances have allowed for deep learning–based algorithms to be widely integrated with everyday applications.¹⁵ Advances in computational power in the form of graphics processing units and parallelization, the existence of large data sets such as the ImageNet database, and the rise of software frameworks have allowed for quick prototyping and deployment of deep learning models.^16,17

Convolutional neural networks have demonstrated potential to excel at a wide range of visual tasks. In dermatology, visual recognition methods often rely on using either a pretrained CNN as a feature extractor for further classification or fine-tuning a pretrained network on dermoscopic images.^18-20 In 2017, a model was trained on 130,000 clinical images of benign and malignant skin lesions. Its performance was found to be in line with that of 21 US board-certified dermatology experts when diagnosing skin cancers from clinical images confirmed by biopsy.²¹

Triage—The AI application Triage is composed of several components contained in a web interface (Figure 2). To use the interface, the user must sign up and upload a photograph to the website. The image first passes through a gated-logic visual classifier that rejects any images that do not contain a visible skin condition. If the image contains a skin condition, the image is passed to a skin classifier that predicts the probability of the image containing 1 of 133 classes of skin conditions, 7 of which the application can diagnose with a dermoscopic image.

Image courtesy of Triage Technologies Inc and Izhaar Tejani, BA (Toronto, Ontario, Canada).

The AI application uses several techniques when training a CNN model. To address skin condition class imbalances (when more examples exist for 1 class than the others) in the training data, additional weights are applied to mistakes made on underrepresented classes, which encourages the model to better detect cases with low prevalence in the data set. Data augmentation techniques such as rotating, zooming, and flipping the training images are applied to allow the model to become more familiar with variability in the input images. Convolutional neural networks are trained using a well-known neural network optimization method called Stochastic gradient descent with momentum.²²

The final predictions are refined by a question-and-answer system that encodes dermatology knowledge and is currently under active development. Finally, the top k most probable conditions are displayed to the user, where k≤5. An initial prototype of the system was described in a published research paper in the 2019 medical imaging workshop of the Neural Information Systems conference.²³

The prototype demonstrated that combining a pretrained CNN with a reinforcement learning agent as a question-answering model increased the classification confidence and accuracy of its visual symptom checker and decreased the average number of questions asked to narrow down the differential diagnosis. The reinforcement learning approach increases the accuracy more than 20% compared with the CNN-only approach, which only uses visual information to predict the condition.²³

This application’s current visual question-answering system is trained on a diverse set of data that includes more than 20 years of clinical encounters and user-uploaded cases submitted by more than 150,000 patients and 10,000 clinicians in more than 150 countries. All crowdsourced images used for training the dermoscopy classifier are biopsy-verified images contributed by dermatologists. These data are made up of case photographs that are tagged with metadata around the patient’s age, sex, symptoms, and diagnoses. The CNN algorithm used covers 133 skin disease classes, representing 588 clinical conditions. It also can automatically detect 7 malignant, premalignant, and benign dermoscopic categories, which is the focus of this study (Table 2). Diagnoses are verified by patient response to treatment, biopsy results, and dermatologist consensus.

In addition to having improved performance, supporting more than 130 disease classes, and having a diverse data set, the application used has beat competing technologies.^20,24 The application currently is available on the internet in more than 30 countries after it received Health Canada Class I medical device approval and the CE mark in Europe.

Can AI Reliably Detect Melanoma?—In our study, of the lesions labeled benign, the higher PPV and NPV of the AI algorithm means that the lesions were more reliably true benign lesions, and the lesions labeled as malignant were more likely to be true malignant lesions. Therefore, the diagnosis given by the AI compared with the medical provider was significantly more likely to be correct. These findings demonstrate that this AI application can reliably detect malignant melanoma using dermoscopic images. However, this study was limited by the small sample size of medical providers. Further studies are necessary to assess whether the high diagnostic accuracy of the application translates to expedited referrals and a decrease in unnecessary biopsies.

Dermoscopy Training—This study looked at dermoscopic images instead of gross examination, as is often done in clinic, which draws into question the dermoscopic training dermatologists receive. The diagnostic accuracy using dermoscopic images has been shown to be higher than evaluation with the naked eye.^5,6 However, there currently is no standard for dermoscopic training in dermatology residencies, and education varies widely.²⁵ These data suggest that there may be a lack of dermoscopic training among dermatologists, which could accentuate the difference in performance between dermatologists and AI. Most primary care providers also lack formal dermoscopy training. Although dermoscopy has been shown to increase the diagnostic efficacy of primary care providers, this increase does not become apparent until the medical provider has had years of formal training in addition to clinical experience, which is not commonly provided in the medical training that primary care providers receive.^8,26

Conclusion

It is anticipated that AI will shape the future of medicine and become incorporated into daily practice.²⁷ Artificial intelligence will not replace physicians but rather assist clinicians and help to streamline medical care. Clinicians will take on the role of interpreting AI output and integrate it into patient care. With this advancement, it is important to highlight that for AI to improve the quality, efficiency, and accessibility of health care, clinicians must be equipped with the right training.^27-29

The incidence of skin cancer continues to increase, and it is by far the most common malignancy in the United States. Based on the sheer incidence and prevalence of skin cancer, early detection and treatment are critical. Looking at melanoma alone, the 5-year survival rate is greater than 99% when detected early but falls to 71% when the disease reaches the lymph nodes and 32% with metastasis to distant organs.¹ Furthermore, a 2018 study found stage I melanoma patients who were treated 4 months after biopsy had a 41% increased risk of death compared with those treated within the first month.² However, many patients are not seen by a dermatologist first for examination of suspicious skin lesions and instead are referred by a general practitioner or primary care mid-level provider. Therefore, many patients experience a longer time to diagnosis or treatment, which directly correlates with survival rate.

Dermoscopy is a noninvasive diagnostic tool for skin lesions, including melanoma. Using a handheld dermoscope (or dermatoscope), a transilluminating light source magnifies skin lesions and allows for the visualization of subsurface skin structures within the epidermis, dermoepidermal junction, and papillary dermis.³ Dermoscopy has been shown to improve a dermatologist’s accuracy in diagnosing malignant melanoma vs clinical evaluation with the unaided eye.^4,5 More recently, dermoscopy has been digitized, allowing for the collection and documentation of case photographs. Dermoscopy also has expanded past the scope of dermatologists and has become increasingly useful in primary care.⁶ Among family physicians, dermoscopy also has been shown to have a higher sensitivity for melanoma detection compared to gross examination.⁷ Therefore, both the increased diagnostic performance of malignant melanoma using a dermoscope and the expanded use of dermoscopy in medical care validate the evaluation of an artificial intelligence (AI) algorithm in diagnosing malignant melanoma using dermoscopic images.

Triage (Triage Technologies Inc) is an AI application that uses a web interface and combines a pretrained convolutional neural network (CNN) with a reinforcement learning agent as a question-answering model. The CNN algorithm can classify 133 different skin diseases, 7 of which it is able to classify using dermoscopic images. This study sought to evaluate the performance of Triage’s dermoscopic classifier in identifying lesions as benign or malignant to determine whether AI could assist in the triage of skin cancer cases to shorten time to diagnosis.

Materials and Methods

The MClass-D test set from the International Skin Imaging Collaboration was assessed by both AI and practicing medical providers. The set was composed of 80 benign nevi and 20 biopsy-verified malignant melanomas. Board-certified US dermatologists (n=23), family physicians (n=7), and primary care mid-level providers (n=12)(ie, nurse practitioners, physician assistants) were asked to label the images as benign or malignant. The results from the medical providers were then compared to the performance of the AI application by looking at the sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV). Statistical significance was determined with a 1 sample t test run through RStudio (Posit Software, PBC), and P<.05 was considered significant.

Results

The AI application performed extremely well in differentiating between benign nevi and malignant melanomas, with a sensitivity of 80%, specificity of 95%, accuracy of 92%, PPV of 80%, and NPV of 95% (Table 1). When compared with practicing medical providers, the AI performed significantly better in almost all categories (P<.05)(Figure 1). With all medical providers combined, the AI had significantly higher accuracy, sensitivity, and specificity (P<.05). The accuracy of the individual medical providers ranged from 32% to 78%.

Compared with dermatologists, the AI was significantly more specific and accurate and demonstrated a higher PPV and NPV (P<.05). There was no significant difference between the AI and dermatologists in sensitivity or labeling the true malignant lesions as malignant. The dermatologists who participated had been practicing from 1.5 years to 44 years, with an average of 16 years of dermatologic experience. There was no correlation between years practicing and performance in determining the malignancy of lesions. Of 14 dermatologists, dermoscopy was used daily by 10 and occasionally by 3, but only 6 dermatologists had any formal training. Dermatologists who used dermoscopy averaged 11 years of use.

The AI also performed significantly better than the primary care providers, including both family physicians and mid-level providers (P<.05). With the family physicians and mid-level provider scores combined, the AI showed a statistically significantly better performance in all categories examined, including sensitivity, specificity, accuracy, PPV, and NPV (P<.05). However, when compared with family physicians alone, the AI did not demonstrate a statistically significant difference in sensitivity.

Automatic Visual Recognition Development—The AI application we studied was developed by dermatologists as a tool to assist in the screening of skin lesions suspicious for melanoma or a benign neoplasm.⁸ Developing AI applications that can reliably recognize objects in photographs has been the subject of considerable research. Notable progress in automatic visual recognition was shown in 2012 when a deep learning model won the ImageNet object recognition challenge and outperformed competing approaches by a large margin.^9,10 The ImageNet competition, which has been held annually since 2010, required participants to build a visual classification system that distinguished among 1000 object categories using 1.2 million labeled images as training data. In 2017, participants developed automated visual systems that surpassed the estimated human performance.¹¹ Given this success, the organization decided to deliver a more challenging competition involving 3D imaging—Medical ImageNet, a petabyte-scale, cloud-based, open repository project—with goals including image classification and annotation.¹²

Convolutional Neural Networks—Convolutional neural networks are computer system architectures commonly employed for making predictions from images.¹³ Convolutional neural networks are based on a set of layers of learned filters that perform convolution, a mathematical operation that reflects the relationship between the 2 functions. The main algorithm that makes the learning possible is called backpropagation, wherein an error is computed at the output and distributed backward through the neural network’s layers.¹⁴ Although CNNs and backpropagation methods have existed since 1989, recent technologic advances have allowed for deep learning–based algorithms to be widely integrated with everyday applications.¹⁵ Advances in computational power in the form of graphics processing units and parallelization, the existence of large data sets such as the ImageNet database, and the rise of software frameworks have allowed for quick prototyping and deployment of deep learning models.^16,17

Convolutional neural networks have demonstrated potential to excel at a wide range of visual tasks. In dermatology, visual recognition methods often rely on using either a pretrained CNN as a feature extractor for further classification or fine-tuning a pretrained network on dermoscopic images.^18-20 In 2017, a model was trained on 130,000 clinical images of benign and malignant skin lesions. Its performance was found to be in line with that of 21 US board-certified dermatology experts when diagnosing skin cancers from clinical images confirmed by biopsy.²¹

Triage—The AI application Triage is composed of several components contained in a web interface (Figure 2). To use the interface, the user must sign up and upload a photograph to the website. The image first passes through a gated-logic visual classifier that rejects any images that do not contain a visible skin condition. If the image contains a skin condition, the image is passed to a skin classifier that predicts the probability of the image containing 1 of 133 classes of skin conditions, 7 of which the application can diagnose with a dermoscopic image.

Image courtesy of Triage Technologies Inc and Izhaar Tejani, BA (Toronto, Ontario, Canada).

The AI application uses several techniques when training a CNN model. To address skin condition class imbalances (when more examples exist for 1 class than the others) in the training data, additional weights are applied to mistakes made on underrepresented classes, which encourages the model to better detect cases with low prevalence in the data set. Data augmentation techniques such as rotating, zooming, and flipping the training images are applied to allow the model to become more familiar with variability in the input images. Convolutional neural networks are trained using a well-known neural network optimization method called Stochastic gradient descent with momentum.²²

The final predictions are refined by a question-and-answer system that encodes dermatology knowledge and is currently under active development. Finally, the top k most probable conditions are displayed to the user, where k≤5. An initial prototype of the system was described in a published research paper in the 2019 medical imaging workshop of the Neural Information Systems conference.²³

The prototype demonstrated that combining a pretrained CNN with a reinforcement learning agent as a question-answering model increased the classification confidence and accuracy of its visual symptom checker and decreased the average number of questions asked to narrow down the differential diagnosis. The reinforcement learning approach increases the accuracy more than 20% compared with the CNN-only approach, which only uses visual information to predict the condition.²³

This application’s current visual question-answering system is trained on a diverse set of data that includes more than 20 years of clinical encounters and user-uploaded cases submitted by more than 150,000 patients and 10,000 clinicians in more than 150 countries. All crowdsourced images used for training the dermoscopy classifier are biopsy-verified images contributed by dermatologists. These data are made up of case photographs that are tagged with metadata around the patient’s age, sex, symptoms, and diagnoses. The CNN algorithm used covers 133 skin disease classes, representing 588 clinical conditions. It also can automatically detect 7 malignant, premalignant, and benign dermoscopic categories, which is the focus of this study (Table 2). Diagnoses are verified by patient response to treatment, biopsy results, and dermatologist consensus.

In addition to having improved performance, supporting more than 130 disease classes, and having a diverse data set, the application used has beat competing technologies.^20,24 The application currently is available on the internet in more than 30 countries after it received Health Canada Class I medical device approval and the CE mark in Europe.

Can AI Reliably Detect Melanoma?—In our study, of the lesions labeled benign, the higher PPV and NPV of the AI algorithm means that the lesions were more reliably true benign lesions, and the lesions labeled as malignant were more likely to be true malignant lesions. Therefore, the diagnosis given by the AI compared with the medical provider was significantly more likely to be correct. These findings demonstrate that this AI application can reliably detect malignant melanoma using dermoscopic images. However, this study was limited by the small sample size of medical providers. Further studies are necessary to assess whether the high diagnostic accuracy of the application translates to expedited referrals and a decrease in unnecessary biopsies.

Dermoscopy Training—This study looked at dermoscopic images instead of gross examination, as is often done in clinic, which draws into question the dermoscopic training dermatologists receive. The diagnostic accuracy using dermoscopic images has been shown to be higher than evaluation with the naked eye.^5,6 However, there currently is no standard for dermoscopic training in dermatology residencies, and education varies widely.²⁵ These data suggest that there may be a lack of dermoscopic training among dermatologists, which could accentuate the difference in performance between dermatologists and AI. Most primary care providers also lack formal dermoscopy training. Although dermoscopy has been shown to increase the diagnostic efficacy of primary care providers, this increase does not become apparent until the medical provider has had years of formal training in addition to clinical experience, which is not commonly provided in the medical training that primary care providers receive.^8,26

Conclusion

It is anticipated that AI will shape the future of medicine and become incorporated into daily practice.²⁷ Artificial intelligence will not replace physicians but rather assist clinicians and help to streamline medical care. Clinicians will take on the role of interpreting AI output and integrate it into patient care. With this advancement, it is important to highlight that for AI to improve the quality, efficiency, and accessibility of health care, clinicians must be equipped with the right training.^27-29

Amyotrophic lateral sclerosis (ALS) is an incurable neurodegenerative disorder that results in progressive deterioration of motor neurons in the ventral horn of the spinal cord, which results in loss of voluntary muscle movements.¹ Eventually, typical daily tasks become difficult to perform, and as the disease progresses, the ability to eat and breathe is impaired.² Reports from 2015 show the annual incidence of ALS is 5 cases per 100,000 people, with the total number of cases reported at more than 16,000 in the United States.³ In clinical practice, disease progression is routinely assessed by the Revised Amyotrophic Lateral Sclerosis Functional Rating Scale (ALSFRS-R). Typical decline is 1 point per month.⁴

Unfortunately, at this time, ALS care focuses on symptom management, including prevention of weight loss; implementation of communication strategies; and management of pain, constipation, excess secretions, cramping, and breathing. Despite copious research into treatment options, few exist. Riluzole is an oral medication administered twice daily and has been on the market since 1995.^5-7 Efficacy was demonstrated in a study showing statistically significant survival at 12 months compared with controls (74% vs 58%, respectively; P = .014).⁶ Since its approval, riluzole has become part of standard-of-care ALS management.

In 2017, the US Food and Drug Administration (FDA) approved edaravone, an IV medication that was found to slow the progression of ALS in some patients.^8-12 Oxidative stress caused by free radicals is hypothesized to increase the progression of ALS by motor neuron degradation.¹³ Edaravone works as a free radical and peroxynitrite scavenger and has been shown to eliminate lipid peroxides and hydroxyl radicals known to damage endothelial and neuronal cells.¹²

Given the mechanism of action of edaravone, it seemed to be a promising option to slow the progression of ALS. A 2019 systematic review analyzed 3 randomized studies with 367 patients and found a statistically significant difference in change in ALSFRS-R scores between patients treated with edaravone for 24 weeks compared with patients treated with the placebo (mean difference, 1.63; 95% CI, 0.26-3.00; P = .02).¹² Secondary endpoints evaluated included percent forced vital capacity (%FVC), grip strength, and pinch strength: All showing no significant difference when comparing IV edaravone with placebo.

A 2022 postmarketing study of 324 patients with ALS evaluated the safety and efficacy of long-term edaravone treatment. IV edaravone therapy for > 24 weeks was well tolerated, although it was not associated with any disease-modifying benefit when comparing ALSFRS-R scores with patients not receiving edaravone over a median 13.9 months (ALSFRS-R points/month, -0.91 vs -0.85; P = .37).¹³ A third ALS treatment medication, sodium phenylbutyrate/taurursodiol was approved in 2022 but not available during our study period and not included here.^14,15

Studies have shown an increased incidence of ALS in the veteran population. Veterans serving in the Gulf War were nearly twice as likely to develop ALS as those not serving in the Gulf.¹⁶ However, existing literature regarding the effectiveness of edaravone does not specifically examine the effect on this unique population. The objective of this study was to assess the effect of IV edaravone on ALS progression in veterans compared with veterans who received standard of care.

Methods

This study was conducted at a large, academic US Department of Veterans Affairs (VA) medical center. Patients with ALS are followed by a multidisciplinary clinic composed of a neurologist, pulmonologist, clinical pharmacist, social worker, speech therapist, physical therapist, occupational therapist, dietician, clinical psychologist, wheelchair clinic representative, and benefits representative. Patients are typically seen for a half-day appointment about every 3 months. During these visits, a comprehensive review of disease progression is performed. This review entails completion of the ALSFRS-R, physical examination, and pulmonary function testing. Speech intelligibility stage (SIS) is assessed by a speech therapist as well. SIS is scored from 1 (no detectable speech disorder) to 5 (no functional speech). All patients followed in this multidisciplinary ALS clinic receive standard-of-care treatment. This includes the discussion of treatment options that if appropriate are provided to help manage a wide range of complications associated with this disease (eg, pain, cramping, constipation, excessive secretions, weight loss, dysphagia). As a part of these personal discussions, treatment with riluzole is also offered as a standard-of-care pharmacologic option.

Study Design

This retrospective case-control study was conducted using electronic health record data to compare ALS progression in patients on IV edaravone therapy with standard of care. The Indiana University/Purdue University, Indianapolis Institutional Review Board and the VA Research and Development Committee approved the study. The control cohort received the standard of care. Patients in the case cohort received standard of care and edaravone 60 mg infusions daily for an initial cycle of 14 days on treatment, followed by 14 days off. All subsequent cycles were 10 of 14 days on treatment followed by 14 days off. The initial 2 doses were administered in the outpatient infusion clinic to monitor for a hypersensitivity reaction. Patients then had a peripherally inserted central catheter line placed and received doses on days 3 through 14 at home. A port was placed for subsequent cycles, which were also completed at home. Appropriateness of edaravone therapy was assessed by the neurologist at each follow-up appointment. Therapy was then discontinued if warranted based on disease progression or patient preference.

Study Population

Patients included were aged 18 to 75 years with diagnosed ALS. Patients with complications that might influence evaluation of medication efficacy (eg, Parkinson disease, schizophrenia, significant dementia, other major medical morbidity) were excluded. Patients were also excluded if they were on continuous bilevel positive airway pressure and/or had a total score of ≤ 3 points on ALSFRS-R items for dyspnea, orthopnea, or respiratory insufficiency. Due to our small sample size, patients were excluded if treatment was < 6 months, which is the gold standard of therapy duration established by clinical trials.^9,11,12

The standard-of-care cohort included patients enrolled in the multidisciplinary clinic September 1, 2014 to August 31, 2017. These patients were compared in a 2:1 ratio with patients who received IV edaravone. The edaravone cohort included patients who initiated treatment with IV edaravone between September 1, 2017, and August 31, 2020. This date range prior to the approval of edaravone was chosen to compare patients at similar stages of disease progression and to have the largest sample size possible.

Data Collection

Data were obtained for eligible patients using the VA Computerized Patient Record System. Demographic data gathered for each patient included age, sex, weight, height, body mass index (BMI), race, and riluzole use.

The primary endpoint was the change in ALSFRS-R score after 6 months of IV edaravone compared with standard-of-care ALS management. Secondary outcomes included change in ALSFRS-R scores 3, 12, 18, and 24 months after therapy initiation, change in %FVC and SIS 3, 6, 12, 18, and 24 months after therapy initiation, duration of edaravone completed (months), time to death (months), and adverse events.

Statistical Analysis

Comparisons between the edaravone and control groups for differences in patient characteristics were made using χ² and 2-sample t tests for categorical and continuous variables, respectively. Comparisons between the 2 groups for differences in study outcomes (ALSFRS-R scores, %FVC, SIS) at each time point were evaluated using 2-sample t tests. Adverse events and adverse drug reactions were compared between groups using χ² tests. Statistical significance was set at 0.05.

We estimated that a sample size of 21 subjects in the edaravone (case) group and 42 in the standard-of-care (control) group would be needed to achieve 80% power to detect a difference of 6.5 between the 2 groups for the change in ALSFRS-R scores. This 80% power was calculated based on a 2-sample t test, and assuming a 2-sided 5% significance level and a within-group SD of 8.5.⁹ Statistical analysis was conducted using Microsoft Excel.

Results

Of the 96 patients, 10 met exclusion criteria. From the remaining 86, 42 were randomly selected for the standard-of-care group. A total of 27 patients seen in multidisciplinary ALS clinic between September 1, 2017, and August 31, 2020, received at least 1 dose of IV edaravone. Of the 27 edaravone patients, 6 were excluded for not completing a total of 6 months of edaravone. Two of the 6 excluded developed a rash, which resolved within 1 week after discontinuing edaravone. The other 4 discontinued edaravone before 6 months because of disease progression.

Baseline Characteristics

Efficacy

Discussion

This 24-month, case-control retrospective study assessed efficacy and safety of IV edaravone for the management of ALS. Although the landmark edaravone study showed slowed progression of ALS at 6 and 12 months, the effectiveness of edaravone outside the clinical trial setting has been less compelling.^9-11,13 A later study showed no difference in change in ALSFRS-R score at 6 months compared with that of the placebo group.⁷ In our study, no statistically significant difference was found for change in ALSFRS-R scores at 6 months.

Our study was unique given we evaluated a veteran population. The link between the military and ALS is largely unknown, although studies have shown increased incidence of ALS in people with a military history compared with that of the general population.^16-18 Our study was also unique because it was single-centered in design and allowed for outcome assessments, including ALSFRS-R scores, SIS, and %FVC measurements, to all be conducted by the same practitioner to limit variability. Unfortunately, our sample size resulted in a cohort that was underpowered at 12, 18, and 24 months. In addition, there was a lack of data on chart review for SIS and %FVC measurements at 24 months. As ALS progresses toward end stage, SIS and %FVC measurements can become difficult and burdensome on the patient to obtain, and the ALS multidisciplinary team may decide not to gather these data points as ALS progresses. As a result, change in SIS and %FVC measurements were unable to be reported due to lack of gathering this information at the 24-month mark in the edaravone group. Due to the cost and administration burden associated with edaravone, it is important that assessment of disease progression is performed regularly to assess benefit and appropriateness of continued therapy. The oral formulation of edaravone was approved in 2022, shortly after the completion of data collection for this study.^19,20 Although our study did not analyze oral edaravone, the administration burden of treatment would be reduced with the oral formulation, and we hypothesize there will be increased patient interest in ALS management with oral vs IV edaravone. Evaluation of long-term treatment for efficacy and safety beyond 24 months has not been evaluated. Future studies should continue to evaluate edaravone use in a larger veteran population.

Limitations

One limitation for our study alluded to earlier in the discussion was sample size. Although this study met power at the 6-month mark, it was limited by the number of patients who received more than 6 months of edaravone (n = 21). As a result, statistical analyses between treatment groups were underpowered at 12, 18, and 24 months. Our study had 80% power to detect a difference of 6.5 between the groups for the change in ALSFRS-R scores. Previous studies detected a statistically significant difference in ALSFRS-R scores, with a difference in ALSFRS-R scores of 2.49 between groups.⁸ Future studies should evaluate a larger sample size of patients who are prescribed edaravone.

Another limitation was that the edaravone and standard-of-care group data were gathered from different time periods. Two different time frames were selected to increase sample size by gathering data over a longer period and to account for patients who may have qualified for IV edaravone but could not receive it as it was not yet available on the market. There were no known changes to the standard of care between the time periods that would affect results. As noted previously, the standard-of-care group had fewer patients taking riluzole compared with the edaravone group, which may have confounded our results. We concluded patients opting for edaravone were more likely to trial riluzole, taken by mouth twice daily, before starting edaravone, a once-daily IV infusion.

Conclusions

No difference in the rate of ALS progression was noted between patients who received IV edaravone vs standard of care at 6 months. In addition, no difference was noted in other objective measures of disease progression, including %FVC, SIS, and time to death. As a result, the decision to initiate and continue edaravone therapy should be made on an individualized basis according to a prescriber’s clinical judgment and a patient’s goals. Edaravone therapy should be discontinued when disease progression occurs or when medication administration becomes a burden.

Acknowledgments

This material is the result of work supported with resources and the use of facilities at Veteran Health Indiana.

Amyotrophic lateral sclerosis (ALS) is an incurable neurodegenerative disorder that results in progressive deterioration of motor neurons in the ventral horn of the spinal cord, which results in loss of voluntary muscle movements.¹ Eventually, typical daily tasks become difficult to perform, and as the disease progresses, the ability to eat and breathe is impaired.² Reports from 2015 show the annual incidence of ALS is 5 cases per 100,000 people, with the total number of cases reported at more than 16,000 in the United States.³ In clinical practice, disease progression is routinely assessed by the Revised Amyotrophic Lateral Sclerosis Functional Rating Scale (ALSFRS-R). Typical decline is 1 point per month.⁴

Unfortunately, at this time, ALS care focuses on symptom management, including prevention of weight loss; implementation of communication strategies; and management of pain, constipation, excess secretions, cramping, and breathing. Despite copious research into treatment options, few exist. Riluzole is an oral medication administered twice daily and has been on the market since 1995.^5-7 Efficacy was demonstrated in a study showing statistically significant survival at 12 months compared with controls (74% vs 58%, respectively; P = .014).⁶ Since its approval, riluzole has become part of standard-of-care ALS management.

In 2017, the US Food and Drug Administration (FDA) approved edaravone, an IV medication that was found to slow the progression of ALS in some patients.^8-12 Oxidative stress caused by free radicals is hypothesized to increase the progression of ALS by motor neuron degradation.¹³ Edaravone works as a free radical and peroxynitrite scavenger and has been shown to eliminate lipid peroxides and hydroxyl radicals known to damage endothelial and neuronal cells.¹²

Given the mechanism of action of edaravone, it seemed to be a promising option to slow the progression of ALS. A 2019 systematic review analyzed 3 randomized studies with 367 patients and found a statistically significant difference in change in ALSFRS-R scores between patients treated with edaravone for 24 weeks compared with patients treated with the placebo (mean difference, 1.63; 95% CI, 0.26-3.00; P = .02).¹² Secondary endpoints evaluated included percent forced vital capacity (%FVC), grip strength, and pinch strength: All showing no significant difference when comparing IV edaravone with placebo.

A 2022 postmarketing study of 324 patients with ALS evaluated the safety and efficacy of long-term edaravone treatment. IV edaravone therapy for > 24 weeks was well tolerated, although it was not associated with any disease-modifying benefit when comparing ALSFRS-R scores with patients not receiving edaravone over a median 13.9 months (ALSFRS-R points/month, -0.91 vs -0.85; P = .37).¹³ A third ALS treatment medication, sodium phenylbutyrate/taurursodiol was approved in 2022 but not available during our study period and not included here.^14,15

Studies have shown an increased incidence of ALS in the veteran population. Veterans serving in the Gulf War were nearly twice as likely to develop ALS as those not serving in the Gulf.¹⁶ However, existing literature regarding the effectiveness of edaravone does not specifically examine the effect on this unique population. The objective of this study was to assess the effect of IV edaravone on ALS progression in veterans compared with veterans who received standard of care.

Methods

This study was conducted at a large, academic US Department of Veterans Affairs (VA) medical center. Patients with ALS are followed by a multidisciplinary clinic composed of a neurologist, pulmonologist, clinical pharmacist, social worker, speech therapist, physical therapist, occupational therapist, dietician, clinical psychologist, wheelchair clinic representative, and benefits representative. Patients are typically seen for a half-day appointment about every 3 months. During these visits, a comprehensive review of disease progression is performed. This review entails completion of the ALSFRS-R, physical examination, and pulmonary function testing. Speech intelligibility stage (SIS) is assessed by a speech therapist as well. SIS is scored from 1 (no detectable speech disorder) to 5 (no functional speech). All patients followed in this multidisciplinary ALS clinic receive standard-of-care treatment. This includes the discussion of treatment options that if appropriate are provided to help manage a wide range of complications associated with this disease (eg, pain, cramping, constipation, excessive secretions, weight loss, dysphagia). As a part of these personal discussions, treatment with riluzole is also offered as a standard-of-care pharmacologic option.

Study Design

This retrospective case-control study was conducted using electronic health record data to compare ALS progression in patients on IV edaravone therapy with standard of care. The Indiana University/Purdue University, Indianapolis Institutional Review Board and the VA Research and Development Committee approved the study. The control cohort received the standard of care. Patients in the case cohort received standard of care and edaravone 60 mg infusions daily for an initial cycle of 14 days on treatment, followed by 14 days off. All subsequent cycles were 10 of 14 days on treatment followed by 14 days off. The initial 2 doses were administered in the outpatient infusion clinic to monitor for a hypersensitivity reaction. Patients then had a peripherally inserted central catheter line placed and received doses on days 3 through 14 at home. A port was placed for subsequent cycles, which were also completed at home. Appropriateness of edaravone therapy was assessed by the neurologist at each follow-up appointment. Therapy was then discontinued if warranted based on disease progression or patient preference.

Study Population

Patients included were aged 18 to 75 years with diagnosed ALS. Patients with complications that might influence evaluation of medication efficacy (eg, Parkinson disease, schizophrenia, significant dementia, other major medical morbidity) were excluded. Patients were also excluded if they were on continuous bilevel positive airway pressure and/or had a total score of ≤ 3 points on ALSFRS-R items for dyspnea, orthopnea, or respiratory insufficiency. Due to our small sample size, patients were excluded if treatment was < 6 months, which is the gold standard of therapy duration established by clinical trials.^9,11,12

The standard-of-care cohort included patients enrolled in the multidisciplinary clinic September 1, 2014 to August 31, 2017. These patients were compared in a 2:1 ratio with patients who received IV edaravone. The edaravone cohort included patients who initiated treatment with IV edaravone between September 1, 2017, and August 31, 2020. This date range prior to the approval of edaravone was chosen to compare patients at similar stages of disease progression and to have the largest sample size possible.

Data Collection

Data were obtained for eligible patients using the VA Computerized Patient Record System. Demographic data gathered for each patient included age, sex, weight, height, body mass index (BMI), race, and riluzole use.

The primary endpoint was the change in ALSFRS-R score after 6 months of IV edaravone compared with standard-of-care ALS management. Secondary outcomes included change in ALSFRS-R scores 3, 12, 18, and 24 months after therapy initiation, change in %FVC and SIS 3, 6, 12, 18, and 24 months after therapy initiation, duration of edaravone completed (months), time to death (months), and adverse events.

Statistical Analysis

Comparisons between the edaravone and control groups for differences in patient characteristics were made using χ² and 2-sample t tests for categorical and continuous variables, respectively. Comparisons between the 2 groups for differences in study outcomes (ALSFRS-R scores, %FVC, SIS) at each time point were evaluated using 2-sample t tests. Adverse events and adverse drug reactions were compared between groups using χ² tests. Statistical significance was set at 0.05.

We estimated that a sample size of 21 subjects in the edaravone (case) group and 42 in the standard-of-care (control) group would be needed to achieve 80% power to detect a difference of 6.5 between the 2 groups for the change in ALSFRS-R scores. This 80% power was calculated based on a 2-sample t test, and assuming a 2-sided 5% significance level and a within-group SD of 8.5.⁹ Statistical analysis was conducted using Microsoft Excel.

Results

Of the 96 patients, 10 met exclusion criteria. From the remaining 86, 42 were randomly selected for the standard-of-care group. A total of 27 patients seen in multidisciplinary ALS clinic between September 1, 2017, and August 31, 2020, received at least 1 dose of IV edaravone. Of the 27 edaravone patients, 6 were excluded for not completing a total of 6 months of edaravone. Two of the 6 excluded developed a rash, which resolved within 1 week after discontinuing edaravone. The other 4 discontinued edaravone before 6 months because of disease progression.

Baseline Characteristics

Efficacy

Discussion

This 24-month, case-control retrospective study assessed efficacy and safety of IV edaravone for the management of ALS. Although the landmark edaravone study showed slowed progression of ALS at 6 and 12 months, the effectiveness of edaravone outside the clinical trial setting has been less compelling.^9-11,13 A later study showed no difference in change in ALSFRS-R score at 6 months compared with that of the placebo group.⁷ In our study, no statistically significant difference was found for change in ALSFRS-R scores at 6 months.

Our study was unique given we evaluated a veteran population. The link between the military and ALS is largely unknown, although studies have shown increased incidence of ALS in people with a military history compared with that of the general population.^16-18 Our study was also unique because it was single-centered in design and allowed for outcome assessments, including ALSFRS-R scores, SIS, and %FVC measurements, to all be conducted by the same practitioner to limit variability. Unfortunately, our sample size resulted in a cohort that was underpowered at 12, 18, and 24 months. In addition, there was a lack of data on chart review for SIS and %FVC measurements at 24 months. As ALS progresses toward end stage, SIS and %FVC measurements can become difficult and burdensome on the patient to obtain, and the ALS multidisciplinary team may decide not to gather these data points as ALS progresses. As a result, change in SIS and %FVC measurements were unable to be reported due to lack of gathering this information at the 24-month mark in the edaravone group. Due to the cost and administration burden associated with edaravone, it is important that assessment of disease progression is performed regularly to assess benefit and appropriateness of continued therapy. The oral formulation of edaravone was approved in 2022, shortly after the completion of data collection for this study.^19,20 Although our study did not analyze oral edaravone, the administration burden of treatment would be reduced with the oral formulation, and we hypothesize there will be increased patient interest in ALS management with oral vs IV edaravone. Evaluation of long-term treatment for efficacy and safety beyond 24 months has not been evaluated. Future studies should continue to evaluate edaravone use in a larger veteran population.

Limitations

One limitation for our study alluded to earlier in the discussion was sample size. Although this study met power at the 6-month mark, it was limited by the number of patients who received more than 6 months of edaravone (n = 21). As a result, statistical analyses between treatment groups were underpowered at 12, 18, and 24 months. Our study had 80% power to detect a difference of 6.5 between the groups for the change in ALSFRS-R scores. Previous studies detected a statistically significant difference in ALSFRS-R scores, with a difference in ALSFRS-R scores of 2.49 between groups.⁸ Future studies should evaluate a larger sample size of patients who are prescribed edaravone.

Another limitation was that the edaravone and standard-of-care group data were gathered from different time periods. Two different time frames were selected to increase sample size by gathering data over a longer period and to account for patients who may have qualified for IV edaravone but could not receive it as it was not yet available on the market. There were no known changes to the standard of care between the time periods that would affect results. As noted previously, the standard-of-care group had fewer patients taking riluzole compared with the edaravone group, which may have confounded our results. We concluded patients opting for edaravone were more likely to trial riluzole, taken by mouth twice daily, before starting edaravone, a once-daily IV infusion.

Conclusions

No difference in the rate of ALS progression was noted between patients who received IV edaravone vs standard of care at 6 months. In addition, no difference was noted in other objective measures of disease progression, including %FVC, SIS, and time to death. As a result, the decision to initiate and continue edaravone therapy should be made on an individualized basis according to a prescriber’s clinical judgment and a patient’s goals. Edaravone therapy should be discontinued when disease progression occurs or when medication administration becomes a burden.

Acknowledgments

This material is the result of work supported with resources and the use of facilities at Veteran Health Indiana.

Amyotrophic lateral sclerosis (ALS) is an incurable neurodegenerative disorder that results in progressive deterioration of motor neurons in the ventral horn of the spinal cord, which results in loss of voluntary muscle movements.¹ Eventually, typical daily tasks become difficult to perform, and as the disease progresses, the ability to eat and breathe is impaired.² Reports from 2015 show the annual incidence of ALS is 5 cases per 100,000 people, with the total number of cases reported at more than 16,000 in the United States.³ In clinical practice, disease progression is routinely assessed by the Revised Amyotrophic Lateral Sclerosis Functional Rating Scale (ALSFRS-R). Typical decline is 1 point per month.⁴

Unfortunately, at this time, ALS care focuses on symptom management, including prevention of weight loss; implementation of communication strategies; and management of pain, constipation, excess secretions, cramping, and breathing. Despite copious research into treatment options, few exist. Riluzole is an oral medication administered twice daily and has been on the market since 1995.^5-7 Efficacy was demonstrated in a study showing statistically significant survival at 12 months compared with controls (74% vs 58%, respectively; P = .014).⁶ Since its approval, riluzole has become part of standard-of-care ALS management.

In 2017, the US Food and Drug Administration (FDA) approved edaravone, an IV medication that was found to slow the progression of ALS in some patients.^8-12 Oxidative stress caused by free radicals is hypothesized to increase the progression of ALS by motor neuron degradation.¹³ Edaravone works as a free radical and peroxynitrite scavenger and has been shown to eliminate lipid peroxides and hydroxyl radicals known to damage endothelial and neuronal cells.¹²

Given the mechanism of action of edaravone, it seemed to be a promising option to slow the progression of ALS. A 2019 systematic review analyzed 3 randomized studies with 367 patients and found a statistically significant difference in change in ALSFRS-R scores between patients treated with edaravone for 24 weeks compared with patients treated with the placebo (mean difference, 1.63; 95% CI, 0.26-3.00; P = .02).¹² Secondary endpoints evaluated included percent forced vital capacity (%FVC), grip strength, and pinch strength: All showing no significant difference when comparing IV edaravone with placebo.

A 2022 postmarketing study of 324 patients with ALS evaluated the safety and efficacy of long-term edaravone treatment. IV edaravone therapy for > 24 weeks was well tolerated, although it was not associated with any disease-modifying benefit when comparing ALSFRS-R scores with patients not receiving edaravone over a median 13.9 months (ALSFRS-R points/month, -0.91 vs -0.85; P = .37).¹³ A third ALS treatment medication, sodium phenylbutyrate/taurursodiol was approved in 2022 but not available during our study period and not included here.^14,15

Studies have shown an increased incidence of ALS in the veteran population. Veterans serving in the Gulf War were nearly twice as likely to develop ALS as those not serving in the Gulf.¹⁶ However, existing literature regarding the effectiveness of edaravone does not specifically examine the effect on this unique population. The objective of this study was to assess the effect of IV edaravone on ALS progression in veterans compared with veterans who received standard of care.

Methods

This study was conducted at a large, academic US Department of Veterans Affairs (VA) medical center. Patients with ALS are followed by a multidisciplinary clinic composed of a neurologist, pulmonologist, clinical pharmacist, social worker, speech therapist, physical therapist, occupational therapist, dietician, clinical psychologist, wheelchair clinic representative, and benefits representative. Patients are typically seen for a half-day appointment about every 3 months. During these visits, a comprehensive review of disease progression is performed. This review entails completion of the ALSFRS-R, physical examination, and pulmonary function testing. Speech intelligibility stage (SIS) is assessed by a speech therapist as well. SIS is scored from 1 (no detectable speech disorder) to 5 (no functional speech). All patients followed in this multidisciplinary ALS clinic receive standard-of-care treatment. This includes the discussion of treatment options that if appropriate are provided to help manage a wide range of complications associated with this disease (eg, pain, cramping, constipation, excessive secretions, weight loss, dysphagia). As a part of these personal discussions, treatment with riluzole is also offered as a standard-of-care pharmacologic option.

Study Design

This retrospective case-control study was conducted using electronic health record data to compare ALS progression in patients on IV edaravone therapy with standard of care. The Indiana University/Purdue University, Indianapolis Institutional Review Board and the VA Research and Development Committee approved the study. The control cohort received the standard of care. Patients in the case cohort received standard of care and edaravone 60 mg infusions daily for an initial cycle of 14 days on treatment, followed by 14 days off. All subsequent cycles were 10 of 14 days on treatment followed by 14 days off. The initial 2 doses were administered in the outpatient infusion clinic to monitor for a hypersensitivity reaction. Patients then had a peripherally inserted central catheter line placed and received doses on days 3 through 14 at home. A port was placed for subsequent cycles, which were also completed at home. Appropriateness of edaravone therapy was assessed by the neurologist at each follow-up appointment. Therapy was then discontinued if warranted based on disease progression or patient preference.

Study Population

Patients included were aged 18 to 75 years with diagnosed ALS. Patients with complications that might influence evaluation of medication efficacy (eg, Parkinson disease, schizophrenia, significant dementia, other major medical morbidity) were excluded. Patients were also excluded if they were on continuous bilevel positive airway pressure and/or had a total score of ≤ 3 points on ALSFRS-R items for dyspnea, orthopnea, or respiratory insufficiency. Due to our small sample size, patients were excluded if treatment was < 6 months, which is the gold standard of therapy duration established by clinical trials.^9,11,12

The standard-of-care cohort included patients enrolled in the multidisciplinary clinic September 1, 2014 to August 31, 2017. These patients were compared in a 2:1 ratio with patients who received IV edaravone. The edaravone cohort included patients who initiated treatment with IV edaravone between September 1, 2017, and August 31, 2020. This date range prior to the approval of edaravone was chosen to compare patients at similar stages of disease progression and to have the largest sample size possible.

Data Collection

Data were obtained for eligible patients using the VA Computerized Patient Record System. Demographic data gathered for each patient included age, sex, weight, height, body mass index (BMI), race, and riluzole use.

The primary endpoint was the change in ALSFRS-R score after 6 months of IV edaravone compared with standard-of-care ALS management. Secondary outcomes included change in ALSFRS-R scores 3, 12, 18, and 24 months after therapy initiation, change in %FVC and SIS 3, 6, 12, 18, and 24 months after therapy initiation, duration of edaravone completed (months), time to death (months), and adverse events.

Statistical Analysis

Comparisons between the edaravone and control groups for differences in patient characteristics were made using χ² and 2-sample t tests for categorical and continuous variables, respectively. Comparisons between the 2 groups for differences in study outcomes (ALSFRS-R scores, %FVC, SIS) at each time point were evaluated using 2-sample t tests. Adverse events and adverse drug reactions were compared between groups using χ² tests. Statistical significance was set at 0.05.

We estimated that a sample size of 21 subjects in the edaravone (case) group and 42 in the standard-of-care (control) group would be needed to achieve 80% power to detect a difference of 6.5 between the 2 groups for the change in ALSFRS-R scores. This 80% power was calculated based on a 2-sample t test, and assuming a 2-sided 5% significance level and a within-group SD of 8.5.⁹ Statistical analysis was conducted using Microsoft Excel.

Results

Of the 96 patients, 10 met exclusion criteria. From the remaining 86, 42 were randomly selected for the standard-of-care group. A total of 27 patients seen in multidisciplinary ALS clinic between September 1, 2017, and August 31, 2020, received at least 1 dose of IV edaravone. Of the 27 edaravone patients, 6 were excluded for not completing a total of 6 months of edaravone. Two of the 6 excluded developed a rash, which resolved within 1 week after discontinuing edaravone. The other 4 discontinued edaravone before 6 months because of disease progression.

Baseline Characteristics

Efficacy

Discussion

This 24-month, case-control retrospective study assessed efficacy and safety of IV edaravone for the management of ALS. Although the landmark edaravone study showed slowed progression of ALS at 6 and 12 months, the effectiveness of edaravone outside the clinical trial setting has been less compelling.^9-11,13 A later study showed no difference in change in ALSFRS-R score at 6 months compared with that of the placebo group.⁷ In our study, no statistically significant difference was found for change in ALSFRS-R scores at 6 months.

Our study was unique given we evaluated a veteran population. The link between the military and ALS is largely unknown, although studies have shown increased incidence of ALS in people with a military history compared with that of the general population.^16-18 Our study was also unique because it was single-centered in design and allowed for outcome assessments, including ALSFRS-R scores, SIS, and %FVC measurements, to all be conducted by the same practitioner to limit variability. Unfortunately, our sample size resulted in a cohort that was underpowered at 12, 18, and 24 months. In addition, there was a lack of data on chart review for SIS and %FVC measurements at 24 months. As ALS progresses toward end stage, SIS and %FVC measurements can become difficult and burdensome on the patient to obtain, and the ALS multidisciplinary team may decide not to gather these data points as ALS progresses. As a result, change in SIS and %FVC measurements were unable to be reported due to lack of gathering this information at the 24-month mark in the edaravone group. Due to the cost and administration burden associated with edaravone, it is important that assessment of disease progression is performed regularly to assess benefit and appropriateness of continued therapy. The oral formulation of edaravone was approved in 2022, shortly after the completion of data collection for this study.^19,20 Although our study did not analyze oral edaravone, the administration burden of treatment would be reduced with the oral formulation, and we hypothesize there will be increased patient interest in ALS management with oral vs IV edaravone. Evaluation of long-term treatment for efficacy and safety beyond 24 months has not been evaluated. Future studies should continue to evaluate edaravone use in a larger veteran population.

Limitations

One limitation for our study alluded to earlier in the discussion was sample size. Although this study met power at the 6-month mark, it was limited by the number of patients who received more than 6 months of edaravone (n = 21). As a result, statistical analyses between treatment groups were underpowered at 12, 18, and 24 months. Our study had 80% power to detect a difference of 6.5 between the groups for the change in ALSFRS-R scores. Previous studies detected a statistically significant difference in ALSFRS-R scores, with a difference in ALSFRS-R scores of 2.49 between groups.⁸ Future studies should evaluate a larger sample size of patients who are prescribed edaravone.

Another limitation was that the edaravone and standard-of-care group data were gathered from different time periods. Two different time frames were selected to increase sample size by gathering data over a longer period and to account for patients who may have qualified for IV edaravone but could not receive it as it was not yet available on the market. There were no known changes to the standard of care between the time periods that would affect results. As noted previously, the standard-of-care group had fewer patients taking riluzole compared with the edaravone group, which may have confounded our results. We concluded patients opting for edaravone were more likely to trial riluzole, taken by mouth twice daily, before starting edaravone, a once-daily IV infusion.

Conclusions

No difference in the rate of ALS progression was noted between patients who received IV edaravone vs standard of care at 6 months. In addition, no difference was noted in other objective measures of disease progression, including %FVC, SIS, and time to death. As a result, the decision to initiate and continue edaravone therapy should be made on an individualized basis according to a prescriber’s clinical judgment and a patient’s goals. Edaravone therapy should be discontinued when disease progression occurs or when medication administration becomes a burden.

Acknowledgments

This material is the result of work supported with resources and the use of facilities at Veteran Health Indiana.

Alcohol use disorder (AUD) is a chronic disease characterized by an impaired ability to control alcohol use that negatively impacts the social, occupational, and health aspects of patients’ lives.¹ It is the third leading modifiable cause of death in the United States.² About 50% of patients with AUD experience alcohol withdrawal syndrome (AWS) following abrupt cessation of alcohol use. AWS often presents with mild symptoms, such as headaches, nausea, vomiting, and anxiety. However, as many as 20% of patients experience severe and potentially life-threatening symptoms, such as tremors, delirium, hallucinations, and seizures within 48 hours of AWS onset.³

Benzodiazepines, such as lorazepam or chlordiazepoxide, are considered the gold standard for AWS.⁴ Benzodiazepines act by potentiation of γ-aminobutyric acid (GABA) receptors that produce inhibitory responses in the central nervous system (CNS). This mechanism is similar to the activity of ethanol, which acts primarily at the GABA-A receptors, resulting in facilitation of GABAergic transmission. The Clinical Institute Withdrawal Assessment (CIWA) of Alcohol scale is a commonly used tool to assess the severity of AWS and the appropriate dosing schedule of benzodiazepines.³ Multiple studies have demonstrated the superiority of using benzodiazepines, as they are beneficial for reducing withdrawal severity and incidence of delirium and seizures.^5,6

Although benzodiazepines are effective, they are associated with serious adverse effects (AEs), such as respiratory depression, excessive sedation, and abuse potential.⁴ Older patients are at higher risk of these AEs, particularly oversedation. In addition, sudden discontinuation of a benzodiazepine treatment can result in anxiety, irritability, and insomnia, which might worsen AWS.

Given the safety concerns of benzodiazepines, alternative treatments for AWS management have been investigated, including gabapentin. Previous studies have demonstrated gabapentin might be effective for mild-to-moderate AWS management.^7-9 Gabapentin exhibits its action by binding to the α2δ subunit of voltage-activated calcium channels with high affinity. Although the exact mechanism of action of gabapentin in AWS is unknown, it has been proposed that gabapentin normalizes GABA activation in the amygdala, which is associated with alcohol dependence.¹⁰ A systemic review conducted by Leung and colleagues found that gabapentin might be an option for the management of mild AWS.¹¹ However, current evidence does not support the use of gabapentin monotherapy in patients with severe AWS, a history of seizures, or those at risk of delirium tremens (DTs) since there is a higher chance of complications.

Baclofen is another medication investigated by researchers for use in patients with AWS. Baclofen works by activating the GABA-B receptor, which results in the downregulation of GABA-A activity. This results in a negative feedback loop leading to a decrease in excitatory neurotransmitters that is similar to the effect produced by alcohol.¹² However, there is limited evidence that baclofen is effective as monotherapy for the treatment of AWS. A Cochrane review previously evaluated baclofen use in AWS but found insufficient evidence of its efficacy and safety for this indication.¹³

Methods

We performed a retrospective cohort chart review at CJALFHCC. Data were collected from the facility’s electronic health record Computerized Patient Record System (CPRS). This study was approved by the Edward Hines Jr. Veterans Affairs Hospital Institutional Review Board.

Patient records were screened and included if they met the following criteria: (1) Patients aged ≥ 18 years who were hospitalized from January 1, 2014, to July 31, 2021, for the primary indication of AWS; (2) Patients who received a g/b or benzodiazepine protocol during AWS hospitalization. If a patient was admitted multiple times for AWS management, only the first admission was included for primary outcome analysis. Exclusion criteria were patients who were active-duty service members, discharged within 24 hours; patients with a primary seizure disorder; patients with known gabapentin, baclofen, or benzodiazepine allergy or intolerance. Patients who used gabapentin, baclofen, or benzodiazepines in an outpatient setting prior to AWS admission; had concurrent intoxication or overdose involving substances other than alcohol; had a concurrent regimen of gabapentin, baclofen, or benzodiazepines; or had initiation on adjuvant medications for AWS management (eg, divalproex, haloperidol, carbamazepine, or clonidine) also were excluded. Patients were categorized as those who received g/b as the initial therapy after admission or patients who received benzodiazepine therapy.

CPRS was used to collect information including baseline demographics, blood alcohol content, CIWA scores throughout hospitalization, number of admissions for alcohol detoxification in the previous year, AWS readmission within 30 days after discharge, prior treatment with g/b, history of alcohol withdrawal seizures and DTs, hospital LOS, outpatient medications for AUD treatment, rates of conversions from g/b protocol to lorazepam, and rates of transition to a higher level of care.

Statistical Analysis

Study data were stored and analyzed using an Excel spreadsheet and IBM SPSS Statistics software. LOS was compared between the g/b and benzodiazepine groups using inferential statistics. An independent 2-sample t test was used to assess the primary outcome if data were normally distributed. If the collected data were not distributed normally, the Mann-Whitney U test was used. All other continuous variables were assessed by using independent t tests and categorical variables by using χ² tests. A P value < .05 was considered statistically significant. Effect size of d = 0.42 was calculated based on a previous study with a similar research design as our study.⁹ We determined that if using an independent 2-sample t test for the primary outcome analysis, an estimated sample size of 178 subjects would provide the study with an 80% power to detect a difference at a 2-sided significance level with α = 0.05. If using the Mann-Whitney U test, 186 subjects would be required to provide identical power.

Results

We reviewed 196 patient health records, and 39 were initially excluded. The most common reason was that AWS was not the primary diagnosis for hospitalization (n = 28).

This retrospective chart review study found that LOS was shorter in patients with AWS treated with g/b compared with those treated with benzodiazepines, with no significant difference in safety outcomes such as seizures, DTs, or intensive care unit transfers. Although there was a statistically significant difference in the primary outcome between the 2 groups, it appears that patients on benzodiazepine therapy originally had more severe AWS presentation as their admission and maximum CIWA scores were statistically significantly higher compared with the g/b group. Thus, patients who were initially started on g/b had less serious AWS presentations. Based on this information we can conclude that the g/b combination may be an effective option for mild AWS management.

To our knowledge, this is the first study that has investigated the combination of g/b compared with benzodiazepines for AWS management in hospitalized patients. The research design of this project was adapted from the Bates and colleagues study that examined gabapentin monotherapy use for the treatment of patients hospitalized with AWS.⁹ We specifically used the primary outcome that they defined in their study since their LOS definition aimed to reflect clinically active withdrawal rather than simply hours of hospitalization, which would decrease the risk of confounding the primary outcome. The results of our research were similar to Bates and colleagues as they found that the gabapentin protocol appeared to be an effective and safe option compared with benzodiazepines for patients hospitalized with AWS.⁹

Limitations

This study has multiple limitations. As it was a retrospective chart review study, the data collection accuracy depends on accurate recordkeeping. Additionally, certain information was missing, such as CIWA scores for some patients. This study has limited external validity as most of the patients were older, White, and male, and the data collection was limited only to a single center. Therefore, it is uncertain whether the results of this study can be generalized to other populations. Also, this study had a small sample size, and we were not able to obtain the intended number of patients to achieve a power of 80%. Lastly, some background characteristics, such as admission and maximum CIWA scores, were not distributed equally between groups. Therefore, future studies are needed with a larger sample size that examine the LOS in the g/b group compared with the benzodiazepine group and in which CIWA scores are matched to reduce the effect of extraneous variables.

Conclusions

Gabapentin and baclofen combination seems to be an effective and safe alternative to benzodiazepines and may be considered for managing mild AWS in hospitalized patients, but additional research is needed to examine this regimen.

Acknowledgments

Research committee: Hong-Yen Vi, PharmD, BCPS; Shaiza Khan, PharmD, BCPS; Yinka Alaka, PharmD; Jennifer Kwon, PharmD, BCOP. Co-investigators: Zachary Rosenfeldt, PharmD, BCPS; Kaylee Caniff, PharmD, BCIDP.

Alcohol use disorder (AUD) is a chronic disease characterized by an impaired ability to control alcohol use that negatively impacts the social, occupational, and health aspects of patients’ lives.¹ It is the third leading modifiable cause of death in the United States.² About 50% of patients with AUD experience alcohol withdrawal syndrome (AWS) following abrupt cessation of alcohol use. AWS often presents with mild symptoms, such as headaches, nausea, vomiting, and anxiety. However, as many as 20% of patients experience severe and potentially life-threatening symptoms, such as tremors, delirium, hallucinations, and seizures within 48 hours of AWS onset.³

Benzodiazepines, such as lorazepam or chlordiazepoxide, are considered the gold standard for AWS.⁴ Benzodiazepines act by potentiation of γ-aminobutyric acid (GABA) receptors that produce inhibitory responses in the central nervous system (CNS). This mechanism is similar to the activity of ethanol, which acts primarily at the GABA-A receptors, resulting in facilitation of GABAergic transmission. The Clinical Institute Withdrawal Assessment (CIWA) of Alcohol scale is a commonly used tool to assess the severity of AWS and the appropriate dosing schedule of benzodiazepines.³ Multiple studies have demonstrated the superiority of using benzodiazepines, as they are beneficial for reducing withdrawal severity and incidence of delirium and seizures.^5,6

Although benzodiazepines are effective, they are associated with serious adverse effects (AEs), such as respiratory depression, excessive sedation, and abuse potential.⁴ Older patients are at higher risk of these AEs, particularly oversedation. In addition, sudden discontinuation of a benzodiazepine treatment can result in anxiety, irritability, and insomnia, which might worsen AWS.

Given the safety concerns of benzodiazepines, alternative treatments for AWS management have been investigated, including gabapentin. Previous studies have demonstrated gabapentin might be effective for mild-to-moderate AWS management.^7-9 Gabapentin exhibits its action by binding to the α2δ subunit of voltage-activated calcium channels with high affinity. Although the exact mechanism of action of gabapentin in AWS is unknown, it has been proposed that gabapentin normalizes GABA activation in the amygdala, which is associated with alcohol dependence.¹⁰ A systemic review conducted by Leung and colleagues found that gabapentin might be an option for the management of mild AWS.¹¹ However, current evidence does not support the use of gabapentin monotherapy in patients with severe AWS, a history of seizures, or those at risk of delirium tremens (DTs) since there is a higher chance of complications.

Baclofen is another medication investigated by researchers for use in patients with AWS. Baclofen works by activating the GABA-B receptor, which results in the downregulation of GABA-A activity. This results in a negative feedback loop leading to a decrease in excitatory neurotransmitters that is similar to the effect produced by alcohol.¹² However, there is limited evidence that baclofen is effective as monotherapy for the treatment of AWS. A Cochrane review previously evaluated baclofen use in AWS but found insufficient evidence of its efficacy and safety for this indication.¹³

Methods

We performed a retrospective cohort chart review at CJALFHCC. Data were collected from the facility’s electronic health record Computerized Patient Record System (CPRS). This study was approved by the Edward Hines Jr. Veterans Affairs Hospital Institutional Review Board.

Patient records were screened and included if they met the following criteria: (1) Patients aged ≥ 18 years who were hospitalized from January 1, 2014, to July 31, 2021, for the primary indication of AWS; (2) Patients who received a g/b or benzodiazepine protocol during AWS hospitalization. If a patient was admitted multiple times for AWS management, only the first admission was included for primary outcome analysis. Exclusion criteria were patients who were active-duty service members, discharged within 24 hours; patients with a primary seizure disorder; patients with known gabapentin, baclofen, or benzodiazepine allergy or intolerance. Patients who used gabapentin, baclofen, or benzodiazepines in an outpatient setting prior to AWS admission; had concurrent intoxication or overdose involving substances other than alcohol; had a concurrent regimen of gabapentin, baclofen, or benzodiazepines; or had initiation on adjuvant medications for AWS management (eg, divalproex, haloperidol, carbamazepine, or clonidine) also were excluded. Patients were categorized as those who received g/b as the initial therapy after admission or patients who received benzodiazepine therapy.

CPRS was used to collect information including baseline demographics, blood alcohol content, CIWA scores throughout hospitalization, number of admissions for alcohol detoxification in the previous year, AWS readmission within 30 days after discharge, prior treatment with g/b, history of alcohol withdrawal seizures and DTs, hospital LOS, outpatient medications for AUD treatment, rates of conversions from g/b protocol to lorazepam, and rates of transition to a higher level of care.

Statistical Analysis

Study data were stored and analyzed using an Excel spreadsheet and IBM SPSS Statistics software. LOS was compared between the g/b and benzodiazepine groups using inferential statistics. An independent 2-sample t test was used to assess the primary outcome if data were normally distributed. If the collected data were not distributed normally, the Mann-Whitney U test was used. All other continuous variables were assessed by using independent t tests and categorical variables by using χ² tests. A P value < .05 was considered statistically significant. Effect size of d = 0.42 was calculated based on a previous study with a similar research design as our study.⁹ We determined that if using an independent 2-sample t test for the primary outcome analysis, an estimated sample size of 178 subjects would provide the study with an 80% power to detect a difference at a 2-sided significance level with α = 0.05. If using the Mann-Whitney U test, 186 subjects would be required to provide identical power.

Results

We reviewed 196 patient health records, and 39 were initially excluded. The most common reason was that AWS was not the primary diagnosis for hospitalization (n = 28).

This retrospective chart review study found that LOS was shorter in patients with AWS treated with g/b compared with those treated with benzodiazepines, with no significant difference in safety outcomes such as seizures, DTs, or intensive care unit transfers. Although there was a statistically significant difference in the primary outcome between the 2 groups, it appears that patients on benzodiazepine therapy originally had more severe AWS presentation as their admission and maximum CIWA scores were statistically significantly higher compared with the g/b group. Thus, patients who were initially started on g/b had less serious AWS presentations. Based on this information we can conclude that the g/b combination may be an effective option for mild AWS management.

To our knowledge, this is the first study that has investigated the combination of g/b compared with benzodiazepines for AWS management in hospitalized patients. The research design of this project was adapted from the Bates and colleagues study that examined gabapentin monotherapy use for the treatment of patients hospitalized with AWS.⁹ We specifically used the primary outcome that they defined in their study since their LOS definition aimed to reflect clinically active withdrawal rather than simply hours of hospitalization, which would decrease the risk of confounding the primary outcome. The results of our research were similar to Bates and colleagues as they found that the gabapentin protocol appeared to be an effective and safe option compared with benzodiazepines for patients hospitalized with AWS.⁹

Limitations

This study has multiple limitations. As it was a retrospective chart review study, the data collection accuracy depends on accurate recordkeeping. Additionally, certain information was missing, such as CIWA scores for some patients. This study has limited external validity as most of the patients were older, White, and male, and the data collection was limited only to a single center. Therefore, it is uncertain whether the results of this study can be generalized to other populations. Also, this study had a small sample size, and we were not able to obtain the intended number of patients to achieve a power of 80%. Lastly, some background characteristics, such as admission and maximum CIWA scores, were not distributed equally between groups. Therefore, future studies are needed with a larger sample size that examine the LOS in the g/b group compared with the benzodiazepine group and in which CIWA scores are matched to reduce the effect of extraneous variables.

Conclusions

Gabapentin and baclofen combination seems to be an effective and safe alternative to benzodiazepines and may be considered for managing mild AWS in hospitalized patients, but additional research is needed to examine this regimen.

Acknowledgments

Research committee: Hong-Yen Vi, PharmD, BCPS; Shaiza Khan, PharmD, BCPS; Yinka Alaka, PharmD; Jennifer Kwon, PharmD, BCOP. Co-investigators: Zachary Rosenfeldt, PharmD, BCPS; Kaylee Caniff, PharmD, BCIDP.

Alcohol use disorder (AUD) is a chronic disease characterized by an impaired ability to control alcohol use that negatively impacts the social, occupational, and health aspects of patients’ lives.¹ It is the third leading modifiable cause of death in the United States.² About 50% of patients with AUD experience alcohol withdrawal syndrome (AWS) following abrupt cessation of alcohol use. AWS often presents with mild symptoms, such as headaches, nausea, vomiting, and anxiety. However, as many as 20% of patients experience severe and potentially life-threatening symptoms, such as tremors, delirium, hallucinations, and seizures within 48 hours of AWS onset.³

Benzodiazepines, such as lorazepam or chlordiazepoxide, are considered the gold standard for AWS.⁴ Benzodiazepines act by potentiation of γ-aminobutyric acid (GABA) receptors that produce inhibitory responses in the central nervous system (CNS). This mechanism is similar to the activity of ethanol, which acts primarily at the GABA-A receptors, resulting in facilitation of GABAergic transmission. The Clinical Institute Withdrawal Assessment (CIWA) of Alcohol scale is a commonly used tool to assess the severity of AWS and the appropriate dosing schedule of benzodiazepines.³ Multiple studies have demonstrated the superiority of using benzodiazepines, as they are beneficial for reducing withdrawal severity and incidence of delirium and seizures.^5,6

Although benzodiazepines are effective, they are associated with serious adverse effects (AEs), such as respiratory depression, excessive sedation, and abuse potential.⁴ Older patients are at higher risk of these AEs, particularly oversedation. In addition, sudden discontinuation of a benzodiazepine treatment can result in anxiety, irritability, and insomnia, which might worsen AWS.

Given the safety concerns of benzodiazepines, alternative treatments for AWS management have been investigated, including gabapentin. Previous studies have demonstrated gabapentin might be effective for mild-to-moderate AWS management.^7-9 Gabapentin exhibits its action by binding to the α2δ subunit of voltage-activated calcium channels with high affinity. Although the exact mechanism of action of gabapentin in AWS is unknown, it has been proposed that gabapentin normalizes GABA activation in the amygdala, which is associated with alcohol dependence.¹⁰ A systemic review conducted by Leung and colleagues found that gabapentin might be an option for the management of mild AWS.¹¹ However, current evidence does not support the use of gabapentin monotherapy in patients with severe AWS, a history of seizures, or those at risk of delirium tremens (DTs) since there is a higher chance of complications.

Baclofen is another medication investigated by researchers for use in patients with AWS. Baclofen works by activating the GABA-B receptor, which results in the downregulation of GABA-A activity. This results in a negative feedback loop leading to a decrease in excitatory neurotransmitters that is similar to the effect produced by alcohol.¹² However, there is limited evidence that baclofen is effective as monotherapy for the treatment of AWS. A Cochrane review previously evaluated baclofen use in AWS but found insufficient evidence of its efficacy and safety for this indication.¹³

Methods

We performed a retrospective cohort chart review at CJALFHCC. Data were collected from the facility’s electronic health record Computerized Patient Record System (CPRS). This study was approved by the Edward Hines Jr. Veterans Affairs Hospital Institutional Review Board.

Patient records were screened and included if they met the following criteria: (1) Patients aged ≥ 18 years who were hospitalized from January 1, 2014, to July 31, 2021, for the primary indication of AWS; (2) Patients who received a g/b or benzodiazepine protocol during AWS hospitalization. If a patient was admitted multiple times for AWS management, only the first admission was included for primary outcome analysis. Exclusion criteria were patients who were active-duty service members, discharged within 24 hours; patients with a primary seizure disorder; patients with known gabapentin, baclofen, or benzodiazepine allergy or intolerance. Patients who used gabapentin, baclofen, or benzodiazepines in an outpatient setting prior to AWS admission; had concurrent intoxication or overdose involving substances other than alcohol; had a concurrent regimen of gabapentin, baclofen, or benzodiazepines; or had initiation on adjuvant medications for AWS management (eg, divalproex, haloperidol, carbamazepine, or clonidine) also were excluded. Patients were categorized as those who received g/b as the initial therapy after admission or patients who received benzodiazepine therapy.

CPRS was used to collect information including baseline demographics, blood alcohol content, CIWA scores throughout hospitalization, number of admissions for alcohol detoxification in the previous year, AWS readmission within 30 days after discharge, prior treatment with g/b, history of alcohol withdrawal seizures and DTs, hospital LOS, outpatient medications for AUD treatment, rates of conversions from g/b protocol to lorazepam, and rates of transition to a higher level of care.

Statistical Analysis

Study data were stored and analyzed using an Excel spreadsheet and IBM SPSS Statistics software. LOS was compared between the g/b and benzodiazepine groups using inferential statistics. An independent 2-sample t test was used to assess the primary outcome if data were normally distributed. If the collected data were not distributed normally, the Mann-Whitney U test was used. All other continuous variables were assessed by using independent t tests and categorical variables by using χ² tests. A P value < .05 was considered statistically significant. Effect size of d = 0.42 was calculated based on a previous study with a similar research design as our study.⁹ We determined that if using an independent 2-sample t test for the primary outcome analysis, an estimated sample size of 178 subjects would provide the study with an 80% power to detect a difference at a 2-sided significance level with α = 0.05. If using the Mann-Whitney U test, 186 subjects would be required to provide identical power.

Results

We reviewed 196 patient health records, and 39 were initially excluded. The most common reason was that AWS was not the primary diagnosis for hospitalization (n = 28).

This retrospective chart review study found that LOS was shorter in patients with AWS treated with g/b compared with those treated with benzodiazepines, with no significant difference in safety outcomes such as seizures, DTs, or intensive care unit transfers. Although there was a statistically significant difference in the primary outcome between the 2 groups, it appears that patients on benzodiazepine therapy originally had more severe AWS presentation as their admission and maximum CIWA scores were statistically significantly higher compared with the g/b group. Thus, patients who were initially started on g/b had less serious AWS presentations. Based on this information we can conclude that the g/b combination may be an effective option for mild AWS management.

To our knowledge, this is the first study that has investigated the combination of g/b compared with benzodiazepines for AWS management in hospitalized patients. The research design of this project was adapted from the Bates and colleagues study that examined gabapentin monotherapy use for the treatment of patients hospitalized with AWS.⁹ We specifically used the primary outcome that they defined in their study since their LOS definition aimed to reflect clinically active withdrawal rather than simply hours of hospitalization, which would decrease the risk of confounding the primary outcome. The results of our research were similar to Bates and colleagues as they found that the gabapentin protocol appeared to be an effective and safe option compared with benzodiazepines for patients hospitalized with AWS.⁹

Limitations

This study has multiple limitations. As it was a retrospective chart review study, the data collection accuracy depends on accurate recordkeeping. Additionally, certain information was missing, such as CIWA scores for some patients. This study has limited external validity as most of the patients were older, White, and male, and the data collection was limited only to a single center. Therefore, it is uncertain whether the results of this study can be generalized to other populations. Also, this study had a small sample size, and we were not able to obtain the intended number of patients to achieve a power of 80%. Lastly, some background characteristics, such as admission and maximum CIWA scores, were not distributed equally between groups. Therefore, future studies are needed with a larger sample size that examine the LOS in the g/b group compared with the benzodiazepine group and in which CIWA scores are matched to reduce the effect of extraneous variables.

Conclusions

Gabapentin and baclofen combination seems to be an effective and safe alternative to benzodiazepines and may be considered for managing mild AWS in hospitalized patients, but additional research is needed to examine this regimen.

Acknowledgments

Research committee: Hong-Yen Vi, PharmD, BCPS; Shaiza Khan, PharmD, BCPS; Yinka Alaka, PharmD; Jennifer Kwon, PharmD, BCOP. Co-investigators: Zachary Rosenfeldt, PharmD, BCPS; Kaylee Caniff, PharmD, BCIDP.

Acute pain is a primary symptom for many patients who present to the emergency department (ED). The ED team is challenged with relieving pain while limiting harm from medications.¹ A 2017 National Health Interview Survey showed that compared with nonveterans, more veterans reported pain in the previous 3 months, and the rate of severe pain was 40% higher in the veteran group especially among those who served during the era of wars in Afghanistan and Iraq.²

The American College of Emergency Physicians guidelines pain management guidelines recommend patient-centered shared decision making that includes patient education about treatment goals and expectations, and short- and long-term risks, as well as a preference toward pharmacologic treatment with nonopioid analgesics except for patients with severe pain or pain refractory to other drug and treatment modalities.³ There is a lack of evidence regarding superior efficacy of either opioid or nonopioid analgesics; therefore, the use of nonopioid analgesics, such as oral or topical nonsteroidal anti-inflammatory drugs (NSAIDs) or central analgesics, such as acetaminophen, is preferred for treating acute pain to mitigate adverse effects (AEs) and risks associated with opioid use.^1,3,4 The US Department of Veterans Affairs (VA) and Department of Defense (DoD) guideline on managing opioid therapy for chronic pain, updated in 2017 and 2022, similarly recommends alternatives to opioids for mild-to-moderate acute pain and encourages multimodal pain care.⁵ However, use of other pharmacologic treatments, such as NSAIDs, is limited by AE profiles, patient contraindications, and severity of acute pain etiologies. There is a need for the expanded use of nonpharmacologic treatments for addressing pain in the veteran population.

The American College of Emergency Physicians guidelines recommend nonpharmacologic modalities, such as applying heat or cold, physical therapy, cognitive behavioral therapy, and acupuncture.³ A 2014 study reported that 37% to 46% of active duty and reserve military personnel use complementary and alternative medicine (CAM) for a variety of ailments, and there is increasing interest in the use of CAM as adjuncts to traditional therapies.⁶ According to one study, some CAM therapies are used significantly more by military personnel than used by civilians.⁷ However, the percentage of the veteran population using acupuncture in this study was small, and more information is needed to assess its use.

Auricular acupuncture originated in traditional Chinese medicine.⁸ Contemporary auricular acupuncture experts view this modality as a self-contained microsystem mapping portions of the ear to specific parts of the body and internal organs. The analgesic effects may be mediated through the central nervous system by local release of endorphins through nerve fiber activation and neurotransmitters—including serotonin, dopamine, and norepinephrine—leading to pre- and postsynaptic suppression of pain transmission.

Battlefield acupuncture (BFA) uses 5 set points anatomically located on each ear.⁹ Practitioners use small semipermanent, dartlike acupuncture needles. Patients could experience pain relief in a few minutes, which can last minutes, hours, days, weeks, or months depending on the pathology of the pain. This procedure developed in 2001 has been studied for different pain types and has shown benefit when used for postsurgical pain, chronic spinal cord injury−related neuropathic pain, and general chronic pain, as well as for other indications, such as insomnia, depression, and weight loss.^8,10-13 In 2018, a randomized controlled trial compared postintervention numeric rating scale (NRS) pain scores in patients presenting to the ED with acute or acute-on-chronic lower back pain who received BFA as an adjunct to standard care vs standard care alone.¹⁴ Patients receiving BFA as an adjunct to standard care were found to have mean postintervention pain scores 1.7 points lower than those receiving standard care alone. This study demonstrated that BFA was feasible and well tolerated for lower back pain in the ED as an adjunct to standard care. The study was limited by the adjunct use of BFA rather than as monotherapy and by the practitioners’ discretion regarding standard care, which was not defined by the study’s authors.

The Jesse Brown Veterans Affairs Medical Center (JBVAMC) in Chicago, Illinois, offers several CAM modalities, such as exercise/movement therapy, chiropractic, art/music therapy, and relaxation workshops, which are widely used by veterans. Recent evidence suggests BFA could reduce pain scores as an adjunct or an alternative to pharmacologic therapy. We are interested in how CAM therapies, such as BFA, can help avoid AEs associated with opioid or NSAID therapy.

At the JBVAMC ED, ketorolac 15 mg is the preferred first-line treatment of acute, noncancer pain, based on the results of previous studies. In 2018 BFA was offered first to veterans presenting with acute or acute-on-chronic pain to the ED; however, its effectiveness for pain reduction vs ketorolac has not been evaluated in this patient population. Limited literature is available on BFA and its use in the ED. To our knowledge, this was the first observational study assessing the difference between a single session of BFA vs a single dose of ketorolac in treating noncancer acute or acute-on-chronic pain in the ED.

Methods

This study was a retrospective chart review of patients who presented to the JBVAMC ED with acute pain or acute-on-chronic pain, who received ketorolac or BFA. The study population was generated from a list of all IV and intramuscular (IM) ketorolac unit dose orders verified from June 1, 2018, through August 30, 2019, and a list of all BFA procedure notes signed from June 1, 2018, through August 30, 2019. Patients were included in the study if they had documented administration of IV or IM ketorolac or BFA between June 1, 2018, and August 30, 2019. Patients who received ketorolac doses other than 15 mg, the intervention was administered outside of the ED, received adjunct treatment in addition to the treatment intervention in the ED, had no baseline NRS pain score documented before the intervention, had an NRS pain score of < 4, had no postintervention NRS pain score documented within 6 hours, had a treatment indication other than pain, or had active cancer were excluded. As in previous JBVAMC studies, we used NRS pain score cutoffs (mild, moderate, severe, and very severe) based on Woo and colleagues’ meta-analysis and excluded scores < 4.¹⁵

Endpoints

The primary endpoint was the mean difference in NRS pain score before and after the intervention, determined by comparing the NRS pain score documented at triage to the ED with the first documented NRS pain score at least 30 minutes to 6 hours after treatment administration. The secondary endpoints included the number of patients prescribed pain medication at discharge, the number of patients who were discharged with no medications, and the number of patients admitted to the hospital. The safety endpoint included any AEs of the intervention. Subgroup analyses were performed comparing the mean difference in NRS pain score among subgroups classified by severity of baseline NRS pain score and pain location.

Statistical Analysis

Baseline characteristics and endpoints were analyzed using descriptive statistics. Categorical data were analyzed using Fisher exact test and z test for proportions, and continuous data were compared using t test and paired t test. An 80% power calculation determined that 84 patients per group were needed to detect a statistically significant difference in pain score reduction of 1.3 at a type-1 error rate of 0.05. The sample size was based on a calculation performed in a previously published study that compared IV ketorolac at 3 single-dose regimens for treating acute pain in the ED.¹⁶ The 1.3 pain score reduction is considered the minimum clinically significant difference in pain that could be detected with the NRS.¹⁷

Results

Sixty-one patients received BFA during the study period: 31 were excluded (26 received adjunct treatment in the ED, 2 had active cancer documented, 2 had an indication other than pain, and 1 received BFA outside of the ED), leaving 30 patients in the BFA cohort. During the study period, 1299 patients received ketorolac. These patients were selected using a random number generator and then screened to determine inclusion or exclusion in the study. We continued to randomly select patients for the ketorolac group until we had a similar number in each treatment group. Of these 148 patients who were randomly selected to be reviewed, 116 were excluded: 48 received adjunct treatment in the ED, 24 had no postintervention NRS pain score documented within 6 hours, 18 received ketorolac doses other than 15 mg, 12 received ketorolac outside the ED, 9 had no baseline NRS pain score documented, 3 presented with a NRS pain score of ≤ 3, and 2 had active cancer documented. The ketorolac cohort comprised 31 patients.

Baseline characteristics were similar between the 2 groups except for the average baseline NRS pain score, which was statistically significantly higher in the BFA vs ketorolac group (8.7 vs 7.7, respectively; P = .02). The mean age was 51 years in the BFA group and 48 years in the ketorolac group. Most patients in each cohort were male: 80% in the BFA group and 71% in the ketorolac group. The most common types of pain documented as the chief ED presentation included back, lower extremity, and head.

Endpoints

The mean difference in NRS pain score was 3.9 for the BFA group and 5.1 for the ketorolac group. Both were clinically and statistically significant reductions (P = .03 and P < .01), but the difference between the intervention groups in NRS score reduction was not statistically significant (P = .07).

For the secondary endpoint of outpatient prescriptions written at discharge, there was no significant difference between the groups except for oral NSAIDs, which were more likely to be prescribed to patients who received ketorolac (P = .01).

Subgroup Analysis

An analysis was performed for subgroups classified by baseline NRS pain score (mild: 4; moderate, 5 - 6; severe, 7 - 9; and very severe, 10). Data for mild pain was limited because a small number of patients received interventions. For moderate pain, the mean difference in NRS pain score for BFA and ketorolac was 3.5 and 3.8, respectively; for severe pain, 3.4 and 5.3; and for very severe pain, 4.6 and 6.4. There was a larger difference in the preintervention and postintervention NRS pain scores within severe pain and very severe pain groups.

Discussion

Both interventions resulted in a significant reduction in the mean NRS pain score of about 4 to 5 points within their group, and BFA resulted in a similar NRS pain score reduction compared with ketorolac 15 mg. Because the baseline NRS pain scores were significantly different between the BFA and ketorolac groups, a subgroup analysis revealed that BFA reduced mean NRS pain score in patients with severe and very severe pain but appears to be less beneficial for moderate pain, unlike the ketorolac results that showed a large reduction in all pain groups except for the small sample of patients with mild pain.

In this study, more patients in the BFA group presented to the ED with lower extremity pain, such as gout or neuropathy, compared with the ketorolac group; however, BFA did not result in a significantly different pain score reduction in this subgroup compared with ketorolac. Patients receiving BFA were more likely to receive topical analgesics or muscle relaxants at discharge; whereas those receiving ketorolac were significantly more likely to receive oral NSAIDs. Patients in this study also were more likely to be admitted to the hospital if they received ketorolac; however, for these patients, pain was secondary to their chief presentation, and the admitting physician’s familiarity with ketorolac might have been the reason for choosing this intervention. Reasons for the admissions were surgical observation, psychiatric stabilization, kidney/gallstones, rule out of acute coronary syndrome, pneumonia, and proctitis in the ketorolac group, and suicidal ideations in the BFA group.

Limitations

As a limited number of patients received BFA at JBVAMC, the study was not sufficiently powered to detect a difference in the primary outcome. Because BFA required a consultation to be entered in the electronic health record, in addition to time needed to perform the procedure, practitioners might have preferred IV/IM ketorolac during busy times in the ED, potentially leading to underrepresentation in the BFA group. Prescribing preferences might have differed among the rotating physicians, timing of the documentation of the NRS pain score could have differed based on the treatment intervention, and the investigators were unable to control or accurately assess whether patients had taken an analgesic medication before presenting to the ED. Because pain and the treating physician are subjective, patients who reported a higher baseline pain severity might have been more likely to be discharged with topical analgesics or muscle relaxants. One way to correct for this subjectivity would be to conduct a larger prospective trial with a single treating physician. Finally, ED encounters in this study were short, and there was no follow-up permitting identification of AEs.

Conclusions

NRS pain score reduction with BFA did not differ compared with ketorolac 15 mg for treating acute and acute-on-chronic pain in the ED. Although this study was underpowered, these results add to the limited existing literature, suggesting that both interventions could result in clinically significant pain score reductions for patients presenting to the ED with severe and very severe pain, making BFA a viable nonpharmacologic option. Future studies could include investigating the benefit of BFA in the veteran population by studying larger samples in the ED, surveying patients after their interventions to identify rates AEs, and exploring the use of BFA for chronic pain in the outpatient setting.

Acute pain is a primary symptom for many patients who present to the emergency department (ED). The ED team is challenged with relieving pain while limiting harm from medications.¹ A 2017 National Health Interview Survey showed that compared with nonveterans, more veterans reported pain in the previous 3 months, and the rate of severe pain was 40% higher in the veteran group especially among those who served during the era of wars in Afghanistan and Iraq.²

The American College of Emergency Physicians guidelines pain management guidelines recommend patient-centered shared decision making that includes patient education about treatment goals and expectations, and short- and long-term risks, as well as a preference toward pharmacologic treatment with nonopioid analgesics except for patients with severe pain or pain refractory to other drug and treatment modalities.³ There is a lack of evidence regarding superior efficacy of either opioid or nonopioid analgesics; therefore, the use of nonopioid analgesics, such as oral or topical nonsteroidal anti-inflammatory drugs (NSAIDs) or central analgesics, such as acetaminophen, is preferred for treating acute pain to mitigate adverse effects (AEs) and risks associated with opioid use.^1,3,4 The US Department of Veterans Affairs (VA) and Department of Defense (DoD) guideline on managing opioid therapy for chronic pain, updated in 2017 and 2022, similarly recommends alternatives to opioids for mild-to-moderate acute pain and encourages multimodal pain care.⁵ However, use of other pharmacologic treatments, such as NSAIDs, is limited by AE profiles, patient contraindications, and severity of acute pain etiologies. There is a need for the expanded use of nonpharmacologic treatments for addressing pain in the veteran population.

The American College of Emergency Physicians guidelines recommend nonpharmacologic modalities, such as applying heat or cold, physical therapy, cognitive behavioral therapy, and acupuncture.³ A 2014 study reported that 37% to 46% of active duty and reserve military personnel use complementary and alternative medicine (CAM) for a variety of ailments, and there is increasing interest in the use of CAM as adjuncts to traditional therapies.⁶ According to one study, some CAM therapies are used significantly more by military personnel than used by civilians.⁷ However, the percentage of the veteran population using acupuncture in this study was small, and more information is needed to assess its use.

Auricular acupuncture originated in traditional Chinese medicine.⁸ Contemporary auricular acupuncture experts view this modality as a self-contained microsystem mapping portions of the ear to specific parts of the body and internal organs. The analgesic effects may be mediated through the central nervous system by local release of endorphins through nerve fiber activation and neurotransmitters—including serotonin, dopamine, and norepinephrine—leading to pre- and postsynaptic suppression of pain transmission.

Battlefield acupuncture (BFA) uses 5 set points anatomically located on each ear.⁹ Practitioners use small semipermanent, dartlike acupuncture needles. Patients could experience pain relief in a few minutes, which can last minutes, hours, days, weeks, or months depending on the pathology of the pain. This procedure developed in 2001 has been studied for different pain types and has shown benefit when used for postsurgical pain, chronic spinal cord injury−related neuropathic pain, and general chronic pain, as well as for other indications, such as insomnia, depression, and weight loss.^8,10-13 In 2018, a randomized controlled trial compared postintervention numeric rating scale (NRS) pain scores in patients presenting to the ED with acute or acute-on-chronic lower back pain who received BFA as an adjunct to standard care vs standard care alone.¹⁴ Patients receiving BFA as an adjunct to standard care were found to have mean postintervention pain scores 1.7 points lower than those receiving standard care alone. This study demonstrated that BFA was feasible and well tolerated for lower back pain in the ED as an adjunct to standard care. The study was limited by the adjunct use of BFA rather than as monotherapy and by the practitioners’ discretion regarding standard care, which was not defined by the study’s authors.

The Jesse Brown Veterans Affairs Medical Center (JBVAMC) in Chicago, Illinois, offers several CAM modalities, such as exercise/movement therapy, chiropractic, art/music therapy, and relaxation workshops, which are widely used by veterans. Recent evidence suggests BFA could reduce pain scores as an adjunct or an alternative to pharmacologic therapy. We are interested in how CAM therapies, such as BFA, can help avoid AEs associated with opioid or NSAID therapy.

At the JBVAMC ED, ketorolac 15 mg is the preferred first-line treatment of acute, noncancer pain, based on the results of previous studies. In 2018 BFA was offered first to veterans presenting with acute or acute-on-chronic pain to the ED; however, its effectiveness for pain reduction vs ketorolac has not been evaluated in this patient population. Limited literature is available on BFA and its use in the ED. To our knowledge, this was the first observational study assessing the difference between a single session of BFA vs a single dose of ketorolac in treating noncancer acute or acute-on-chronic pain in the ED.

Methods

This study was a retrospective chart review of patients who presented to the JBVAMC ED with acute pain or acute-on-chronic pain, who received ketorolac or BFA. The study population was generated from a list of all IV and intramuscular (IM) ketorolac unit dose orders verified from June 1, 2018, through August 30, 2019, and a list of all BFA procedure notes signed from June 1, 2018, through August 30, 2019. Patients were included in the study if they had documented administration of IV or IM ketorolac or BFA between June 1, 2018, and August 30, 2019. Patients who received ketorolac doses other than 15 mg, the intervention was administered outside of the ED, received adjunct treatment in addition to the treatment intervention in the ED, had no baseline NRS pain score documented before the intervention, had an NRS pain score of < 4, had no postintervention NRS pain score documented within 6 hours, had a treatment indication other than pain, or had active cancer were excluded. As in previous JBVAMC studies, we used NRS pain score cutoffs (mild, moderate, severe, and very severe) based on Woo and colleagues’ meta-analysis and excluded scores < 4.¹⁵

Endpoints

The primary endpoint was the mean difference in NRS pain score before and after the intervention, determined by comparing the NRS pain score documented at triage to the ED with the first documented NRS pain score at least 30 minutes to 6 hours after treatment administration. The secondary endpoints included the number of patients prescribed pain medication at discharge, the number of patients who were discharged with no medications, and the number of patients admitted to the hospital. The safety endpoint included any AEs of the intervention. Subgroup analyses were performed comparing the mean difference in NRS pain score among subgroups classified by severity of baseline NRS pain score and pain location.

Statistical Analysis

Baseline characteristics and endpoints were analyzed using descriptive statistics. Categorical data were analyzed using Fisher exact test and z test for proportions, and continuous data were compared using t test and paired t test. An 80% power calculation determined that 84 patients per group were needed to detect a statistically significant difference in pain score reduction of 1.3 at a type-1 error rate of 0.05. The sample size was based on a calculation performed in a previously published study that compared IV ketorolac at 3 single-dose regimens for treating acute pain in the ED.¹⁶ The 1.3 pain score reduction is considered the minimum clinically significant difference in pain that could be detected with the NRS.¹⁷

Results

Sixty-one patients received BFA during the study period: 31 were excluded (26 received adjunct treatment in the ED, 2 had active cancer documented, 2 had an indication other than pain, and 1 received BFA outside of the ED), leaving 30 patients in the BFA cohort. During the study period, 1299 patients received ketorolac. These patients were selected using a random number generator and then screened to determine inclusion or exclusion in the study. We continued to randomly select patients for the ketorolac group until we had a similar number in each treatment group. Of these 148 patients who were randomly selected to be reviewed, 116 were excluded: 48 received adjunct treatment in the ED, 24 had no postintervention NRS pain score documented within 6 hours, 18 received ketorolac doses other than 15 mg, 12 received ketorolac outside the ED, 9 had no baseline NRS pain score documented, 3 presented with a NRS pain score of ≤ 3, and 2 had active cancer documented. The ketorolac cohort comprised 31 patients.

Baseline characteristics were similar between the 2 groups except for the average baseline NRS pain score, which was statistically significantly higher in the BFA vs ketorolac group (8.7 vs 7.7, respectively; P = .02). The mean age was 51 years in the BFA group and 48 years in the ketorolac group. Most patients in each cohort were male: 80% in the BFA group and 71% in the ketorolac group. The most common types of pain documented as the chief ED presentation included back, lower extremity, and head.

Endpoints

The mean difference in NRS pain score was 3.9 for the BFA group and 5.1 for the ketorolac group. Both were clinically and statistically significant reductions (P = .03 and P < .01), but the difference between the intervention groups in NRS score reduction was not statistically significant (P = .07).

For the secondary endpoint of outpatient prescriptions written at discharge, there was no significant difference between the groups except for oral NSAIDs, which were more likely to be prescribed to patients who received ketorolac (P = .01).

Subgroup Analysis

An analysis was performed for subgroups classified by baseline NRS pain score (mild: 4; moderate, 5 - 6; severe, 7 - 9; and very severe, 10). Data for mild pain was limited because a small number of patients received interventions. For moderate pain, the mean difference in NRS pain score for BFA and ketorolac was 3.5 and 3.8, respectively; for severe pain, 3.4 and 5.3; and for very severe pain, 4.6 and 6.4. There was a larger difference in the preintervention and postintervention NRS pain scores within severe pain and very severe pain groups.

Discussion

Both interventions resulted in a significant reduction in the mean NRS pain score of about 4 to 5 points within their group, and BFA resulted in a similar NRS pain score reduction compared with ketorolac 15 mg. Because the baseline NRS pain scores were significantly different between the BFA and ketorolac groups, a subgroup analysis revealed that BFA reduced mean NRS pain score in patients with severe and very severe pain but appears to be less beneficial for moderate pain, unlike the ketorolac results that showed a large reduction in all pain groups except for the small sample of patients with mild pain.

In this study, more patients in the BFA group presented to the ED with lower extremity pain, such as gout or neuropathy, compared with the ketorolac group; however, BFA did not result in a significantly different pain score reduction in this subgroup compared with ketorolac. Patients receiving BFA were more likely to receive topical analgesics or muscle relaxants at discharge; whereas those receiving ketorolac were significantly more likely to receive oral NSAIDs. Patients in this study also were more likely to be admitted to the hospital if they received ketorolac; however, for these patients, pain was secondary to their chief presentation, and the admitting physician’s familiarity with ketorolac might have been the reason for choosing this intervention. Reasons for the admissions were surgical observation, psychiatric stabilization, kidney/gallstones, rule out of acute coronary syndrome, pneumonia, and proctitis in the ketorolac group, and suicidal ideations in the BFA group.

Limitations

As a limited number of patients received BFA at JBVAMC, the study was not sufficiently powered to detect a difference in the primary outcome. Because BFA required a consultation to be entered in the electronic health record, in addition to time needed to perform the procedure, practitioners might have preferred IV/IM ketorolac during busy times in the ED, potentially leading to underrepresentation in the BFA group. Prescribing preferences might have differed among the rotating physicians, timing of the documentation of the NRS pain score could have differed based on the treatment intervention, and the investigators were unable to control or accurately assess whether patients had taken an analgesic medication before presenting to the ED. Because pain and the treating physician are subjective, patients who reported a higher baseline pain severity might have been more likely to be discharged with topical analgesics or muscle relaxants. One way to correct for this subjectivity would be to conduct a larger prospective trial with a single treating physician. Finally, ED encounters in this study were short, and there was no follow-up permitting identification of AEs.

Conclusions

NRS pain score reduction with BFA did not differ compared with ketorolac 15 mg for treating acute and acute-on-chronic pain in the ED. Although this study was underpowered, these results add to the limited existing literature, suggesting that both interventions could result in clinically significant pain score reductions for patients presenting to the ED with severe and very severe pain, making BFA a viable nonpharmacologic option. Future studies could include investigating the benefit of BFA in the veteran population by studying larger samples in the ED, surveying patients after their interventions to identify rates AEs, and exploring the use of BFA for chronic pain in the outpatient setting.

Acute pain is a primary symptom for many patients who present to the emergency department (ED). The ED team is challenged with relieving pain while limiting harm from medications.¹ A 2017 National Health Interview Survey showed that compared with nonveterans, more veterans reported pain in the previous 3 months, and the rate of severe pain was 40% higher in the veteran group especially among those who served during the era of wars in Afghanistan and Iraq.²

The American College of Emergency Physicians guidelines pain management guidelines recommend patient-centered shared decision making that includes patient education about treatment goals and expectations, and short- and long-term risks, as well as a preference toward pharmacologic treatment with nonopioid analgesics except for patients with severe pain or pain refractory to other drug and treatment modalities.³ There is a lack of evidence regarding superior efficacy of either opioid or nonopioid analgesics; therefore, the use of nonopioid analgesics, such as oral or topical nonsteroidal anti-inflammatory drugs (NSAIDs) or central analgesics, such as acetaminophen, is preferred for treating acute pain to mitigate adverse effects (AEs) and risks associated with opioid use.^1,3,4 The US Department of Veterans Affairs (VA) and Department of Defense (DoD) guideline on managing opioid therapy for chronic pain, updated in 2017 and 2022, similarly recommends alternatives to opioids for mild-to-moderate acute pain and encourages multimodal pain care.⁵ However, use of other pharmacologic treatments, such as NSAIDs, is limited by AE profiles, patient contraindications, and severity of acute pain etiologies. There is a need for the expanded use of nonpharmacologic treatments for addressing pain in the veteran population.

The American College of Emergency Physicians guidelines recommend nonpharmacologic modalities, such as applying heat or cold, physical therapy, cognitive behavioral therapy, and acupuncture.³ A 2014 study reported that 37% to 46% of active duty and reserve military personnel use complementary and alternative medicine (CAM) for a variety of ailments, and there is increasing interest in the use of CAM as adjuncts to traditional therapies.⁶ According to one study, some CAM therapies are used significantly more by military personnel than used by civilians.⁷ However, the percentage of the veteran population using acupuncture in this study was small, and more information is needed to assess its use.

Auricular acupuncture originated in traditional Chinese medicine.⁸ Contemporary auricular acupuncture experts view this modality as a self-contained microsystem mapping portions of the ear to specific parts of the body and internal organs. The analgesic effects may be mediated through the central nervous system by local release of endorphins through nerve fiber activation and neurotransmitters—including serotonin, dopamine, and norepinephrine—leading to pre- and postsynaptic suppression of pain transmission.

Battlefield acupuncture (BFA) uses 5 set points anatomically located on each ear.⁹ Practitioners use small semipermanent, dartlike acupuncture needles. Patients could experience pain relief in a few minutes, which can last minutes, hours, days, weeks, or months depending on the pathology of the pain. This procedure developed in 2001 has been studied for different pain types and has shown benefit when used for postsurgical pain, chronic spinal cord injury−related neuropathic pain, and general chronic pain, as well as for other indications, such as insomnia, depression, and weight loss.^8,10-13 In 2018, a randomized controlled trial compared postintervention numeric rating scale (NRS) pain scores in patients presenting to the ED with acute or acute-on-chronic lower back pain who received BFA as an adjunct to standard care vs standard care alone.¹⁴ Patients receiving BFA as an adjunct to standard care were found to have mean postintervention pain scores 1.7 points lower than those receiving standard care alone. This study demonstrated that BFA was feasible and well tolerated for lower back pain in the ED as an adjunct to standard care. The study was limited by the adjunct use of BFA rather than as monotherapy and by the practitioners’ discretion regarding standard care, which was not defined by the study’s authors.

The Jesse Brown Veterans Affairs Medical Center (JBVAMC) in Chicago, Illinois, offers several CAM modalities, such as exercise/movement therapy, chiropractic, art/music therapy, and relaxation workshops, which are widely used by veterans. Recent evidence suggests BFA could reduce pain scores as an adjunct or an alternative to pharmacologic therapy. We are interested in how CAM therapies, such as BFA, can help avoid AEs associated with opioid or NSAID therapy.

At the JBVAMC ED, ketorolac 15 mg is the preferred first-line treatment of acute, noncancer pain, based on the results of previous studies. In 2018 BFA was offered first to veterans presenting with acute or acute-on-chronic pain to the ED; however, its effectiveness for pain reduction vs ketorolac has not been evaluated in this patient population. Limited literature is available on BFA and its use in the ED. To our knowledge, this was the first observational study assessing the difference between a single session of BFA vs a single dose of ketorolac in treating noncancer acute or acute-on-chronic pain in the ED.

Methods

This study was a retrospective chart review of patients who presented to the JBVAMC ED with acute pain or acute-on-chronic pain, who received ketorolac or BFA. The study population was generated from a list of all IV and intramuscular (IM) ketorolac unit dose orders verified from June 1, 2018, through August 30, 2019, and a list of all BFA procedure notes signed from June 1, 2018, through August 30, 2019. Patients were included in the study if they had documented administration of IV or IM ketorolac or BFA between June 1, 2018, and August 30, 2019. Patients who received ketorolac doses other than 15 mg, the intervention was administered outside of the ED, received adjunct treatment in addition to the treatment intervention in the ED, had no baseline NRS pain score documented before the intervention, had an NRS pain score of < 4, had no postintervention NRS pain score documented within 6 hours, had a treatment indication other than pain, or had active cancer were excluded. As in previous JBVAMC studies, we used NRS pain score cutoffs (mild, moderate, severe, and very severe) based on Woo and colleagues’ meta-analysis and excluded scores < 4.¹⁵

Endpoints

The primary endpoint was the mean difference in NRS pain score before and after the intervention, determined by comparing the NRS pain score documented at triage to the ED with the first documented NRS pain score at least 30 minutes to 6 hours after treatment administration. The secondary endpoints included the number of patients prescribed pain medication at discharge, the number of patients who were discharged with no medications, and the number of patients admitted to the hospital. The safety endpoint included any AEs of the intervention. Subgroup analyses were performed comparing the mean difference in NRS pain score among subgroups classified by severity of baseline NRS pain score and pain location.

Statistical Analysis

Baseline characteristics and endpoints were analyzed using descriptive statistics. Categorical data were analyzed using Fisher exact test and z test for proportions, and continuous data were compared using t test and paired t test. An 80% power calculation determined that 84 patients per group were needed to detect a statistically significant difference in pain score reduction of 1.3 at a type-1 error rate of 0.05. The sample size was based on a calculation performed in a previously published study that compared IV ketorolac at 3 single-dose regimens for treating acute pain in the ED.¹⁶ The 1.3 pain score reduction is considered the minimum clinically significant difference in pain that could be detected with the NRS.¹⁷

Results

Sixty-one patients received BFA during the study period: 31 were excluded (26 received adjunct treatment in the ED, 2 had active cancer documented, 2 had an indication other than pain, and 1 received BFA outside of the ED), leaving 30 patients in the BFA cohort. During the study period, 1299 patients received ketorolac. These patients were selected using a random number generator and then screened to determine inclusion or exclusion in the study. We continued to randomly select patients for the ketorolac group until we had a similar number in each treatment group. Of these 148 patients who were randomly selected to be reviewed, 116 were excluded: 48 received adjunct treatment in the ED, 24 had no postintervention NRS pain score documented within 6 hours, 18 received ketorolac doses other than 15 mg, 12 received ketorolac outside the ED, 9 had no baseline NRS pain score documented, 3 presented with a NRS pain score of ≤ 3, and 2 had active cancer documented. The ketorolac cohort comprised 31 patients.

Baseline characteristics were similar between the 2 groups except for the average baseline NRS pain score, which was statistically significantly higher in the BFA vs ketorolac group (8.7 vs 7.7, respectively; P = .02). The mean age was 51 years in the BFA group and 48 years in the ketorolac group. Most patients in each cohort were male: 80% in the BFA group and 71% in the ketorolac group. The most common types of pain documented as the chief ED presentation included back, lower extremity, and head.

Endpoints

The mean difference in NRS pain score was 3.9 for the BFA group and 5.1 for the ketorolac group. Both were clinically and statistically significant reductions (P = .03 and P < .01), but the difference between the intervention groups in NRS score reduction was not statistically significant (P = .07).

For the secondary endpoint of outpatient prescriptions written at discharge, there was no significant difference between the groups except for oral NSAIDs, which were more likely to be prescribed to patients who received ketorolac (P = .01).

Subgroup Analysis

An analysis was performed for subgroups classified by baseline NRS pain score (mild: 4; moderate, 5 - 6; severe, 7 - 9; and very severe, 10). Data for mild pain was limited because a small number of patients received interventions. For moderate pain, the mean difference in NRS pain score for BFA and ketorolac was 3.5 and 3.8, respectively; for severe pain, 3.4 and 5.3; and for very severe pain, 4.6 and 6.4. There was a larger difference in the preintervention and postintervention NRS pain scores within severe pain and very severe pain groups.

Discussion

Both interventions resulted in a significant reduction in the mean NRS pain score of about 4 to 5 points within their group, and BFA resulted in a similar NRS pain score reduction compared with ketorolac 15 mg. Because the baseline NRS pain scores were significantly different between the BFA and ketorolac groups, a subgroup analysis revealed that BFA reduced mean NRS pain score in patients with severe and very severe pain but appears to be less beneficial for moderate pain, unlike the ketorolac results that showed a large reduction in all pain groups except for the small sample of patients with mild pain.

In this study, more patients in the BFA group presented to the ED with lower extremity pain, such as gout or neuropathy, compared with the ketorolac group; however, BFA did not result in a significantly different pain score reduction in this subgroup compared with ketorolac. Patients receiving BFA were more likely to receive topical analgesics or muscle relaxants at discharge; whereas those receiving ketorolac were significantly more likely to receive oral NSAIDs. Patients in this study also were more likely to be admitted to the hospital if they received ketorolac; however, for these patients, pain was secondary to their chief presentation, and the admitting physician’s familiarity with ketorolac might have been the reason for choosing this intervention. Reasons for the admissions were surgical observation, psychiatric stabilization, kidney/gallstones, rule out of acute coronary syndrome, pneumonia, and proctitis in the ketorolac group, and suicidal ideations in the BFA group.

Limitations

As a limited number of patients received BFA at JBVAMC, the study was not sufficiently powered to detect a difference in the primary outcome. Because BFA required a consultation to be entered in the electronic health record, in addition to time needed to perform the procedure, practitioners might have preferred IV/IM ketorolac during busy times in the ED, potentially leading to underrepresentation in the BFA group. Prescribing preferences might have differed among the rotating physicians, timing of the documentation of the NRS pain score could have differed based on the treatment intervention, and the investigators were unable to control or accurately assess whether patients had taken an analgesic medication before presenting to the ED. Because pain and the treating physician are subjective, patients who reported a higher baseline pain severity might have been more likely to be discharged with topical analgesics or muscle relaxants. One way to correct for this subjectivity would be to conduct a larger prospective trial with a single treating physician. Finally, ED encounters in this study were short, and there was no follow-up permitting identification of AEs.

Conclusions

NRS pain score reduction with BFA did not differ compared with ketorolac 15 mg for treating acute and acute-on-chronic pain in the ED. Although this study was underpowered, these results add to the limited existing literature, suggesting that both interventions could result in clinically significant pain score reductions for patients presenting to the ED with severe and very severe pain, making BFA a viable nonpharmacologic option. Future studies could include investigating the benefit of BFA in the veteran population by studying larger samples in the ED, surveying patients after their interventions to identify rates AEs, and exploring the use of BFA for chronic pain in the outpatient setting.

CASE A difficult withdrawal

Three days after he stops drinking alcohol, Mr. G, age 49, presents to a detoxification center with his wife, who drove him there because she was concerned about his condition. She says her husband had been drinking alcohol every night for as long as she can remember. Despite numerous admissions to rehabilitation centers, Mr. G usually would resume drinking soon after he was discharged. Three days ago, Mr. G’s wife had told him she “could not take it anymore,” so he got rid of all his alcohol and stopped drinking. Mr. G’s wife felt he was doing fine the first day, but his condition increasingly worsened the second and third days. The triage nurse who attempts to interview Mr. G finds him tremulous, vomiting, and sweating. She notices that he seems preoccupied with pulling at his shirt, appearing to pick at things that are not there.

HISTORY Untreated depression, other comorbidities

Mr. G’s wife says he has never been psychiatrically hospitalized or exhibited suicidal behavior. Mr. G previously received care from a psychiatrist, who diagnosed him with major depressive disorder (MDD) and prescribed an antidepressant, though his wife cannot recall which specific medication. She shares it has been “a long time” since Mr. G has taken the antidepressant and the last time he received treatment for his MDD was 5 years ago. Mr. G’s wife says her husband had once abstained from alcohol use for >6 months following one of his stints at a rehabilitation center. She is not able to share many other details about Mr. G’s previous stays at rehabilitation centers, but says he always had “a rough time.”

She says Mr. G had been drinking an average of 10 drinks each night, usually within 4 hours. He has no history of nicotine or illicit substance use and has held a corporate job for the last 18 years. Several years ago, a physician had diagnosed Mr. G with hypertension and high cholesterol, but he did not follow up for treatment. Mr. G’s wife also recalls a physician told her husband he had a fatty liver. His family history includes heart disease and cancer.

[polldaddy:12041618]

The author’s observations

The treatment team observed several elements of alcohol withdrawal and classified Mr. G as a priority patient. If the team had completed the Clinical Institute Withdrawal Assessment for Alcohol–Revised scale (CIWA-Ar) (Table 1¹), Mr. G would score ≥10. While the protocol for initiating treatment for patients experiencing alcohol withdrawal varies by institution, patients with moderate to severe scores on the CIWA-Ar when experiencing withdrawal typically are managed with pharmacotherapy to address their symptoms.¹ Given the timeline of his last drink as reported by his wife, Mr. G is on the brink of experiencing a cascade of symptoms concerning for delirium tremens (DTs).²Table 2² provides a timeline and symptoms related to alcohol withdrawal. To prevent further exacerbation of symptoms, which could lead to DTs, Mr. G’s treatment team will likely initiate a benzodiazepine, using either scheduled or symptom-driven dosing.³

Two neurotransmitters that play a role in DTs are glutamate (excitatory) and GABA (inhibitory). In a normal state, the competing actions of these neurotransmitters balance each other. Acute alcohol intake causes a shift in the excitatory and inhibitory levels, with more inhibition taking place, thus causing disequilibrium. If chronic alcohol use continues, the amount of GABA inhibition reduction is related to downregulation of receptors.^2,4 Excitation increases by way of upregulation of the N-methyl-D-aspartate receptors.^2,4,5 The goal is to achieve equilibrium of the neurotransmitters, even though the balance is different from when alcohol was not present.^2,4

If alcohol is suddenly removed following chronic use, there is unchecked glutamate excitation related to a blunted GABA state. This added increase in the excitation of glutamate leads to withdrawal symptoms.^2,4Table 3^2,4,5 depicts the neuro­transmitter equilibrium of GABA and glutamate relative to alcohol use.

EVALUATION Bleeding gums and bruising

The treatment team admits Mr. G to the triage bay and contacts the addiction psychiatrist. The physician orders laboratory tests to assess nutritional deficits and electrolyte abnormalities. Mr. G is also placed on routine assessments with symptom-triggered therapy. An assessment reveals bleeding gums and bruises, which are believed to be a result of thrombocytopenia (low blood platelet count).

[polldaddy:12041627]

Continue to: The author's observations

Though regular clinical assessment of PEth varies, it is considered to have high sensitivity and specificity to detect alcohol use.⁶ When ethanol is present, the phospholipase D enzyme acts upon phosphatidylcholine, forming a direct biomarker, PEth, on the surface of the red blood cell.^6,7 PEth’s half-life ranges from 4.5 to 12 days,⁶ and it can be detected in blood for 3 to 4 weeks after alcohol ingestion.^6,7 A PEth value <20 ng/mL indicates light or no alcohol consumption; 20 to 199 ng/mL indicates significant consumption; and >200 ng/mL indicates heavy consumption.⁷ Since Mr. G has a history of chronic alcohol use, his PEth level is expected to be >200 ng/mL.

AST/ALT and MCV are indirect biomarkers, meaning the tests are not alcohol-specific and the role of alcohol is instead observed by the damage to the body with excessive use over time.⁷ The expected AST:ALT ratio is 2:1. This is related to 3 mechanisms. The first is a decrease in ALT usually relative to B6 deficiency in individuals with alcohol use disorder (AUD). Another mechanism is related to alcohol’s propensity to affect mitochondria, which is a source for AST. Additionally, AST is also found in higher proportions in the kidneys, heart, and muscles.⁸

An MCV <100 fL would be within the normal range (80 to 100 fL) for red blood cells. While the reasons for an enlarged red blood cell (or macrocyte) are extensive, alcohol can be a factor once other causes are excluded. Additional laboratory tests and a peripheral blood smear test can help in this investigation.Alcohol disrupts the complete maturation of red blood cells.^9,10 If the cause of the macrocyte is alcohol-related and alcohol use is terminated, those enlarged cells can resolve in an average of 3 months.⁹

Vitamin B1 levels >200 nmol/L would be within normal range (74 to 222 nmol/L). Mr. G’s chronic alcohol use would likely cause him to be vitamin B1–deficient. The deficiency is usually related to diet, malabsorption, and the cells’ impaired ability to utilize vitamin B1. A consequence of vitamin B1 deficiency is Wernicke-Korsakoff syndrome.¹¹

Due to his chronic alcohol use, Mr. G’s magnesium stores most likely would be below normal range (1.7 to 2.2 mg/dL). Acting as a diuretic, alcohol depletes magnesium and other electrolytes. The intracellular shift that occurs to balance the deficit causes the body to use its normal stores of magnesium, which leads to further magnesium depletion. Other common causes include nutritional deficiency and decreased gastrointestinal absorption.¹² The bleeding the physician suspected was a result of drinking likely occurred through direct and indirect mechanisms that affect platelets.^9,13 Platelets can show improvement 1 week after drinking cessation. Some evidence suggests the risk of seizure or DTs increases significantly with a platelet count <119,000 µL per unit of blood.¹³

Continue to: TREATMENT Pharmacotherapy for alcohol use disorder

TREATMENT Pharmacotherapy for alcohol use disorder

As Mr. G’s condition starts to stabilize, he discusses treatment options for AUD with his physician. At the end of the discussion, Mr. G expresses an interest in starting a medication. The doctor reviews his laboratory results and available treatment options.

[polldaddy:12041630]

The author’s observations

Of the 3 FDA-approved medications for treating AUD (disulfiram, acamprosate, and naltrexone), naltrexone has been shown to decrease heavy drinking days^5,14 and comes in oral and injectable forms. Reducing drinking is achieved by reducing the rewarding effects of alcohol^5,14 and alcohol cravings.⁵ Disulfiram often has poor adherence, and like acamprosate it may be more helpful for maintenance of abstinence.Neither topiramate nor gabapentin are FDA-approved for AUD but may be used for their affects on GABA.⁵ Gabapentin may also help patients experiencing alcohol withdrawal syndrome.^5,15 Mr. G did not have any concomitant medications or comorbid medical conditions, but these factors as well as any renal or hepatic dysfunction must be considered before initiating any medications.

OUTCOME Improved well-being

Mr. G’s treatment team initiates oral naltrexone 50 mg/d, which he tolerates well without complications. He stops drinking entirely and expresses an interest in transitioning to an injectable form of naltrexone in the future. In addition to taking medication, Mr. G wants to participate in psychotherapy. Mr. G thanks his team for the care he received in the hospital, telling them, “You all saved my life.” As he discusses his past issues with alcohol, Mr. G asks his physician how he could get involved to make changes to reduce excessive alcohol consumption in his community (Box^5,15-21).

Box

Bottom Line

Patients experiencing alcohol withdrawal can present with several layers of complexity. Failure to achieve acute stabilization may be life-threatening. After providing critical care, promptly start alcohol use disorder treatment for patients who expresses a desire to change.

Related Resources

• American Society of Addiction Medicine. The ASAM Clinical Practice Guideline on Alcohol Withdrawal Management. https://www.asam.org/quality-care/clinical-guidelines/alcohol-withdrawal-management-guideline
• American Psychiatric Association. The American Psychiatric Association Practice Guideline for the Pharmacological Treatment of Patients With Alcohol Use Disorder. American Psychiatric Association Publishing; 2018.

Drug Brand Names

Acamprosate • Campral
Disulfiram • Antabuse
Gabapentin • Neurontin
Naltrexone (injection) • Vivitrol
Naltrexone (oral) • ReVia
Topiramate • Topamax

CASE A difficult withdrawal

Three days after he stops drinking alcohol, Mr. G, age 49, presents to a detoxification center with his wife, who drove him there because she was concerned about his condition. She says her husband had been drinking alcohol every night for as long as she can remember. Despite numerous admissions to rehabilitation centers, Mr. G usually would resume drinking soon after he was discharged. Three days ago, Mr. G’s wife had told him she “could not take it anymore,” so he got rid of all his alcohol and stopped drinking. Mr. G’s wife felt he was doing fine the first day, but his condition increasingly worsened the second and third days. The triage nurse who attempts to interview Mr. G finds him tremulous, vomiting, and sweating. She notices that he seems preoccupied with pulling at his shirt, appearing to pick at things that are not there.

HISTORY Untreated depression, other comorbidities

Mr. G’s wife says he has never been psychiatrically hospitalized or exhibited suicidal behavior. Mr. G previously received care from a psychiatrist, who diagnosed him with major depressive disorder (MDD) and prescribed an antidepressant, though his wife cannot recall which specific medication. She shares it has been “a long time” since Mr. G has taken the antidepressant and the last time he received treatment for his MDD was 5 years ago. Mr. G’s wife says her husband had once abstained from alcohol use for >6 months following one of his stints at a rehabilitation center. She is not able to share many other details about Mr. G’s previous stays at rehabilitation centers, but says he always had “a rough time.”

She says Mr. G had been drinking an average of 10 drinks each night, usually within 4 hours. He has no history of nicotine or illicit substance use and has held a corporate job for the last 18 years. Several years ago, a physician had diagnosed Mr. G with hypertension and high cholesterol, but he did not follow up for treatment. Mr. G’s wife also recalls a physician told her husband he had a fatty liver. His family history includes heart disease and cancer.

[polldaddy:12041618]

The author’s observations

The treatment team observed several elements of alcohol withdrawal and classified Mr. G as a priority patient. If the team had completed the Clinical Institute Withdrawal Assessment for Alcohol–Revised scale (CIWA-Ar) (Table 1¹), Mr. G would score ≥10. While the protocol for initiating treatment for patients experiencing alcohol withdrawal varies by institution, patients with moderate to severe scores on the CIWA-Ar when experiencing withdrawal typically are managed with pharmacotherapy to address their symptoms.¹ Given the timeline of his last drink as reported by his wife, Mr. G is on the brink of experiencing a cascade of symptoms concerning for delirium tremens (DTs).²Table 2² provides a timeline and symptoms related to alcohol withdrawal. To prevent further exacerbation of symptoms, which could lead to DTs, Mr. G’s treatment team will likely initiate a benzodiazepine, using either scheduled or symptom-driven dosing.³

Two neurotransmitters that play a role in DTs are glutamate (excitatory) and GABA (inhibitory). In a normal state, the competing actions of these neurotransmitters balance each other. Acute alcohol intake causes a shift in the excitatory and inhibitory levels, with more inhibition taking place, thus causing disequilibrium. If chronic alcohol use continues, the amount of GABA inhibition reduction is related to downregulation of receptors.^2,4 Excitation increases by way of upregulation of the N-methyl-D-aspartate receptors.^2,4,5 The goal is to achieve equilibrium of the neurotransmitters, even though the balance is different from when alcohol was not present.^2,4

If alcohol is suddenly removed following chronic use, there is unchecked glutamate excitation related to a blunted GABA state. This added increase in the excitation of glutamate leads to withdrawal symptoms.^2,4Table 3^2,4,5 depicts the neuro­transmitter equilibrium of GABA and glutamate relative to alcohol use.

EVALUATION Bleeding gums and bruising

The treatment team admits Mr. G to the triage bay and contacts the addiction psychiatrist. The physician orders laboratory tests to assess nutritional deficits and electrolyte abnormalities. Mr. G is also placed on routine assessments with symptom-triggered therapy. An assessment reveals bleeding gums and bruises, which are believed to be a result of thrombocytopenia (low blood platelet count).

[polldaddy:12041627]

Continue to: The author's observations

Though regular clinical assessment of PEth varies, it is considered to have high sensitivity and specificity to detect alcohol use.⁶ When ethanol is present, the phospholipase D enzyme acts upon phosphatidylcholine, forming a direct biomarker, PEth, on the surface of the red blood cell.^6,7 PEth’s half-life ranges from 4.5 to 12 days,⁶ and it can be detected in blood for 3 to 4 weeks after alcohol ingestion.^6,7 A PEth value <20 ng/mL indicates light or no alcohol consumption; 20 to 199 ng/mL indicates significant consumption; and >200 ng/mL indicates heavy consumption.⁷ Since Mr. G has a history of chronic alcohol use, his PEth level is expected to be >200 ng/mL.

AST/ALT and MCV are indirect biomarkers, meaning the tests are not alcohol-specific and the role of alcohol is instead observed by the damage to the body with excessive use over time.⁷ The expected AST:ALT ratio is 2:1. This is related to 3 mechanisms. The first is a decrease in ALT usually relative to B6 deficiency in individuals with alcohol use disorder (AUD). Another mechanism is related to alcohol’s propensity to affect mitochondria, which is a source for AST. Additionally, AST is also found in higher proportions in the kidneys, heart, and muscles.⁸

An MCV <100 fL would be within the normal range (80 to 100 fL) for red blood cells. While the reasons for an enlarged red blood cell (or macrocyte) are extensive, alcohol can be a factor once other causes are excluded. Additional laboratory tests and a peripheral blood smear test can help in this investigation.Alcohol disrupts the complete maturation of red blood cells.^9,10 If the cause of the macrocyte is alcohol-related and alcohol use is terminated, those enlarged cells can resolve in an average of 3 months.⁹

Vitamin B1 levels >200 nmol/L would be within normal range (74 to 222 nmol/L). Mr. G’s chronic alcohol use would likely cause him to be vitamin B1–deficient. The deficiency is usually related to diet, malabsorption, and the cells’ impaired ability to utilize vitamin B1. A consequence of vitamin B1 deficiency is Wernicke-Korsakoff syndrome.¹¹

Due to his chronic alcohol use, Mr. G’s magnesium stores most likely would be below normal range (1.7 to 2.2 mg/dL). Acting as a diuretic, alcohol depletes magnesium and other electrolytes. The intracellular shift that occurs to balance the deficit causes the body to use its normal stores of magnesium, which leads to further magnesium depletion. Other common causes include nutritional deficiency and decreased gastrointestinal absorption.¹² The bleeding the physician suspected was a result of drinking likely occurred through direct and indirect mechanisms that affect platelets.^9,13 Platelets can show improvement 1 week after drinking cessation. Some evidence suggests the risk of seizure or DTs increases significantly with a platelet count <119,000 µL per unit of blood.¹³

Continue to: TREATMENT Pharmacotherapy for alcohol use disorder

TREATMENT Pharmacotherapy for alcohol use disorder

As Mr. G’s condition starts to stabilize, he discusses treatment options for AUD with his physician. At the end of the discussion, Mr. G expresses an interest in starting a medication. The doctor reviews his laboratory results and available treatment options.

[polldaddy:12041630]

The author’s observations

Of the 3 FDA-approved medications for treating AUD (disulfiram, acamprosate, and naltrexone), naltrexone has been shown to decrease heavy drinking days^5,14 and comes in oral and injectable forms. Reducing drinking is achieved by reducing the rewarding effects of alcohol^5,14 and alcohol cravings.⁵ Disulfiram often has poor adherence, and like acamprosate it may be more helpful for maintenance of abstinence.Neither topiramate nor gabapentin are FDA-approved for AUD but may be used for their affects on GABA.⁵ Gabapentin may also help patients experiencing alcohol withdrawal syndrome.^5,15 Mr. G did not have any concomitant medications or comorbid medical conditions, but these factors as well as any renal or hepatic dysfunction must be considered before initiating any medications.

OUTCOME Improved well-being

Mr. G’s treatment team initiates oral naltrexone 50 mg/d, which he tolerates well without complications. He stops drinking entirely and expresses an interest in transitioning to an injectable form of naltrexone in the future. In addition to taking medication, Mr. G wants to participate in psychotherapy. Mr. G thanks his team for the care he received in the hospital, telling them, “You all saved my life.” As he discusses his past issues with alcohol, Mr. G asks his physician how he could get involved to make changes to reduce excessive alcohol consumption in his community (Box^5,15-21).

Box

Bottom Line

Patients experiencing alcohol withdrawal can present with several layers of complexity. Failure to achieve acute stabilization may be life-threatening. After providing critical care, promptly start alcohol use disorder treatment for patients who expresses a desire to change.

Related Resources

• American Society of Addiction Medicine. The ASAM Clinical Practice Guideline on Alcohol Withdrawal Management. https://www.asam.org/quality-care/clinical-guidelines/alcohol-withdrawal-management-guideline
• American Psychiatric Association. The American Psychiatric Association Practice Guideline for the Pharmacological Treatment of Patients With Alcohol Use Disorder. American Psychiatric Association Publishing; 2018.

Drug Brand Names

Acamprosate • Campral
Disulfiram • Antabuse
Gabapentin • Neurontin
Naltrexone (injection) • Vivitrol
Naltrexone (oral) • ReVia
Topiramate • Topamax

CASE A difficult withdrawal

Three days after he stops drinking alcohol, Mr. G, age 49, presents to a detoxification center with his wife, who drove him there because she was concerned about his condition. She says her husband had been drinking alcohol every night for as long as she can remember. Despite numerous admissions to rehabilitation centers, Mr. G usually would resume drinking soon after he was discharged. Three days ago, Mr. G’s wife had told him she “could not take it anymore,” so he got rid of all his alcohol and stopped drinking. Mr. G’s wife felt he was doing fine the first day, but his condition increasingly worsened the second and third days. The triage nurse who attempts to interview Mr. G finds him tremulous, vomiting, and sweating. She notices that he seems preoccupied with pulling at his shirt, appearing to pick at things that are not there.

HISTORY Untreated depression, other comorbidities

Mr. G’s wife says he has never been psychiatrically hospitalized or exhibited suicidal behavior. Mr. G previously received care from a psychiatrist, who diagnosed him with major depressive disorder (MDD) and prescribed an antidepressant, though his wife cannot recall which specific medication. She shares it has been “a long time” since Mr. G has taken the antidepressant and the last time he received treatment for his MDD was 5 years ago. Mr. G’s wife says her husband had once abstained from alcohol use for >6 months following one of his stints at a rehabilitation center. She is not able to share many other details about Mr. G’s previous stays at rehabilitation centers, but says he always had “a rough time.”

She says Mr. G had been drinking an average of 10 drinks each night, usually within 4 hours. He has no history of nicotine or illicit substance use and has held a corporate job for the last 18 years. Several years ago, a physician had diagnosed Mr. G with hypertension and high cholesterol, but he did not follow up for treatment. Mr. G’s wife also recalls a physician told her husband he had a fatty liver. His family history includes heart disease and cancer.

[polldaddy:12041618]

The author’s observations

The treatment team observed several elements of alcohol withdrawal and classified Mr. G as a priority patient. If the team had completed the Clinical Institute Withdrawal Assessment for Alcohol–Revised scale (CIWA-Ar) (Table 1¹), Mr. G would score ≥10. While the protocol for initiating treatment for patients experiencing alcohol withdrawal varies by institution, patients with moderate to severe scores on the CIWA-Ar when experiencing withdrawal typically are managed with pharmacotherapy to address their symptoms.¹ Given the timeline of his last drink as reported by his wife, Mr. G is on the brink of experiencing a cascade of symptoms concerning for delirium tremens (DTs).²Table 2² provides a timeline and symptoms related to alcohol withdrawal. To prevent further exacerbation of symptoms, which could lead to DTs, Mr. G’s treatment team will likely initiate a benzodiazepine, using either scheduled or symptom-driven dosing.³

Two neurotransmitters that play a role in DTs are glutamate (excitatory) and GABA (inhibitory). In a normal state, the competing actions of these neurotransmitters balance each other. Acute alcohol intake causes a shift in the excitatory and inhibitory levels, with more inhibition taking place, thus causing disequilibrium. If chronic alcohol use continues, the amount of GABA inhibition reduction is related to downregulation of receptors.^2,4 Excitation increases by way of upregulation of the N-methyl-D-aspartate receptors.^2,4,5 The goal is to achieve equilibrium of the neurotransmitters, even though the balance is different from when alcohol was not present.^2,4

If alcohol is suddenly removed following chronic use, there is unchecked glutamate excitation related to a blunted GABA state. This added increase in the excitation of glutamate leads to withdrawal symptoms.^2,4Table 3^2,4,5 depicts the neuro­transmitter equilibrium of GABA and glutamate relative to alcohol use.

EVALUATION Bleeding gums and bruising

The treatment team admits Mr. G to the triage bay and contacts the addiction psychiatrist. The physician orders laboratory tests to assess nutritional deficits and electrolyte abnormalities. Mr. G is also placed on routine assessments with symptom-triggered therapy. An assessment reveals bleeding gums and bruises, which are believed to be a result of thrombocytopenia (low blood platelet count).

[polldaddy:12041627]

Continue to: The author's observations

Though regular clinical assessment of PEth varies, it is considered to have high sensitivity and specificity to detect alcohol use.⁶ When ethanol is present, the phospholipase D enzyme acts upon phosphatidylcholine, forming a direct biomarker, PEth, on the surface of the red blood cell.^6,7 PEth’s half-life ranges from 4.5 to 12 days,⁶ and it can be detected in blood for 3 to 4 weeks after alcohol ingestion.^6,7 A PEth value <20 ng/mL indicates light or no alcohol consumption; 20 to 199 ng/mL indicates significant consumption; and >200 ng/mL indicates heavy consumption.⁷ Since Mr. G has a history of chronic alcohol use, his PEth level is expected to be >200 ng/mL.

AST/ALT and MCV are indirect biomarkers, meaning the tests are not alcohol-specific and the role of alcohol is instead observed by the damage to the body with excessive use over time.⁷ The expected AST:ALT ratio is 2:1. This is related to 3 mechanisms. The first is a decrease in ALT usually relative to B6 deficiency in individuals with alcohol use disorder (AUD). Another mechanism is related to alcohol’s propensity to affect mitochondria, which is a source for AST. Additionally, AST is also found in higher proportions in the kidneys, heart, and muscles.⁸

An MCV <100 fL would be within the normal range (80 to 100 fL) for red blood cells. While the reasons for an enlarged red blood cell (or macrocyte) are extensive, alcohol can be a factor once other causes are excluded. Additional laboratory tests and a peripheral blood smear test can help in this investigation.Alcohol disrupts the complete maturation of red blood cells.^9,10 If the cause of the macrocyte is alcohol-related and alcohol use is terminated, those enlarged cells can resolve in an average of 3 months.⁹

Vitamin B1 levels >200 nmol/L would be within normal range (74 to 222 nmol/L). Mr. G’s chronic alcohol use would likely cause him to be vitamin B1–deficient. The deficiency is usually related to diet, malabsorption, and the cells’ impaired ability to utilize vitamin B1. A consequence of vitamin B1 deficiency is Wernicke-Korsakoff syndrome.¹¹

Due to his chronic alcohol use, Mr. G’s magnesium stores most likely would be below normal range (1.7 to 2.2 mg/dL). Acting as a diuretic, alcohol depletes magnesium and other electrolytes. The intracellular shift that occurs to balance the deficit causes the body to use its normal stores of magnesium, which leads to further magnesium depletion. Other common causes include nutritional deficiency and decreased gastrointestinal absorption.¹² The bleeding the physician suspected was a result of drinking likely occurred through direct and indirect mechanisms that affect platelets.^9,13 Platelets can show improvement 1 week after drinking cessation. Some evidence suggests the risk of seizure or DTs increases significantly with a platelet count <119,000 µL per unit of blood.¹³

Continue to: TREATMENT Pharmacotherapy for alcohol use disorder

TREATMENT Pharmacotherapy for alcohol use disorder

As Mr. G’s condition starts to stabilize, he discusses treatment options for AUD with his physician. At the end of the discussion, Mr. G expresses an interest in starting a medication. The doctor reviews his laboratory results and available treatment options.

[polldaddy:12041630]

The author’s observations

Of the 3 FDA-approved medications for treating AUD (disulfiram, acamprosate, and naltrexone), naltrexone has been shown to decrease heavy drinking days^5,14 and comes in oral and injectable forms. Reducing drinking is achieved by reducing the rewarding effects of alcohol^5,14 and alcohol cravings.⁵ Disulfiram often has poor adherence, and like acamprosate it may be more helpful for maintenance of abstinence.Neither topiramate nor gabapentin are FDA-approved for AUD but may be used for their affects on GABA.⁵ Gabapentin may also help patients experiencing alcohol withdrawal syndrome.^5,15 Mr. G did not have any concomitant medications or comorbid medical conditions, but these factors as well as any renal or hepatic dysfunction must be considered before initiating any medications.

OUTCOME Improved well-being

Mr. G’s treatment team initiates oral naltrexone 50 mg/d, which he tolerates well without complications. He stops drinking entirely and expresses an interest in transitioning to an injectable form of naltrexone in the future. In addition to taking medication, Mr. G wants to participate in psychotherapy. Mr. G thanks his team for the care he received in the hospital, telling them, “You all saved my life.” As he discusses his past issues with alcohol, Mr. G asks his physician how he could get involved to make changes to reduce excessive alcohol consumption in his community (Box^5,15-21).

Box

Bottom Line

Patients experiencing alcohol withdrawal can present with several layers of complexity. Failure to achieve acute stabilization may be life-threatening. After providing critical care, promptly start alcohol use disorder treatment for patients who expresses a desire to change.

Related Resources

• American Society of Addiction Medicine. The ASAM Clinical Practice Guideline on Alcohol Withdrawal Management. https://www.asam.org/quality-care/clinical-guidelines/alcohol-withdrawal-management-guideline
• American Psychiatric Association. The American Psychiatric Association Practice Guideline for the Pharmacological Treatment of Patients With Alcohol Use Disorder. American Psychiatric Association Publishing; 2018.

Drug Brand Names

Acamprosate • Campral
Disulfiram • Antabuse
Gabapentin • Neurontin
Naltrexone (injection) • Vivitrol
Naltrexone (oral) • ReVia
Topiramate • Topamax

Diversity of health care professionals improves medical outcomes and quality of life in patients. ¹ There is a lack of diversity in dermatology, with only 4.2% of dermatologists identifying as Hispanic and 3% identifying as African American, ² possibly due to a lack of early exposure to dermatology among high school and undergraduate students, a low number of underrepresented students in medical school, a lack of formal mentorship programs geared to underrepresented students, and implicit biases. ^1-4 Furthermore, the field is competitive, with many more applicants than available positions. In 2022, there were 851 applicants competing for 492 residency positions in dermatology. ⁵ Thus, it is important to educate young students about dermatology and understand root causes as to why the number of u nderrepresented in medicine (UiM) dermatologists remains stagnant.

According to Pritchett et al,⁴ it is crucial for dermatologists to interact with high school and college students to foster an early interest in dermatology. Many racial minority students do not progress from high school to college and then from college to medical school, which leaves a substantially reduced number of eligible UiM applicants who can progress into dermatology.⁶ Increasing the amount of UiM students going to medical school requires early mediation. Collaborating with pre-existing premedical school organizations through presentations and workshops is another way to promote an early interest in dermatology.⁴ Special consideration should be given to students who are UiM.

Among the general medical school curriculum, requirements for exposure to dermatology are not high. In one study, the median number of clinical and preclinical hours required was 10. Furthermore, 20% of 33 medical schools did not require preclinical dermatology hours (hours done before medical school rotations begin and in an academic setting), 36% required no clinical hours (rotational hours), 8% required no dermatology hours whatsoever, and only 10% required clinical dermatology rotation.³ Based on these findings, it is clear that dermatology is not well incorporated into medical school curricula. Furthermore, curricula have historically neglected to display adequate representation of skin of color.⁷ As a result, medical students generally have limited exposure to dermatology³ and are exposed even less to presentations of dermatologic issues in historically marginalized populations.⁷

Given the paucity of research on UiM students’ perceptions of dermatology prior to medical school, our cross-sectional survey study sought to evaluate the level of interest in dermatology of UiM premedical undergraduates. This survey specifically evaluated exposure to dermatology, preconceived notions about the field, and mentorship opportunities. By understanding these factors, dermatologists and dermatology residency programs can use this information to create mentorship opportunities and better adjust existing programs to meet students’ needs.

Methods

A 19-question multiple-choice survey was administered electronically (SurveyMonkey) in May 2020 to premedical students at Howard University (Washington, DC). One screening question was used: “What is your major?” Those who considered themselves a science major and/or with premedical interest were allowed to complete the survey. All students surveyed were members of the Health Professions Society at Howard University. Students who were interested in pursuing medical school were invited to respond. Approval for this study was obtained from the Howard University institutional review board (FWA00000891).

The survey was divided into 3 sections: Demographics, Exposure to Medicine and Dermatology, and Perceptions of Dermatology. The Demographics section addressed gender, age, and race/ethnicity. The Exposure to Medicine and Dermatology section addressed interest in attending medical school, shadowing experience, exposure to dermatology, and mentoring. The Perceptions of Dermatology section addressed preconceived notions about the field (eg, “dermatology is interesting and exciting”).

Statistical Analysis—The data represented are percentages based on the number of respondents who answered each question. Answers in response to “Please enter any comments” were organized into themes, and the number of respondents who discussed each theme was quantified into a table.

A total of 271 survey invitations were sent to premedical students at Howard University. Students were informed of the study protocol and asked to consent before proceeding to have their responses anonymously collected. Based on the screening question, 152 participants qualified for the survey, and 152 participants completed it (response rate, 56%; completion rate, 100%). Participants were asked to complete the survey only once.

Demographics—Eighty-four percent of respondents identified as science majors, and the remaining 16% identified as nonscience premedical. Ninety-four percent of participants identified as Black or African American; 3% as Asian or Asian American; and the remaining 3% as Other. Most respondents were female (82%), 16% were male, and 2% were either nonbinary or preferred not to answer. Ninety-nine percent were aged 18 to 24 years, and 1% were aged 25 to 34 years (Table 1).

Exposure to Medicine and Dermatology—Ninety-three percent of participants planned on attending medical school, and most students developed an interest in medicine from an early age. Ninety-six percent cited that they became interested in medicine prior to beginning their undergraduate education, and 4% developed an interest as freshmen or sophomores. When asked what led to their interest in medicine, family influence had the single greatest impact on students’ decision to pursue medicine (33%). Classes/school were the second most influential factor (24%), followed by volunteering (15%), shadowing (13%), other (7%), and peer influence (3%)(Figure 1).

Many (56%) premedical students surveyed had shadowing experience to varying degrees. Approximately 18% had fewer than 8 hours of shadowing experience, 24% had 8 to 40 hours, and 14% had more than 40 hours. However, many (43%) premedical students had no shadowing experience (Figure 2). Similarly, 30% of premedical students responded to having a physician as a mentor.

Regarding exposure to dermatology, 42% of premedical students had none. However, 58% of students had exposure to dermatology by being a patient themselves, 40% through seeing a dermatologist with a family member, 21% through seeing a dermatologist on television or social media, 5% through shadowing or volunteering, 3% through mentorship, and 1% through dermatology research (Figure 3).

Of students who said they were interested in dermatology (32%), 16% developed their interest before undergraduate education, while 9% developed interest in their freshman or sophomore year and 7% in their junior or senior year of undergraduate education. Three percent of respondents indicated that they had a dermatology mentorship.

Perceptions of Dermatology—To further evaluate the level of interest that UiM premedical students have in the field of dermatology, students were asked how much they agree or disagree on whether the field of dermatology is interesting. Sixty-three percent of the students agreed that the field of dermatology is interesting, 34% remained uncertain, and 3% disagreed. Additionally, students were asked whether they would consider dermatology as a career; 54% of respondents would consider dermatology as a career, 30% remained uncertain, and 16% would not consider dermatology as a career choice.

Nearly all (95%) students agreed that dermatologists do valuable work that goes beyond the scope of cosmetic procedures such as neuromodulators, fillers, chemical peels, and lasers. Some students also noted they had personal experiences interacting with a dermatologist. For example, one student described visiting the dermatologist many times to get a treatment regimen for their eczema.

Overall themes from the survey are depicted in Table 2. Major themes found in the comments included the desire for more dermatology-related opportunities, mentorship, exposure, connections, and a discussion of disparities faced by Black patients and students within dermatology. Students also expressed an interest in dermatology and the desire to learn more about the specialty.

Comment

Interest in Dermatology—In this cross-sectional survey study of 152 UiM undergraduate students, it was found that many students were interested in dermatology as a career, and more than 70% would be interested in attending events that increased exposure to the field of dermatology. Of the students who had any exposure to dermatology, less than 5% had shadowed an actual dermatologist. The survey showed that there is great potential interest in exposing UiM undergraduate students to the field of dermatology. We found that UiM students are interested in learning more about dermatology, with 80% indicating that they would be willing to participate in dermatology-focused events if they were available. Overall, students mentioned a lack of opportunities, mentorship, exposure, and connections in dermatology despite their interest in the field.

Racial Disparities in Dermatology—Additionally, students discussed disparities they encountered with dermatology due to a lack of patient-provider race concordance and the perceived difference in care when encountering a race-concordant dermatologist. One student noted that they went to multiple White dermatologists for their eczema, and “it wasn’t until I was evaluated by a Black dermatologist (diagnosed with eczema as well) [that I was] prescribed . . . the perfect medication.” Another student noted how a Black dermatologist sparked their interest in getting to know more about the field and remarked that they “think it is an important field that lacks representation for Black people.” This research stresses the need for more dermatology mentorship among UiM undergraduates.

Family Influence on Career Selection—The majority of UiM students in our study became interested in medicine because of family, which is consistent with other studies. In a cross-sectional survey of 300 Pakistani students (150 medical and 150 nonmedical), 87% of students stated that their family had an influence on their career selection.⁸ In another study of 15 junior doctors in Sierra Leone, the most common reasons for pursuing medicine were the desire to help and familial and peer influence.⁹ This again showcases how family can have a positive impact on career selection for medical professionals and highlights the need for early intervention.

Shadowing—One way in which student exposure to dermatology can be effectively increased is by shadowing. In a study evaluating a 30-week shadowing program at the Pediatric Continuity Clinic in Los Angeles, California, a greater proportion of premedical students believed they had a good understanding of the job of a resident physician after the program’s completion compared to before starting the program (an increase from 78% to 100%).¹⁰ The proportion of students reporting a good understanding of the patient-physician relationship after completing the program also increased from 33% to 78%. Furthermore, 72% of the residents stated that having the undergraduates in the clinic was a positive experience.¹⁰ Thus, increasing shadowing opportunities is one extremely effective way to increase student knowledge and awareness of and exposure to dermatology.

Dermatology Mentors—Although 32% of students were interested in dermatology, 3% of students had mentorship in dermatology. In prior studies, it has been shown that mentorship is of great importance in student success and interest in pursuing a specialty. A report from the Association of American Medical Colleges 2019 Medical School Graduation Questionnaire found that the third most influential factor (52.1%) in specialty selection was role model influence.¹¹ In fact, having a role model is consistently one of the top 3 influences on student specialty choice and interest in the last 5 years of survey research. Some studies also have shown mentorship as a positive influence in specialty interest at the undergraduate and graduate levels. A study on an undergraduate student interest group noted that surgeon mentorship and exposure were positive factors to students’ interests in surgery.¹² In fact, the Association of American Medical Colleges noted that some surgical specialties, such as orthopedic surgery, had 45% of respondents who were interested in the specialty before medical school pursue their initial preference in medical school.¹³ Another survey corroborated these findings; more orthopedic-bound students compared with other specialties indicated they were more likely to pursue their field because of experiences prior to medical school.¹⁴

One of the reasons students might not have been exposed to as many opportunities for mentorship in dermatology is because the specialty is one of the smaller fields in medicine and tends to be concentrated in more well-resourced metropolitan areas.¹⁵ Dermatologists make up only 1.3% of the physician workforce.¹⁶ Because there might not be as much exposure to the field, students might also explore their interests in dermatology through other fields, such as through shadowing and observing primary care physicians who often treat patients with dermatologic issues. Skin diseases are a common reason for primary care visits, and one study suggested dermatologic diseases can make up approximately 8.4% of visits in primary care.¹⁷

Moreover, only 1% of medical schools require an elective in dermatology.¹⁸ With exposure being a crucial component to pursuing the specialty, it also is important to pursue formal mentorship within the specialty itself. One study noted that formal mentorship in dermatology was important for most (67%) respondents when considering the specialty; however, 39% of respondents mentioned receiving mentorship in the past. In fact, dermatology was one of the top 3 specialties for which respondents agreed that formal mentorship was important.¹⁹

Mentorship also has been shown to provide students with a variety of opportunities to develop personally and professionally. Some of these opportunities include increased confidence in their personal and professional success, increased desire to pursue a career in a field of interest, networking opportunities, career coaching, and support and research guidance.²⁰ A research study among medical students at Albert Einstein College of Medicine in New York, New York, found that US Medical Licensing Examination Step 1 scores, clinical grades, and the chance of not matching were important factors preventing them from applying to dermatology.²¹

Factors in Dermatology Residency Selection—A survey was conducted wherein 95 of 114 dermatology program directors expressed that among the top 5 criteria for dermatology resident selection were Step 1 scores and clinical grades, supporting the notion that academic factors were given a great emphasis during residency selection.²² Furthermore, among underrepresented minority medical students, a lack of diversity, the belief that minority students are seen negatively by residencies, socioeconomic factors, and not having mentors were major reasons for being dissuaded from applying to dermatology.²¹ These results showcase the heightened importance of mentors for underrepresented minority medical students in particular.

In graduate medical education, resources such as wikis, social networking sites, and blogs provide media through which trainees can communicate, exchange ideas, and enhance their medical knowledge.^23,24 A survey of 9606 osteopathic medical students showed that 35% of 992 respondents had used social media to learn more about residencies, and 10% believed that social media had influenced their choice of residency.²⁵ Given the impact social media has on recruitment, it also can be employed in a similar manner by dermatologists and dermatology residency programs to attract younger students to the field.

Access to More Opportunities to Learn About Dermatology—Besides shadowing and mentorship, other avenues of exposure to dermatology are possible and should be considered. In our study, 80% of students agreed that they would attend an event that increases exposure to dermatology if held by the premedical group, which suggests that students are eager to learn more about the field and want access to more opportunities, which could include learning procedures such as suturing or how to use a dermatoscope, attending guest speaker events, or participating in Learn2Derm volunteer events.

Learn2Derm was a skin cancer prevention fair first organized by medical students at George Washington University in Washington, DC. Students and residents sought to deliver sunscreens to underserved areas in Washington, DC, as well as teach residents about the importance of skin health. Participating in such events could be an excellent opportunity for all students to gain exposure to important topics in dermatology.²⁶

General Opinions of Dermatology—General opinions about dermatology and medicine were collected from the students through the optional “Additional Comments” section. Major themes found in the comments included the desire for more opportunities, mentorship, exposure, connections, and a discussion of disparities faced by Black patients/students within dermatology. Students also expressed an interest in dermatology and the desire to learn more about the specialty. From these themes, it can be gleaned that students are open to and eager for more opportunities to gain exposure and connections, and increasing the number of minority dermatologists is of importance.

Limitations—An important limitation of this study was the potential for selection bias, as the sample was chosen from a population at one university, which is not representative of the general population. Further, we only sampled students who were premedical and likely from a UiM racial group due to the demographics of the student population at the university, but given that the goal of the survey was to understand exposure to dermatology in underrepresented groups, we believe it was the appropriate population to target. Additionally, results were not compared with other more represented racial groups to see if these findings were unique to UiM undergraduate students.

Conclusion

Among premedical students, dermatology is an area of great interest with minimal opportunities available for exposure and learning because it is a smaller specialty with fewer experiences available for shadowing and mentorship. Although most UiM premedical students who were surveyed were exposed to the field through either the media or being a dermatology patient, fewer were exposed to the field through clinical experiences (such as shadowing) or mentorship. Most respondents found dermatology to be interesting and have considered pursuing it as a career. In particular, race-concordant mentoring in dermatologic care was valued by many students in garnering their interest in the field.

Most UiM students wanted more exposure to dermatology-related opportunities as well as mentorship and connections. Increasing shadowing, research, pipeline programs, and general events geared to dermatology are some modalities that could help improve exposure to dermatology for UiM students, especially for those interested in pursuing the field. This increased exposure can help positively influence more UiM students to pursue dermatology and help close the diversity gap in the field. Additionally, many were interested in attending potential dermatology informational events.

Given the fact that dermatology is a small field and mentorship may be hard to access, increasing informational events may be a more reasonable approach to inspiring and supporting interest. These events could include learning how to use certain tools and techniques, guest speaker events, or participating in educational volunteer efforts such as Learn2Derm.²⁶

Future research should focus on identifying beneficial factors of UiM premedical students who retain an interest in dermatology throughout their careers and actually apply to dermatology programs and become dermatologists. Those who do not apply to the specialty can be identified to understand potential dissuading factors and obstacles. Ultimately, more research and development of exposure opportunities, including mentorship programs and informational events, can be used to close the gap and improve diversity and health outcomes in dermatology.

Diversity of health care professionals improves medical outcomes and quality of life in patients. ¹ There is a lack of diversity in dermatology, with only 4.2% of dermatologists identifying as Hispanic and 3% identifying as African American, ² possibly due to a lack of early exposure to dermatology among high school and undergraduate students, a low number of underrepresented students in medical school, a lack of formal mentorship programs geared to underrepresented students, and implicit biases. ^1-4 Furthermore, the field is competitive, with many more applicants than available positions. In 2022, there were 851 applicants competing for 492 residency positions in dermatology. ⁵ Thus, it is important to educate young students about dermatology and understand root causes as to why the number of u nderrepresented in medicine (UiM) dermatologists remains stagnant.

According to Pritchett et al,⁴ it is crucial for dermatologists to interact with high school and college students to foster an early interest in dermatology. Many racial minority students do not progress from high school to college and then from college to medical school, which leaves a substantially reduced number of eligible UiM applicants who can progress into dermatology.⁶ Increasing the amount of UiM students going to medical school requires early mediation. Collaborating with pre-existing premedical school organizations through presentations and workshops is another way to promote an early interest in dermatology.⁴ Special consideration should be given to students who are UiM.

Among the general medical school curriculum, requirements for exposure to dermatology are not high. In one study, the median number of clinical and preclinical hours required was 10. Furthermore, 20% of 33 medical schools did not require preclinical dermatology hours (hours done before medical school rotations begin and in an academic setting), 36% required no clinical hours (rotational hours), 8% required no dermatology hours whatsoever, and only 10% required clinical dermatology rotation.³ Based on these findings, it is clear that dermatology is not well incorporated into medical school curricula. Furthermore, curricula have historically neglected to display adequate representation of skin of color.⁷ As a result, medical students generally have limited exposure to dermatology³ and are exposed even less to presentations of dermatologic issues in historically marginalized populations.⁷

Given the paucity of research on UiM students’ perceptions of dermatology prior to medical school, our cross-sectional survey study sought to evaluate the level of interest in dermatology of UiM premedical undergraduates. This survey specifically evaluated exposure to dermatology, preconceived notions about the field, and mentorship opportunities. By understanding these factors, dermatologists and dermatology residency programs can use this information to create mentorship opportunities and better adjust existing programs to meet students’ needs.

Methods

A 19-question multiple-choice survey was administered electronically (SurveyMonkey) in May 2020 to premedical students at Howard University (Washington, DC). One screening question was used: “What is your major?” Those who considered themselves a science major and/or with premedical interest were allowed to complete the survey. All students surveyed were members of the Health Professions Society at Howard University. Students who were interested in pursuing medical school were invited to respond. Approval for this study was obtained from the Howard University institutional review board (FWA00000891).

The survey was divided into 3 sections: Demographics, Exposure to Medicine and Dermatology, and Perceptions of Dermatology. The Demographics section addressed gender, age, and race/ethnicity. The Exposure to Medicine and Dermatology section addressed interest in attending medical school, shadowing experience, exposure to dermatology, and mentoring. The Perceptions of Dermatology section addressed preconceived notions about the field (eg, “dermatology is interesting and exciting”).

Statistical Analysis—The data represented are percentages based on the number of respondents who answered each question. Answers in response to “Please enter any comments” were organized into themes, and the number of respondents who discussed each theme was quantified into a table.

A total of 271 survey invitations were sent to premedical students at Howard University. Students were informed of the study protocol and asked to consent before proceeding to have their responses anonymously collected. Based on the screening question, 152 participants qualified for the survey, and 152 participants completed it (response rate, 56%; completion rate, 100%). Participants were asked to complete the survey only once.

Demographics—Eighty-four percent of respondents identified as science majors, and the remaining 16% identified as nonscience premedical. Ninety-four percent of participants identified as Black or African American; 3% as Asian or Asian American; and the remaining 3% as Other. Most respondents were female (82%), 16% were male, and 2% were either nonbinary or preferred not to answer. Ninety-nine percent were aged 18 to 24 years, and 1% were aged 25 to 34 years (Table 1).

Exposure to Medicine and Dermatology—Ninety-three percent of participants planned on attending medical school, and most students developed an interest in medicine from an early age. Ninety-six percent cited that they became interested in medicine prior to beginning their undergraduate education, and 4% developed an interest as freshmen or sophomores. When asked what led to their interest in medicine, family influence had the single greatest impact on students’ decision to pursue medicine (33%). Classes/school were the second most influential factor (24%), followed by volunteering (15%), shadowing (13%), other (7%), and peer influence (3%)(Figure 1).

Many (56%) premedical students surveyed had shadowing experience to varying degrees. Approximately 18% had fewer than 8 hours of shadowing experience, 24% had 8 to 40 hours, and 14% had more than 40 hours. However, many (43%) premedical students had no shadowing experience (Figure 2). Similarly, 30% of premedical students responded to having a physician as a mentor.

Regarding exposure to dermatology, 42% of premedical students had none. However, 58% of students had exposure to dermatology by being a patient themselves, 40% through seeing a dermatologist with a family member, 21% through seeing a dermatologist on television or social media, 5% through shadowing or volunteering, 3% through mentorship, and 1% through dermatology research (Figure 3).

Of students who said they were interested in dermatology (32%), 16% developed their interest before undergraduate education, while 9% developed interest in their freshman or sophomore year and 7% in their junior or senior year of undergraduate education. Three percent of respondents indicated that they had a dermatology mentorship.

Perceptions of Dermatology—To further evaluate the level of interest that UiM premedical students have in the field of dermatology, students were asked how much they agree or disagree on whether the field of dermatology is interesting. Sixty-three percent of the students agreed that the field of dermatology is interesting, 34% remained uncertain, and 3% disagreed. Additionally, students were asked whether they would consider dermatology as a career; 54% of respondents would consider dermatology as a career, 30% remained uncertain, and 16% would not consider dermatology as a career choice.

Nearly all (95%) students agreed that dermatologists do valuable work that goes beyond the scope of cosmetic procedures such as neuromodulators, fillers, chemical peels, and lasers. Some students also noted they had personal experiences interacting with a dermatologist. For example, one student described visiting the dermatologist many times to get a treatment regimen for their eczema.

Overall themes from the survey are depicted in Table 2. Major themes found in the comments included the desire for more dermatology-related opportunities, mentorship, exposure, connections, and a discussion of disparities faced by Black patients and students within dermatology. Students also expressed an interest in dermatology and the desire to learn more about the specialty.

Comment

Interest in Dermatology—In this cross-sectional survey study of 152 UiM undergraduate students, it was found that many students were interested in dermatology as a career, and more than 70% would be interested in attending events that increased exposure to the field of dermatology. Of the students who had any exposure to dermatology, less than 5% had shadowed an actual dermatologist. The survey showed that there is great potential interest in exposing UiM undergraduate students to the field of dermatology. We found that UiM students are interested in learning more about dermatology, with 80% indicating that they would be willing to participate in dermatology-focused events if they were available. Overall, students mentioned a lack of opportunities, mentorship, exposure, and connections in dermatology despite their interest in the field.

Racial Disparities in Dermatology—Additionally, students discussed disparities they encountered with dermatology due to a lack of patient-provider race concordance and the perceived difference in care when encountering a race-concordant dermatologist. One student noted that they went to multiple White dermatologists for their eczema, and “it wasn’t until I was evaluated by a Black dermatologist (diagnosed with eczema as well) [that I was] prescribed . . . the perfect medication.” Another student noted how a Black dermatologist sparked their interest in getting to know more about the field and remarked that they “think it is an important field that lacks representation for Black people.” This research stresses the need for more dermatology mentorship among UiM undergraduates.

Family Influence on Career Selection—The majority of UiM students in our study became interested in medicine because of family, which is consistent with other studies. In a cross-sectional survey of 300 Pakistani students (150 medical and 150 nonmedical), 87% of students stated that their family had an influence on their career selection.⁸ In another study of 15 junior doctors in Sierra Leone, the most common reasons for pursuing medicine were the desire to help and familial and peer influence.⁹ This again showcases how family can have a positive impact on career selection for medical professionals and highlights the need for early intervention.

Shadowing—One way in which student exposure to dermatology can be effectively increased is by shadowing. In a study evaluating a 30-week shadowing program at the Pediatric Continuity Clinic in Los Angeles, California, a greater proportion of premedical students believed they had a good understanding of the job of a resident physician after the program’s completion compared to before starting the program (an increase from 78% to 100%).¹⁰ The proportion of students reporting a good understanding of the patient-physician relationship after completing the program also increased from 33% to 78%. Furthermore, 72% of the residents stated that having the undergraduates in the clinic was a positive experience.¹⁰ Thus, increasing shadowing opportunities is one extremely effective way to increase student knowledge and awareness of and exposure to dermatology.

Dermatology Mentors—Although 32% of students were interested in dermatology, 3% of students had mentorship in dermatology. In prior studies, it has been shown that mentorship is of great importance in student success and interest in pursuing a specialty. A report from the Association of American Medical Colleges 2019 Medical School Graduation Questionnaire found that the third most influential factor (52.1%) in specialty selection was role model influence.¹¹ In fact, having a role model is consistently one of the top 3 influences on student specialty choice and interest in the last 5 years of survey research. Some studies also have shown mentorship as a positive influence in specialty interest at the undergraduate and graduate levels. A study on an undergraduate student interest group noted that surgeon mentorship and exposure were positive factors to students’ interests in surgery.¹² In fact, the Association of American Medical Colleges noted that some surgical specialties, such as orthopedic surgery, had 45% of respondents who were interested in the specialty before medical school pursue their initial preference in medical school.¹³ Another survey corroborated these findings; more orthopedic-bound students compared with other specialties indicated they were more likely to pursue their field because of experiences prior to medical school.¹⁴

One of the reasons students might not have been exposed to as many opportunities for mentorship in dermatology is because the specialty is one of the smaller fields in medicine and tends to be concentrated in more well-resourced metropolitan areas.¹⁵ Dermatologists make up only 1.3% of the physician workforce.¹⁶ Because there might not be as much exposure to the field, students might also explore their interests in dermatology through other fields, such as through shadowing and observing primary care physicians who often treat patients with dermatologic issues. Skin diseases are a common reason for primary care visits, and one study suggested dermatologic diseases can make up approximately 8.4% of visits in primary care.¹⁷

Moreover, only 1% of medical schools require an elective in dermatology.¹⁸ With exposure being a crucial component to pursuing the specialty, it also is important to pursue formal mentorship within the specialty itself. One study noted that formal mentorship in dermatology was important for most (67%) respondents when considering the specialty; however, 39% of respondents mentioned receiving mentorship in the past. In fact, dermatology was one of the top 3 specialties for which respondents agreed that formal mentorship was important.¹⁹

Mentorship also has been shown to provide students with a variety of opportunities to develop personally and professionally. Some of these opportunities include increased confidence in their personal and professional success, increased desire to pursue a career in a field of interest, networking opportunities, career coaching, and support and research guidance.²⁰ A research study among medical students at Albert Einstein College of Medicine in New York, New York, found that US Medical Licensing Examination Step 1 scores, clinical grades, and the chance of not matching were important factors preventing them from applying to dermatology.²¹

Factors in Dermatology Residency Selection—A survey was conducted wherein 95 of 114 dermatology program directors expressed that among the top 5 criteria for dermatology resident selection were Step 1 scores and clinical grades, supporting the notion that academic factors were given a great emphasis during residency selection.²² Furthermore, among underrepresented minority medical students, a lack of diversity, the belief that minority students are seen negatively by residencies, socioeconomic factors, and not having mentors were major reasons for being dissuaded from applying to dermatology.²¹ These results showcase the heightened importance of mentors for underrepresented minority medical students in particular.

In graduate medical education, resources such as wikis, social networking sites, and blogs provide media through which trainees can communicate, exchange ideas, and enhance their medical knowledge.^23,24 A survey of 9606 osteopathic medical students showed that 35% of 992 respondents had used social media to learn more about residencies, and 10% believed that social media had influenced their choice of residency.²⁵ Given the impact social media has on recruitment, it also can be employed in a similar manner by dermatologists and dermatology residency programs to attract younger students to the field.

Access to More Opportunities to Learn About Dermatology—Besides shadowing and mentorship, other avenues of exposure to dermatology are possible and should be considered. In our study, 80% of students agreed that they would attend an event that increases exposure to dermatology if held by the premedical group, which suggests that students are eager to learn more about the field and want access to more opportunities, which could include learning procedures such as suturing or how to use a dermatoscope, attending guest speaker events, or participating in Learn2Derm volunteer events.

Learn2Derm was a skin cancer prevention fair first organized by medical students at George Washington University in Washington, DC. Students and residents sought to deliver sunscreens to underserved areas in Washington, DC, as well as teach residents about the importance of skin health. Participating in such events could be an excellent opportunity for all students to gain exposure to important topics in dermatology.²⁶

General Opinions of Dermatology—General opinions about dermatology and medicine were collected from the students through the optional “Additional Comments” section. Major themes found in the comments included the desire for more opportunities, mentorship, exposure, connections, and a discussion of disparities faced by Black patients/students within dermatology. Students also expressed an interest in dermatology and the desire to learn more about the specialty. From these themes, it can be gleaned that students are open to and eager for more opportunities to gain exposure and connections, and increasing the number of minority dermatologists is of importance.

Limitations—An important limitation of this study was the potential for selection bias, as the sample was chosen from a population at one university, which is not representative of the general population. Further, we only sampled students who were premedical and likely from a UiM racial group due to the demographics of the student population at the university, but given that the goal of the survey was to understand exposure to dermatology in underrepresented groups, we believe it was the appropriate population to target. Additionally, results were not compared with other more represented racial groups to see if these findings were unique to UiM undergraduate students.

Conclusion

Among premedical students, dermatology is an area of great interest with minimal opportunities available for exposure and learning because it is a smaller specialty with fewer experiences available for shadowing and mentorship. Although most UiM premedical students who were surveyed were exposed to the field through either the media or being a dermatology patient, fewer were exposed to the field through clinical experiences (such as shadowing) or mentorship. Most respondents found dermatology to be interesting and have considered pursuing it as a career. In particular, race-concordant mentoring in dermatologic care was valued by many students in garnering their interest in the field.

Most UiM students wanted more exposure to dermatology-related opportunities as well as mentorship and connections. Increasing shadowing, research, pipeline programs, and general events geared to dermatology are some modalities that could help improve exposure to dermatology for UiM students, especially for those interested in pursuing the field. This increased exposure can help positively influence more UiM students to pursue dermatology and help close the diversity gap in the field. Additionally, many were interested in attending potential dermatology informational events.

Given the fact that dermatology is a small field and mentorship may be hard to access, increasing informational events may be a more reasonable approach to inspiring and supporting interest. These events could include learning how to use certain tools and techniques, guest speaker events, or participating in educational volunteer efforts such as Learn2Derm.²⁶

Future research should focus on identifying beneficial factors of UiM premedical students who retain an interest in dermatology throughout their careers and actually apply to dermatology programs and become dermatologists. Those who do not apply to the specialty can be identified to understand potential dissuading factors and obstacles. Ultimately, more research and development of exposure opportunities, including mentorship programs and informational events, can be used to close the gap and improve diversity and health outcomes in dermatology.

Diversity of health care professionals improves medical outcomes and quality of life in patients. ¹ There is a lack of diversity in dermatology, with only 4.2% of dermatologists identifying as Hispanic and 3% identifying as African American, ² possibly due to a lack of early exposure to dermatology among high school and undergraduate students, a low number of underrepresented students in medical school, a lack of formal mentorship programs geared to underrepresented students, and implicit biases. ^1-4 Furthermore, the field is competitive, with many more applicants than available positions. In 2022, there were 851 applicants competing for 492 residency positions in dermatology. ⁵ Thus, it is important to educate young students about dermatology and understand root causes as to why the number of u nderrepresented in medicine (UiM) dermatologists remains stagnant.

According to Pritchett et al,⁴ it is crucial for dermatologists to interact with high school and college students to foster an early interest in dermatology. Many racial minority students do not progress from high school to college and then from college to medical school, which leaves a substantially reduced number of eligible UiM applicants who can progress into dermatology.⁶ Increasing the amount of UiM students going to medical school requires early mediation. Collaborating with pre-existing premedical school organizations through presentations and workshops is another way to promote an early interest in dermatology.⁴ Special consideration should be given to students who are UiM.

Among the general medical school curriculum, requirements for exposure to dermatology are not high. In one study, the median number of clinical and preclinical hours required was 10. Furthermore, 20% of 33 medical schools did not require preclinical dermatology hours (hours done before medical school rotations begin and in an academic setting), 36% required no clinical hours (rotational hours), 8% required no dermatology hours whatsoever, and only 10% required clinical dermatology rotation.³ Based on these findings, it is clear that dermatology is not well incorporated into medical school curricula. Furthermore, curricula have historically neglected to display adequate representation of skin of color.⁷ As a result, medical students generally have limited exposure to dermatology³ and are exposed even less to presentations of dermatologic issues in historically marginalized populations.⁷

Given the paucity of research on UiM students’ perceptions of dermatology prior to medical school, our cross-sectional survey study sought to evaluate the level of interest in dermatology of UiM premedical undergraduates. This survey specifically evaluated exposure to dermatology, preconceived notions about the field, and mentorship opportunities. By understanding these factors, dermatologists and dermatology residency programs can use this information to create mentorship opportunities and better adjust existing programs to meet students’ needs.

Methods

A 19-question multiple-choice survey was administered electronically (SurveyMonkey) in May 2020 to premedical students at Howard University (Washington, DC). One screening question was used: “What is your major?” Those who considered themselves a science major and/or with premedical interest were allowed to complete the survey. All students surveyed were members of the Health Professions Society at Howard University. Students who were interested in pursuing medical school were invited to respond. Approval for this study was obtained from the Howard University institutional review board (FWA00000891).

The survey was divided into 3 sections: Demographics, Exposure to Medicine and Dermatology, and Perceptions of Dermatology. The Demographics section addressed gender, age, and race/ethnicity. The Exposure to Medicine and Dermatology section addressed interest in attending medical school, shadowing experience, exposure to dermatology, and mentoring. The Perceptions of Dermatology section addressed preconceived notions about the field (eg, “dermatology is interesting and exciting”).

Statistical Analysis—The data represented are percentages based on the number of respondents who answered each question. Answers in response to “Please enter any comments” were organized into themes, and the number of respondents who discussed each theme was quantified into a table.

A total of 271 survey invitations were sent to premedical students at Howard University. Students were informed of the study protocol and asked to consent before proceeding to have their responses anonymously collected. Based on the screening question, 152 participants qualified for the survey, and 152 participants completed it (response rate, 56%; completion rate, 100%). Participants were asked to complete the survey only once.

Demographics—Eighty-four percent of respondents identified as science majors, and the remaining 16% identified as nonscience premedical. Ninety-four percent of participants identified as Black or African American; 3% as Asian or Asian American; and the remaining 3% as Other. Most respondents were female (82%), 16% were male, and 2% were either nonbinary or preferred not to answer. Ninety-nine percent were aged 18 to 24 years, and 1% were aged 25 to 34 years (Table 1).

Exposure to Medicine and Dermatology—Ninety-three percent of participants planned on attending medical school, and most students developed an interest in medicine from an early age. Ninety-six percent cited that they became interested in medicine prior to beginning their undergraduate education, and 4% developed an interest as freshmen or sophomores. When asked what led to their interest in medicine, family influence had the single greatest impact on students’ decision to pursue medicine (33%). Classes/school were the second most influential factor (24%), followed by volunteering (15%), shadowing (13%), other (7%), and peer influence (3%)(Figure 1).

Many (56%) premedical students surveyed had shadowing experience to varying degrees. Approximately 18% had fewer than 8 hours of shadowing experience, 24% had 8 to 40 hours, and 14% had more than 40 hours. However, many (43%) premedical students had no shadowing experience (Figure 2). Similarly, 30% of premedical students responded to having a physician as a mentor.

Regarding exposure to dermatology, 42% of premedical students had none. However, 58% of students had exposure to dermatology by being a patient themselves, 40% through seeing a dermatologist with a family member, 21% through seeing a dermatologist on television or social media, 5% through shadowing or volunteering, 3% through mentorship, and 1% through dermatology research (Figure 3).

Of students who said they were interested in dermatology (32%), 16% developed their interest before undergraduate education, while 9% developed interest in their freshman or sophomore year and 7% in their junior or senior year of undergraduate education. Three percent of respondents indicated that they had a dermatology mentorship.

Perceptions of Dermatology—To further evaluate the level of interest that UiM premedical students have in the field of dermatology, students were asked how much they agree or disagree on whether the field of dermatology is interesting. Sixty-three percent of the students agreed that the field of dermatology is interesting, 34% remained uncertain, and 3% disagreed. Additionally, students were asked whether they would consider dermatology as a career; 54% of respondents would consider dermatology as a career, 30% remained uncertain, and 16% would not consider dermatology as a career choice.

Nearly all (95%) students agreed that dermatologists do valuable work that goes beyond the scope of cosmetic procedures such as neuromodulators, fillers, chemical peels, and lasers. Some students also noted they had personal experiences interacting with a dermatologist. For example, one student described visiting the dermatologist many times to get a treatment regimen for their eczema.

Overall themes from the survey are depicted in Table 2. Major themes found in the comments included the desire for more dermatology-related opportunities, mentorship, exposure, connections, and a discussion of disparities faced by Black patients and students within dermatology. Students also expressed an interest in dermatology and the desire to learn more about the specialty.

Comment

Interest in Dermatology—In this cross-sectional survey study of 152 UiM undergraduate students, it was found that many students were interested in dermatology as a career, and more than 70% would be interested in attending events that increased exposure to the field of dermatology. Of the students who had any exposure to dermatology, less than 5% had shadowed an actual dermatologist. The survey showed that there is great potential interest in exposing UiM undergraduate students to the field of dermatology. We found that UiM students are interested in learning more about dermatology, with 80% indicating that they would be willing to participate in dermatology-focused events if they were available. Overall, students mentioned a lack of opportunities, mentorship, exposure, and connections in dermatology despite their interest in the field.

Racial Disparities in Dermatology—Additionally, students discussed disparities they encountered with dermatology due to a lack of patient-provider race concordance and the perceived difference in care when encountering a race-concordant dermatologist. One student noted that they went to multiple White dermatologists for their eczema, and “it wasn’t until I was evaluated by a Black dermatologist (diagnosed with eczema as well) [that I was] prescribed . . . the perfect medication.” Another student noted how a Black dermatologist sparked their interest in getting to know more about the field and remarked that they “think it is an important field that lacks representation for Black people.” This research stresses the need for more dermatology mentorship among UiM undergraduates.

Family Influence on Career Selection—The majority of UiM students in our study became interested in medicine because of family, which is consistent with other studies. In a cross-sectional survey of 300 Pakistani students (150 medical and 150 nonmedical), 87% of students stated that their family had an influence on their career selection.⁸ In another study of 15 junior doctors in Sierra Leone, the most common reasons for pursuing medicine were the desire to help and familial and peer influence.⁹ This again showcases how family can have a positive impact on career selection for medical professionals and highlights the need for early intervention.

Shadowing—One way in which student exposure to dermatology can be effectively increased is by shadowing. In a study evaluating a 30-week shadowing program at the Pediatric Continuity Clinic in Los Angeles, California, a greater proportion of premedical students believed they had a good understanding of the job of a resident physician after the program’s completion compared to before starting the program (an increase from 78% to 100%).¹⁰ The proportion of students reporting a good understanding of the patient-physician relationship after completing the program also increased from 33% to 78%. Furthermore, 72% of the residents stated that having the undergraduates in the clinic was a positive experience.¹⁰ Thus, increasing shadowing opportunities is one extremely effective way to increase student knowledge and awareness of and exposure to dermatology.

Dermatology Mentors—Although 32% of students were interested in dermatology, 3% of students had mentorship in dermatology. In prior studies, it has been shown that mentorship is of great importance in student success and interest in pursuing a specialty. A report from the Association of American Medical Colleges 2019 Medical School Graduation Questionnaire found that the third most influential factor (52.1%) in specialty selection was role model influence.¹¹ In fact, having a role model is consistently one of the top 3 influences on student specialty choice and interest in the last 5 years of survey research. Some studies also have shown mentorship as a positive influence in specialty interest at the undergraduate and graduate levels. A study on an undergraduate student interest group noted that surgeon mentorship and exposure were positive factors to students’ interests in surgery.¹² In fact, the Association of American Medical Colleges noted that some surgical specialties, such as orthopedic surgery, had 45% of respondents who were interested in the specialty before medical school pursue their initial preference in medical school.¹³ Another survey corroborated these findings; more orthopedic-bound students compared with other specialties indicated they were more likely to pursue their field because of experiences prior to medical school.¹⁴

One of the reasons students might not have been exposed to as many opportunities for mentorship in dermatology is because the specialty is one of the smaller fields in medicine and tends to be concentrated in more well-resourced metropolitan areas.¹⁵ Dermatologists make up only 1.3% of the physician workforce.¹⁶ Because there might not be as much exposure to the field, students might also explore their interests in dermatology through other fields, such as through shadowing and observing primary care physicians who often treat patients with dermatologic issues. Skin diseases are a common reason for primary care visits, and one study suggested dermatologic diseases can make up approximately 8.4% of visits in primary care.¹⁷

Moreover, only 1% of medical schools require an elective in dermatology.¹⁸ With exposure being a crucial component to pursuing the specialty, it also is important to pursue formal mentorship within the specialty itself. One study noted that formal mentorship in dermatology was important for most (67%) respondents when considering the specialty; however, 39% of respondents mentioned receiving mentorship in the past. In fact, dermatology was one of the top 3 specialties for which respondents agreed that formal mentorship was important.¹⁹

Mentorship also has been shown to provide students with a variety of opportunities to develop personally and professionally. Some of these opportunities include increased confidence in their personal and professional success, increased desire to pursue a career in a field of interest, networking opportunities, career coaching, and support and research guidance.²⁰ A research study among medical students at Albert Einstein College of Medicine in New York, New York, found that US Medical Licensing Examination Step 1 scores, clinical grades, and the chance of not matching were important factors preventing them from applying to dermatology.²¹

Factors in Dermatology Residency Selection—A survey was conducted wherein 95 of 114 dermatology program directors expressed that among the top 5 criteria for dermatology resident selection were Step 1 scores and clinical grades, supporting the notion that academic factors were given a great emphasis during residency selection.²² Furthermore, among underrepresented minority medical students, a lack of diversity, the belief that minority students are seen negatively by residencies, socioeconomic factors, and not having mentors were major reasons for being dissuaded from applying to dermatology.²¹ These results showcase the heightened importance of mentors for underrepresented minority medical students in particular.

In graduate medical education, resources such as wikis, social networking sites, and blogs provide media through which trainees can communicate, exchange ideas, and enhance their medical knowledge.^23,24 A survey of 9606 osteopathic medical students showed that 35% of 992 respondents had used social media to learn more about residencies, and 10% believed that social media had influenced their choice of residency.²⁵ Given the impact social media has on recruitment, it also can be employed in a similar manner by dermatologists and dermatology residency programs to attract younger students to the field.

Access to More Opportunities to Learn About Dermatology—Besides shadowing and mentorship, other avenues of exposure to dermatology are possible and should be considered. In our study, 80% of students agreed that they would attend an event that increases exposure to dermatology if held by the premedical group, which suggests that students are eager to learn more about the field and want access to more opportunities, which could include learning procedures such as suturing or how to use a dermatoscope, attending guest speaker events, or participating in Learn2Derm volunteer events.

Learn2Derm was a skin cancer prevention fair first organized by medical students at George Washington University in Washington, DC. Students and residents sought to deliver sunscreens to underserved areas in Washington, DC, as well as teach residents about the importance of skin health. Participating in such events could be an excellent opportunity for all students to gain exposure to important topics in dermatology.²⁶

General Opinions of Dermatology—General opinions about dermatology and medicine were collected from the students through the optional “Additional Comments” section. Major themes found in the comments included the desire for more opportunities, mentorship, exposure, connections, and a discussion of disparities faced by Black patients/students within dermatology. Students also expressed an interest in dermatology and the desire to learn more about the specialty. From these themes, it can be gleaned that students are open to and eager for more opportunities to gain exposure and connections, and increasing the number of minority dermatologists is of importance.

Limitations—An important limitation of this study was the potential for selection bias, as the sample was chosen from a population at one university, which is not representative of the general population. Further, we only sampled students who were premedical and likely from a UiM racial group due to the demographics of the student population at the university, but given that the goal of the survey was to understand exposure to dermatology in underrepresented groups, we believe it was the appropriate population to target. Additionally, results were not compared with other more represented racial groups to see if these findings were unique to UiM undergraduate students.

Conclusion

Among premedical students, dermatology is an area of great interest with minimal opportunities available for exposure and learning because it is a smaller specialty with fewer experiences available for shadowing and mentorship. Although most UiM premedical students who were surveyed were exposed to the field through either the media or being a dermatology patient, fewer were exposed to the field through clinical experiences (such as shadowing) or mentorship. Most respondents found dermatology to be interesting and have considered pursuing it as a career. In particular, race-concordant mentoring in dermatologic care was valued by many students in garnering their interest in the field.

Most UiM students wanted more exposure to dermatology-related opportunities as well as mentorship and connections. Increasing shadowing, research, pipeline programs, and general events geared to dermatology are some modalities that could help improve exposure to dermatology for UiM students, especially for those interested in pursuing the field. This increased exposure can help positively influence more UiM students to pursue dermatology and help close the diversity gap in the field. Additionally, many were interested in attending potential dermatology informational events.

Given the fact that dermatology is a small field and mentorship may be hard to access, increasing informational events may be a more reasonable approach to inspiring and supporting interest. These events could include learning how to use certain tools and techniques, guest speaker events, or participating in educational volunteer efforts such as Learn2Derm.²⁶

Future research should focus on identifying beneficial factors of UiM premedical students who retain an interest in dermatology throughout their careers and actually apply to dermatology programs and become dermatologists. Those who do not apply to the specialty can be identified to understand potential dissuading factors and obstacles. Ultimately, more research and development of exposure opportunities, including mentorship programs and informational events, can be used to close the gap and improve diversity and health outcomes in dermatology.

From the University of Nebraska, Lincoln (Mr. Puckett and Dr. Ryherd), University of Nebraska Medical Center, Omaha (Dr. Manley), and the University of Nebraska, Omaha (Dr. Ryan).

ABSTRACT

Objective: This study evaluated relationships between physical characteristics of nursing home residences and quality-of-care measures.

Design: This was a cross-sectional ecologic study. The dependent variables were 5 Centers for Medicare & Medicaid Services (CMS) Nursing Home Compare database long-stay quality measures (QMs) during 2019: percentage of residents who displayed depressive symptoms, percentage of residents who were physically restrained, percentage of residents who experienced 1 or more falls resulting in injury, percentage of residents who received antipsychotic medication, and percentage of residents who received anti-anxiety medication. The independent variables were 4 residence characteristics: ownership type, size, occupancy, and region within the United States. We explored how different types of each residence characteristic compare for each QM.

Setting, participants, and measurements: Quality measure values from 15,420 CMS-supported nursing homes across the United States averaged over the 4 quarters of 2019 reporting were used. Welch’s analysis of variance was performed to examine whether the mean QM values for groups within each residential characteristic were statistically different.

Results: Publicly owned and low-occupancy residences had the highest mean QM values, indicating the poorest performance. Nonprofit and high-occupancy residences generally had the lowest (ie, best) mean QM values. There were significant differences in mean QM values among nursing home sizes and regions.

Conclusion: This study suggests that residence characteristics are related to 5 nursing home QMs. Results suggest that physical characteristics may be related to overall quality of life in nursing homes.

Keywords: quality of care, quality measures, residence characteristics, Alzheimer’s disease and related dementias.

More than 55 million people worldwide are living with Alzheimer’s disease and related dementias (ADRD).¹ With the aging of the Baby Boomer population, this number is expected to rise to more than 78 million worldwide by 2030.¹ Given the growing number of cognitively impaired older adults, there is an increased need for residences designed for the specialized care of this population. Although there are dozens of living options for the elderly, and although most specialized establishments have the resources to meet the immediate needs of their residents, many facilities lack universal design features that support a high quality of life for someone with ADRD or mild cognitive impairment. Previous research has shown relationships between behavioral and psychological symptoms of dementia (BPSD) and environmental characteristics such as acoustics, lighting, and indoor air temperature.^2,3 Physical behaviors of BPSD, including aggression and wandering, and psychological symptoms, such as depression, anxiety, and delusions, put residents at risk of injury.⁴ Additionally, BPSD is correlated with caregiver burden and stress.^5-8 Patients with dementia may also experience a lower stress threshold, changes in perception of space, and decreased short-term memory, creating environmental difficulties for those with ADRD⁹ that lead them to exhibit BPSD due to poor environmental design. Thus, there is a need to learn more about design features that minimize BPSD and promote a high quality of life for those with ADRD.¹⁰

Although research has shown relationships between physical environmental characteristics and BPSD, in this work we study relationships between possible BPSD indicators and 4 residence-level characteristics: ownership type, size, occupancy, and region in the United States (determined by location of the Centers for Medicare & Medicaid Services [CMS] regional offices). We analyzed data from the CMS Nursing Home Compare database for the year 2019.¹¹ This database publishes quarterly data and star ratings for quality-of-care measures (QMs), staffing levels, and health inspections for every nursing home supported by CMS. Previous research has investigated the accuracy of QM reporting for resident falls, the impact of residential characteristics on administration of antipsychotic medication, the influence of profit status on resident outcomes and quality of care, and the effect of nursing home size on quality of life.^12-16 Additionally, research suggests that residential characteristics such as size and location could be associated with infection control in nursing homes.¹⁷

Certain QMs, such as psychotropic drug administration, resident falls, and physical restraint, provide indicators of agitation, disorientation, or aggression, which are often signals of BPSD episodes. We hypothesized that residence types are associated with different QM scores, which could indicate different occurrences of BPSD. We selected 5 QMs for long-stay residents that could potentially be used as indicators of BPSD. Short-stay resident data were not included in this work to control for BPSD that could be a result of sheer unfamiliarity with the environment and confusion from being in a new home.

Methods

Design and Data Collection

This was a cross-sectional ecologic study aimed at exploring relationships between aggregate residential characteristics and QMs. Data were retrieved from the 2019 annual archives found in the CMS provider data catalog on nursing homes, including rehabilitation services.¹¹ The dataset provides general residence information, such as ownership, number of beds, number of residents, and location, as well as residence quality metrics, such as QMs, staffing data, and inspection data. Residence characteristics and 4-quarter averages of QMs were retrieved and used as cross-sectional data. The data used are from 15,420 residences across the United States. Nursing homes located in Guam, the US Pacific Territories, Puerto Rico, and the US Virgin Islands, while supported by CMS and included in the dataset, were excluded from the study due to a severe absence of QM data.

Dependent Variables

We investigated 5 QMs that were averaged across the 4 quarters of 2019. The QMs used as dependent variables were percentage of residents who displayed depressive symptoms (depression), percentage of residents who were physically restrained (restraint), percentage of residents who experienced 1 or more falls resulting in a major injury (falls), percentage of residents who received antipsychotic medication (antipsychotic medication), and percentage of residents who received anti-anxiety or hypnotic medication (anti-anxiety medication).

A total of 2471 QM values were unreported across the 5 QM analyzed: 501 residences did not report depression data; 479 did not report restraint data; 477 did not report falls data; 508 did not report antipsychotic medication data; and 506 did not report anti-anxiety medication data. A residence with a missing QM value was excluded from that respective analysis.

To assess the relationships among the different QMs, a Pearson correlation coefficient r was computed for each unique pair of QMs (Figure). All QMs studied were found to be very weakly or weakly correlated with one another using the Evans classification for very weak and weak correlations (r < 0.20 and 0.20 < r < 0.39, respectively).¹⁸

Independent Variables

A total of 15,420 residences were included in the study. Seventy-nine residences did not report occupancy data, however, so those residences were excluded from the occupancy analyses. We categorized the ownership of each nursing home as for-profit, nonprofit, or public. We categorized nursing home size, based on quartiles of the size distribution, as large (> 127 beds), medium (64 to 126 beds), and small (< 64 beds). This method for categorizing the residential characteristics was similar to that used in previous work.¹⁹ Similarly, we categorized nursing home occupancy as high (> 92% occupancy), medium (73% to 91% occupancy), and low (< 73% occupancy) based on quartiles of the occupancy distribution. For the regional analysis, we grouped states together based on the CMS regional offices: Atlanta, Georgia; Boston, Massachusetts; Chicago, Illinois; Dallas, Texas; Denver, Colorado; Kansas City, Missouri; New York, New York; Philadelphia, Pennsylvania; San Francisco, California; and Seattle, Washington.²⁰

Analyses

We used Levene’s test to determine whether variances among the residential groups were equal for each QM, using an a priori α = 0.05. For all 20 tests conducted (4 residential characteristics for all 5 QMs), the resulting F-statistics were significant, indicating that the assumption of homogeneity of variance was not met.

We therefore used Welch’s analysis of variance (ANOVA) to evaluate whether the groups within each residential characteristic were the same on their QM means. For example, we tested whether for-profit, nonprofit, and public residences had significantly different mean depression rates. For statistically significant differences, a Games-Howell post-hoc test was conducted to test the difference between all unique pairwise comparisons. An a priori α = 0.05 was used for both Welch’s ANOVA and post-hoc testing. All analyses were conducted in RStudio Version 1.2.5033 (Posit Software, PBC).

Mean Differences

Mean QM scores for the 5 QMs investigated, grouped by residential characteristic for the 2019 year of reporting, are shown in Table 1. It should be noted that the number of residences that reported occupancy data (n = 15,341) does not equal the total number of residences included in the study (N = 15,420) because 79 residences did not report occupancy data. For all QMs reported in Table 1, lower scores are better. Table 2 and Table 3 show results from pairwise comparisons of mean differences for the different residential characteristic and QM groupings. Mean differences and 95% CI are presented along with an indication of statistical significance (when applicable).

Ownership

Nonprofit residences had significantly lower (ie, better) mean scores than for-profit and public residences for 3 QMs: resident depression, antipsychotic medication use, and anti-anxiety medication use. For-profit and public residences did not significantly differ in their mean values for these QMs. For-profit residences had a significantly lower mean score for resident falls than both nonprofit and public residences, but no significant difference existed between scores for nonprofit and public residence falls. There were no statistically significant differences between mean restraint scores among the ownership types.

Size

Large (ie, high-capacity) residences had a significantly higher mean depression score than both medium and small residences, but there was not a significant difference between medium and small residences. Large residences had the significantly lowest mean score for resident falls, and medium residences scored significantly lower than small residences. Medium residences had a significantly higher mean score for anti-anxiety medication use than both small and large residences, but there was no significant difference between small and large residences. There were no statistically significant differences between mean scores for restraint and antipsychotic medication use among the nursing home sizes.

Occupancy

The mean scores for 4 out of the 5 QMs exhibited similar relationships with occupancy rates: resident depression, falls, and antipsychotic and anti-anxiety medication use. Low-occupancy residences consistently scored significantly higher than both medium- and high-occupancy residences, and medium-occupancy residences consistently scored significantly higher than high-occupancy residences. On average, high-occupancy (≥ 92%) residences reported better QM scores than low-occupancy (< 73%) and medium-occupancy (73% to 91%) residences for all the QMs studied except physical restraint, which yielded no significant results. These findings indicate a possible inverse relationship between building occupancy rate and these 4 QMs.

Region

Pairwise comparisons of mean QM scores by region are shown in Table 3. The Chicago region had a significantly higher mean depression score than all other regions, while the San Francisco region’s score was significantly lower than all other regions, except Atlanta and Boston. The Kansas City region had a significantly higher mean score for resident falls than all other regions, with the exception of Denver, and the San Francisco region scored significantly lower than all other regions in falls. The Boston region had a significantly higher mean score for administering antipsychotic medication than all other regions, except for Kansas City and Seattle, and the New York and San Francisco regions both had significantly lower scores than all other regions except for each other. The Atlanta region reported a significantly higher mean score for administering antianxiety medication than all other regions, and the Seattle region’s score for anti-anxiety medication use was significantly lower than all other regions except for San Francisco.

This study presented mean percentages for 5 QMs reported in the Nursing Home Compare database for the year 2019: depression, restraint, falls, antipsychotic medication use, and anti-anxiety medication use. We investigated these scores by 4 residential characteristics: ownership type, size, occupancy, and region. In general, publicly owned and low-occupancy residences had the highest scores, and thus the poorest performances, for the 5 chosen QMs during 2019. Nonprofit and high-occupancy residences generally had the lowest (ie, better) scores, and this result agrees with previous findings on long-stay nursing home residents.²¹ One possible explanation for better performance by high-occupancy buildings could be that increased social interaction is beneficial to nursing home residents as compared with low-occupancy buildings, where less social interaction is probable. It is difficult to draw conclusions regarding nursing home size and region; however, there are significant differences among sizes for 3 out of the 5 QMs and significant differences among regions for all 5 QMs. The analyses suggest that residence-level characteristics are related to QM scores. Although reported QMs are not a direct representation of resident quality of life, this work agrees with previous research that residential characteristics have some impact on the lives of nursing home residents.^13-17 Improvements in QM reporting and changes in quality improvement goals since the formation of Nursing Home Compare exist, suggesting that nursing homes’ awareness of their reporting duties may impact quality of care or reporting tendencies.^21,22 Future research should consider investigating the impacts of the COVID-19 pandemic on quality-reporting trends and QM scores.

Other physical characteristics of nursing homes, such as noise, lighting levels, and air quality, may also have an impact on QMs and possibly nursing home residents themselves. This type of data exploration could be included in future research. Additionally, future research could include a similar analysis over a longer period, rather than the 1-year period examined here, to investigate which types of residences consistently have high or low scores or how different types of residences have evolved over the years, particularly considering the impact of the COVID-19 pandemic. Information such as staffing levels, building renovations, and inspection data could be accounted for in future studies. Different QMs could also be investigated to better understand the influence of residential characteristics on quality of care.

Conclusion

This study suggests that residence-level characteristics are related to 5 reported nursing home QMs. Overall, nonprofit and high-occupancy residences had the lowest QM scores, indicating the highest performance. Although the results do not necessarily suggest that residence-level characteristics impact individual nursing home residents’ quality of life, they suggest that physical characteristics affect overall quality of life in nursing homes. Future research is needed to determine the specific physical characteristics of these residences that affect QM scores.

Corresponding author: Brian J. Puckett, [email protected].

Disclosures: None reported.

From the University of Nebraska, Lincoln (Mr. Puckett and Dr. Ryherd), University of Nebraska Medical Center, Omaha (Dr. Manley), and the University of Nebraska, Omaha (Dr. Ryan).

ABSTRACT

Objective: This study evaluated relationships between physical characteristics of nursing home residences and quality-of-care measures.

Design: This was a cross-sectional ecologic study. The dependent variables were 5 Centers for Medicare & Medicaid Services (CMS) Nursing Home Compare database long-stay quality measures (QMs) during 2019: percentage of residents who displayed depressive symptoms, percentage of residents who were physically restrained, percentage of residents who experienced 1 or more falls resulting in injury, percentage of residents who received antipsychotic medication, and percentage of residents who received anti-anxiety medication. The independent variables were 4 residence characteristics: ownership type, size, occupancy, and region within the United States. We explored how different types of each residence characteristic compare for each QM.

Setting, participants, and measurements: Quality measure values from 15,420 CMS-supported nursing homes across the United States averaged over the 4 quarters of 2019 reporting were used. Welch’s analysis of variance was performed to examine whether the mean QM values for groups within each residential characteristic were statistically different.

Results: Publicly owned and low-occupancy residences had the highest mean QM values, indicating the poorest performance. Nonprofit and high-occupancy residences generally had the lowest (ie, best) mean QM values. There were significant differences in mean QM values among nursing home sizes and regions.

Conclusion: This study suggests that residence characteristics are related to 5 nursing home QMs. Results suggest that physical characteristics may be related to overall quality of life in nursing homes.

Keywords: quality of care, quality measures, residence characteristics, Alzheimer’s disease and related dementias.

More than 55 million people worldwide are living with Alzheimer’s disease and related dementias (ADRD).¹ With the aging of the Baby Boomer population, this number is expected to rise to more than 78 million worldwide by 2030.¹ Given the growing number of cognitively impaired older adults, there is an increased need for residences designed for the specialized care of this population. Although there are dozens of living options for the elderly, and although most specialized establishments have the resources to meet the immediate needs of their residents, many facilities lack universal design features that support a high quality of life for someone with ADRD or mild cognitive impairment. Previous research has shown relationships between behavioral and psychological symptoms of dementia (BPSD) and environmental characteristics such as acoustics, lighting, and indoor air temperature.^2,3 Physical behaviors of BPSD, including aggression and wandering, and psychological symptoms, such as depression, anxiety, and delusions, put residents at risk of injury.⁴ Additionally, BPSD is correlated with caregiver burden and stress.^5-8 Patients with dementia may also experience a lower stress threshold, changes in perception of space, and decreased short-term memory, creating environmental difficulties for those with ADRD⁹ that lead them to exhibit BPSD due to poor environmental design. Thus, there is a need to learn more about design features that minimize BPSD and promote a high quality of life for those with ADRD.¹⁰

Although research has shown relationships between physical environmental characteristics and BPSD, in this work we study relationships between possible BPSD indicators and 4 residence-level characteristics: ownership type, size, occupancy, and region in the United States (determined by location of the Centers for Medicare & Medicaid Services [CMS] regional offices). We analyzed data from the CMS Nursing Home Compare database for the year 2019.¹¹ This database publishes quarterly data and star ratings for quality-of-care measures (QMs), staffing levels, and health inspections for every nursing home supported by CMS. Previous research has investigated the accuracy of QM reporting for resident falls, the impact of residential characteristics on administration of antipsychotic medication, the influence of profit status on resident outcomes and quality of care, and the effect of nursing home size on quality of life.^12-16 Additionally, research suggests that residential characteristics such as size and location could be associated with infection control in nursing homes.¹⁷

Certain QMs, such as psychotropic drug administration, resident falls, and physical restraint, provide indicators of agitation, disorientation, or aggression, which are often signals of BPSD episodes. We hypothesized that residence types are associated with different QM scores, which could indicate different occurrences of BPSD. We selected 5 QMs for long-stay residents that could potentially be used as indicators of BPSD. Short-stay resident data were not included in this work to control for BPSD that could be a result of sheer unfamiliarity with the environment and confusion from being in a new home.

Methods

Design and Data Collection

This was a cross-sectional ecologic study aimed at exploring relationships between aggregate residential characteristics and QMs. Data were retrieved from the 2019 annual archives found in the CMS provider data catalog on nursing homes, including rehabilitation services.¹¹ The dataset provides general residence information, such as ownership, number of beds, number of residents, and location, as well as residence quality metrics, such as QMs, staffing data, and inspection data. Residence characteristics and 4-quarter averages of QMs were retrieved and used as cross-sectional data. The data used are from 15,420 residences across the United States. Nursing homes located in Guam, the US Pacific Territories, Puerto Rico, and the US Virgin Islands, while supported by CMS and included in the dataset, were excluded from the study due to a severe absence of QM data.

Dependent Variables

We investigated 5 QMs that were averaged across the 4 quarters of 2019. The QMs used as dependent variables were percentage of residents who displayed depressive symptoms (depression), percentage of residents who were physically restrained (restraint), percentage of residents who experienced 1 or more falls resulting in a major injury (falls), percentage of residents who received antipsychotic medication (antipsychotic medication), and percentage of residents who received anti-anxiety or hypnotic medication (anti-anxiety medication).

A total of 2471 QM values were unreported across the 5 QM analyzed: 501 residences did not report depression data; 479 did not report restraint data; 477 did not report falls data; 508 did not report antipsychotic medication data; and 506 did not report anti-anxiety medication data. A residence with a missing QM value was excluded from that respective analysis.

To assess the relationships among the different QMs, a Pearson correlation coefficient r was computed for each unique pair of QMs (Figure). All QMs studied were found to be very weakly or weakly correlated with one another using the Evans classification for very weak and weak correlations (r < 0.20 and 0.20 < r < 0.39, respectively).¹⁸

Independent Variables

A total of 15,420 residences were included in the study. Seventy-nine residences did not report occupancy data, however, so those residences were excluded from the occupancy analyses. We categorized the ownership of each nursing home as for-profit, nonprofit, or public. We categorized nursing home size, based on quartiles of the size distribution, as large (> 127 beds), medium (64 to 126 beds), and small (< 64 beds). This method for categorizing the residential characteristics was similar to that used in previous work.¹⁹ Similarly, we categorized nursing home occupancy as high (> 92% occupancy), medium (73% to 91% occupancy), and low (< 73% occupancy) based on quartiles of the occupancy distribution. For the regional analysis, we grouped states together based on the CMS regional offices: Atlanta, Georgia; Boston, Massachusetts; Chicago, Illinois; Dallas, Texas; Denver, Colorado; Kansas City, Missouri; New York, New York; Philadelphia, Pennsylvania; San Francisco, California; and Seattle, Washington.²⁰

Analyses

We used Levene’s test to determine whether variances among the residential groups were equal for each QM, using an a priori α = 0.05. For all 20 tests conducted (4 residential characteristics for all 5 QMs), the resulting F-statistics were significant, indicating that the assumption of homogeneity of variance was not met.

We therefore used Welch’s analysis of variance (ANOVA) to evaluate whether the groups within each residential characteristic were the same on their QM means. For example, we tested whether for-profit, nonprofit, and public residences had significantly different mean depression rates. For statistically significant differences, a Games-Howell post-hoc test was conducted to test the difference between all unique pairwise comparisons. An a priori α = 0.05 was used for both Welch’s ANOVA and post-hoc testing. All analyses were conducted in RStudio Version 1.2.5033 (Posit Software, PBC).

Mean Differences

Mean QM scores for the 5 QMs investigated, grouped by residential characteristic for the 2019 year of reporting, are shown in Table 1. It should be noted that the number of residences that reported occupancy data (n = 15,341) does not equal the total number of residences included in the study (N = 15,420) because 79 residences did not report occupancy data. For all QMs reported in Table 1, lower scores are better. Table 2 and Table 3 show results from pairwise comparisons of mean differences for the different residential characteristic and QM groupings. Mean differences and 95% CI are presented along with an indication of statistical significance (when applicable).

Ownership

Nonprofit residences had significantly lower (ie, better) mean scores than for-profit and public residences for 3 QMs: resident depression, antipsychotic medication use, and anti-anxiety medication use. For-profit and public residences did not significantly differ in their mean values for these QMs. For-profit residences had a significantly lower mean score for resident falls than both nonprofit and public residences, but no significant difference existed between scores for nonprofit and public residence falls. There were no statistically significant differences between mean restraint scores among the ownership types.

Size

Large (ie, high-capacity) residences had a significantly higher mean depression score than both medium and small residences, but there was not a significant difference between medium and small residences. Large residences had the significantly lowest mean score for resident falls, and medium residences scored significantly lower than small residences. Medium residences had a significantly higher mean score for anti-anxiety medication use than both small and large residences, but there was no significant difference between small and large residences. There were no statistically significant differences between mean scores for restraint and antipsychotic medication use among the nursing home sizes.

Occupancy

The mean scores for 4 out of the 5 QMs exhibited similar relationships with occupancy rates: resident depression, falls, and antipsychotic and anti-anxiety medication use. Low-occupancy residences consistently scored significantly higher than both medium- and high-occupancy residences, and medium-occupancy residences consistently scored significantly higher than high-occupancy residences. On average, high-occupancy (≥ 92%) residences reported better QM scores than low-occupancy (< 73%) and medium-occupancy (73% to 91%) residences for all the QMs studied except physical restraint, which yielded no significant results. These findings indicate a possible inverse relationship between building occupancy rate and these 4 QMs.

Region

Pairwise comparisons of mean QM scores by region are shown in Table 3. The Chicago region had a significantly higher mean depression score than all other regions, while the San Francisco region’s score was significantly lower than all other regions, except Atlanta and Boston. The Kansas City region had a significantly higher mean score for resident falls than all other regions, with the exception of Denver, and the San Francisco region scored significantly lower than all other regions in falls. The Boston region had a significantly higher mean score for administering antipsychotic medication than all other regions, except for Kansas City and Seattle, and the New York and San Francisco regions both had significantly lower scores than all other regions except for each other. The Atlanta region reported a significantly higher mean score for administering antianxiety medication than all other regions, and the Seattle region’s score for anti-anxiety medication use was significantly lower than all other regions except for San Francisco.

This study presented mean percentages for 5 QMs reported in the Nursing Home Compare database for the year 2019: depression, restraint, falls, antipsychotic medication use, and anti-anxiety medication use. We investigated these scores by 4 residential characteristics: ownership type, size, occupancy, and region. In general, publicly owned and low-occupancy residences had the highest scores, and thus the poorest performances, for the 5 chosen QMs during 2019. Nonprofit and high-occupancy residences generally had the lowest (ie, better) scores, and this result agrees with previous findings on long-stay nursing home residents.²¹ One possible explanation for better performance by high-occupancy buildings could be that increased social interaction is beneficial to nursing home residents as compared with low-occupancy buildings, where less social interaction is probable. It is difficult to draw conclusions regarding nursing home size and region; however, there are significant differences among sizes for 3 out of the 5 QMs and significant differences among regions for all 5 QMs. The analyses suggest that residence-level characteristics are related to QM scores. Although reported QMs are not a direct representation of resident quality of life, this work agrees with previous research that residential characteristics have some impact on the lives of nursing home residents.^13-17 Improvements in QM reporting and changes in quality improvement goals since the formation of Nursing Home Compare exist, suggesting that nursing homes’ awareness of their reporting duties may impact quality of care or reporting tendencies.^21,22 Future research should consider investigating the impacts of the COVID-19 pandemic on quality-reporting trends and QM scores.

Other physical characteristics of nursing homes, such as noise, lighting levels, and air quality, may also have an impact on QMs and possibly nursing home residents themselves. This type of data exploration could be included in future research. Additionally, future research could include a similar analysis over a longer period, rather than the 1-year period examined here, to investigate which types of residences consistently have high or low scores or how different types of residences have evolved over the years, particularly considering the impact of the COVID-19 pandemic. Information such as staffing levels, building renovations, and inspection data could be accounted for in future studies. Different QMs could also be investigated to better understand the influence of residential characteristics on quality of care.

Conclusion

This study suggests that residence-level characteristics are related to 5 reported nursing home QMs. Overall, nonprofit and high-occupancy residences had the lowest QM scores, indicating the highest performance. Although the results do not necessarily suggest that residence-level characteristics impact individual nursing home residents’ quality of life, they suggest that physical characteristics affect overall quality of life in nursing homes. Future research is needed to determine the specific physical characteristics of these residences that affect QM scores.

Corresponding author: Brian J. Puckett, [email protected].

Disclosures: None reported.

From the University of Nebraska, Lincoln (Mr. Puckett and Dr. Ryherd), University of Nebraska Medical Center, Omaha (Dr. Manley), and the University of Nebraska, Omaha (Dr. Ryan).

ABSTRACT

Objective: This study evaluated relationships between physical characteristics of nursing home residences and quality-of-care measures.

Design: This was a cross-sectional ecologic study. The dependent variables were 5 Centers for Medicare & Medicaid Services (CMS) Nursing Home Compare database long-stay quality measures (QMs) during 2019: percentage of residents who displayed depressive symptoms, percentage of residents who were physically restrained, percentage of residents who experienced 1 or more falls resulting in injury, percentage of residents who received antipsychotic medication, and percentage of residents who received anti-anxiety medication. The independent variables were 4 residence characteristics: ownership type, size, occupancy, and region within the United States. We explored how different types of each residence characteristic compare for each QM.

Setting, participants, and measurements: Quality measure values from 15,420 CMS-supported nursing homes across the United States averaged over the 4 quarters of 2019 reporting were used. Welch’s analysis of variance was performed to examine whether the mean QM values for groups within each residential characteristic were statistically different.

Results: Publicly owned and low-occupancy residences had the highest mean QM values, indicating the poorest performance. Nonprofit and high-occupancy residences generally had the lowest (ie, best) mean QM values. There were significant differences in mean QM values among nursing home sizes and regions.

Conclusion: This study suggests that residence characteristics are related to 5 nursing home QMs. Results suggest that physical characteristics may be related to overall quality of life in nursing homes.

Keywords: quality of care, quality measures, residence characteristics, Alzheimer’s disease and related dementias.

More than 55 million people worldwide are living with Alzheimer’s disease and related dementias (ADRD).¹ With the aging of the Baby Boomer population, this number is expected to rise to more than 78 million worldwide by 2030.¹ Given the growing number of cognitively impaired older adults, there is an increased need for residences designed for the specialized care of this population. Although there are dozens of living options for the elderly, and although most specialized establishments have the resources to meet the immediate needs of their residents, many facilities lack universal design features that support a high quality of life for someone with ADRD or mild cognitive impairment. Previous research has shown relationships between behavioral and psychological symptoms of dementia (BPSD) and environmental characteristics such as acoustics, lighting, and indoor air temperature.^2,3 Physical behaviors of BPSD, including aggression and wandering, and psychological symptoms, such as depression, anxiety, and delusions, put residents at risk of injury.⁴ Additionally, BPSD is correlated with caregiver burden and stress.^5-8 Patients with dementia may also experience a lower stress threshold, changes in perception of space, and decreased short-term memory, creating environmental difficulties for those with ADRD⁹ that lead them to exhibit BPSD due to poor environmental design. Thus, there is a need to learn more about design features that minimize BPSD and promote a high quality of life for those with ADRD.¹⁰

Although research has shown relationships between physical environmental characteristics and BPSD, in this work we study relationships between possible BPSD indicators and 4 residence-level characteristics: ownership type, size, occupancy, and region in the United States (determined by location of the Centers for Medicare & Medicaid Services [CMS] regional offices). We analyzed data from the CMS Nursing Home Compare database for the year 2019.¹¹ This database publishes quarterly data and star ratings for quality-of-care measures (QMs), staffing levels, and health inspections for every nursing home supported by CMS. Previous research has investigated the accuracy of QM reporting for resident falls, the impact of residential characteristics on administration of antipsychotic medication, the influence of profit status on resident outcomes and quality of care, and the effect of nursing home size on quality of life.^12-16 Additionally, research suggests that residential characteristics such as size and location could be associated with infection control in nursing homes.¹⁷

Certain QMs, such as psychotropic drug administration, resident falls, and physical restraint, provide indicators of agitation, disorientation, or aggression, which are often signals of BPSD episodes. We hypothesized that residence types are associated with different QM scores, which could indicate different occurrences of BPSD. We selected 5 QMs for long-stay residents that could potentially be used as indicators of BPSD. Short-stay resident data were not included in this work to control for BPSD that could be a result of sheer unfamiliarity with the environment and confusion from being in a new home.

Methods

Design and Data Collection

This was a cross-sectional ecologic study aimed at exploring relationships between aggregate residential characteristics and QMs. Data were retrieved from the 2019 annual archives found in the CMS provider data catalog on nursing homes, including rehabilitation services.¹¹ The dataset provides general residence information, such as ownership, number of beds, number of residents, and location, as well as residence quality metrics, such as QMs, staffing data, and inspection data. Residence characteristics and 4-quarter averages of QMs were retrieved and used as cross-sectional data. The data used are from 15,420 residences across the United States. Nursing homes located in Guam, the US Pacific Territories, Puerto Rico, and the US Virgin Islands, while supported by CMS and included in the dataset, were excluded from the study due to a severe absence of QM data.

Dependent Variables

We investigated 5 QMs that were averaged across the 4 quarters of 2019. The QMs used as dependent variables were percentage of residents who displayed depressive symptoms (depression), percentage of residents who were physically restrained (restraint), percentage of residents who experienced 1 or more falls resulting in a major injury (falls), percentage of residents who received antipsychotic medication (antipsychotic medication), and percentage of residents who received anti-anxiety or hypnotic medication (anti-anxiety medication).

A total of 2471 QM values were unreported across the 5 QM analyzed: 501 residences did not report depression data; 479 did not report restraint data; 477 did not report falls data; 508 did not report antipsychotic medication data; and 506 did not report anti-anxiety medication data. A residence with a missing QM value was excluded from that respective analysis.

To assess the relationships among the different QMs, a Pearson correlation coefficient r was computed for each unique pair of QMs (Figure). All QMs studied were found to be very weakly or weakly correlated with one another using the Evans classification for very weak and weak correlations (r < 0.20 and 0.20 < r < 0.39, respectively).¹⁸

Independent Variables

A total of 15,420 residences were included in the study. Seventy-nine residences did not report occupancy data, however, so those residences were excluded from the occupancy analyses. We categorized the ownership of each nursing home as for-profit, nonprofit, or public. We categorized nursing home size, based on quartiles of the size distribution, as large (> 127 beds), medium (64 to 126 beds), and small (< 64 beds). This method for categorizing the residential characteristics was similar to that used in previous work.¹⁹ Similarly, we categorized nursing home occupancy as high (> 92% occupancy), medium (73% to 91% occupancy), and low (< 73% occupancy) based on quartiles of the occupancy distribution. For the regional analysis, we grouped states together based on the CMS regional offices: Atlanta, Georgia; Boston, Massachusetts; Chicago, Illinois; Dallas, Texas; Denver, Colorado; Kansas City, Missouri; New York, New York; Philadelphia, Pennsylvania; San Francisco, California; and Seattle, Washington.²⁰

Analyses

We used Levene’s test to determine whether variances among the residential groups were equal for each QM, using an a priori α = 0.05. For all 20 tests conducted (4 residential characteristics for all 5 QMs), the resulting F-statistics were significant, indicating that the assumption of homogeneity of variance was not met.

We therefore used Welch’s analysis of variance (ANOVA) to evaluate whether the groups within each residential characteristic were the same on their QM means. For example, we tested whether for-profit, nonprofit, and public residences had significantly different mean depression rates. For statistically significant differences, a Games-Howell post-hoc test was conducted to test the difference between all unique pairwise comparisons. An a priori α = 0.05 was used for both Welch’s ANOVA and post-hoc testing. All analyses were conducted in RStudio Version 1.2.5033 (Posit Software, PBC).

Mean Differences

Mean QM scores for the 5 QMs investigated, grouped by residential characteristic for the 2019 year of reporting, are shown in Table 1. It should be noted that the number of residences that reported occupancy data (n = 15,341) does not equal the total number of residences included in the study (N = 15,420) because 79 residences did not report occupancy data. For all QMs reported in Table 1, lower scores are better. Table 2 and Table 3 show results from pairwise comparisons of mean differences for the different residential characteristic and QM groupings. Mean differences and 95% CI are presented along with an indication of statistical significance (when applicable).

Ownership

Nonprofit residences had significantly lower (ie, better) mean scores than for-profit and public residences for 3 QMs: resident depression, antipsychotic medication use, and anti-anxiety medication use. For-profit and public residences did not significantly differ in their mean values for these QMs. For-profit residences had a significantly lower mean score for resident falls than both nonprofit and public residences, but no significant difference existed between scores for nonprofit and public residence falls. There were no statistically significant differences between mean restraint scores among the ownership types.

Size

Large (ie, high-capacity) residences had a significantly higher mean depression score than both medium and small residences, but there was not a significant difference between medium and small residences. Large residences had the significantly lowest mean score for resident falls, and medium residences scored significantly lower than small residences. Medium residences had a significantly higher mean score for anti-anxiety medication use than both small and large residences, but there was no significant difference between small and large residences. There were no statistically significant differences between mean scores for restraint and antipsychotic medication use among the nursing home sizes.

Occupancy

The mean scores for 4 out of the 5 QMs exhibited similar relationships with occupancy rates: resident depression, falls, and antipsychotic and anti-anxiety medication use. Low-occupancy residences consistently scored significantly higher than both medium- and high-occupancy residences, and medium-occupancy residences consistently scored significantly higher than high-occupancy residences. On average, high-occupancy (≥ 92%) residences reported better QM scores than low-occupancy (< 73%) and medium-occupancy (73% to 91%) residences for all the QMs studied except physical restraint, which yielded no significant results. These findings indicate a possible inverse relationship between building occupancy rate and these 4 QMs.

Region

Pairwise comparisons of mean QM scores by region are shown in Table 3. The Chicago region had a significantly higher mean depression score than all other regions, while the San Francisco region’s score was significantly lower than all other regions, except Atlanta and Boston. The Kansas City region had a significantly higher mean score for resident falls than all other regions, with the exception of Denver, and the San Francisco region scored significantly lower than all other regions in falls. The Boston region had a significantly higher mean score for administering antipsychotic medication than all other regions, except for Kansas City and Seattle, and the New York and San Francisco regions both had significantly lower scores than all other regions except for each other. The Atlanta region reported a significantly higher mean score for administering antianxiety medication than all other regions, and the Seattle region’s score for anti-anxiety medication use was significantly lower than all other regions except for San Francisco.

This study presented mean percentages for 5 QMs reported in the Nursing Home Compare database for the year 2019: depression, restraint, falls, antipsychotic medication use, and anti-anxiety medication use. We investigated these scores by 4 residential characteristics: ownership type, size, occupancy, and region. In general, publicly owned and low-occupancy residences had the highest scores, and thus the poorest performances, for the 5 chosen QMs during 2019. Nonprofit and high-occupancy residences generally had the lowest (ie, better) scores, and this result agrees with previous findings on long-stay nursing home residents.²¹ One possible explanation for better performance by high-occupancy buildings could be that increased social interaction is beneficial to nursing home residents as compared with low-occupancy buildings, where less social interaction is probable. It is difficult to draw conclusions regarding nursing home size and region; however, there are significant differences among sizes for 3 out of the 5 QMs and significant differences among regions for all 5 QMs. The analyses suggest that residence-level characteristics are related to QM scores. Although reported QMs are not a direct representation of resident quality of life, this work agrees with previous research that residential characteristics have some impact on the lives of nursing home residents.^13-17 Improvements in QM reporting and changes in quality improvement goals since the formation of Nursing Home Compare exist, suggesting that nursing homes’ awareness of their reporting duties may impact quality of care or reporting tendencies.^21,22 Future research should consider investigating the impacts of the COVID-19 pandemic on quality-reporting trends and QM scores.

Other physical characteristics of nursing homes, such as noise, lighting levels, and air quality, may also have an impact on QMs and possibly nursing home residents themselves. This type of data exploration could be included in future research. Additionally, future research could include a similar analysis over a longer period, rather than the 1-year period examined here, to investigate which types of residences consistently have high or low scores or how different types of residences have evolved over the years, particularly considering the impact of the COVID-19 pandemic. Information such as staffing levels, building renovations, and inspection data could be accounted for in future studies. Different QMs could also be investigated to better understand the influence of residential characteristics on quality of care.

Conclusion

This study suggests that residence-level characteristics are related to 5 reported nursing home QMs. Overall, nonprofit and high-occupancy residences had the lowest QM scores, indicating the highest performance. Although the results do not necessarily suggest that residence-level characteristics impact individual nursing home residents’ quality of life, they suggest that physical characteristics affect overall quality of life in nursing homes. Future research is needed to determine the specific physical characteristics of these residences that affect QM scores.

Corresponding author: Brian J. Puckett, [email protected].

Disclosures: None reported.

Tuba City Regional Health Care Corporation (TCRHCC) is located on the Navajo Reservation in northeastern Arizona and provides medical coverage to a 6000-square-mile area, serving more than 33,000 residents of the Navajo, Hopi, and San Juan Southern Paiute tribes.^1,2 In 2021, there were 334,497 outpatient visits. TCRHCC departments involved in prescribing outpatient antibiotics include the emergency, internal medicine, family medicine, pediatrics, dentistry, surgery, podiatry, obstetrics and gynecology, and midwifery.

Antimicrobial resistance is one of the largest public health threats, causing an estimated 2 million infections and 23,000 deaths every year in the United States.³ This can lead to increased health care costs, morbidity, and mortality. A large, modifiable risk factor is the inappropriate prescribing of antibiotics: An estimated half of all outpatient antibiotics prescribed may be inappropriate in some manner, such as antibiotic choice, dosing, or duration. In addition, at least 30% of US antibiotic prescriptions are unnecessary, leading to significant overuse.³ As such, antimicrobial stewardship is a cornerstone of improving antibiotic use, patient care, and safety.

The goals of antimicrobial stewardship are to measure antimicrobial prescribing, improve clinician prescribing, minimize misdiagnosis or delayed diagnoses, and ensure the right drug, dose, and duration are selected when antimicrobial therapy is appropriate.³ The Centers for Disease Control and Prevention recommends 4 core elements of outpatient antimicrobial stewardship: commitment, action for policy and practice, tracking and reporting, and education and expertise.³ This study focuses on the pillars of action for policy and practice and tracking and reporting.

Methods

The study objectives were not designed to achieve statistical power. A retrospective chart review was performed for patients of any age who were seen in an ambulatory care setting at TCRHCC from August 1, 2020, to August 1, 2021, with a visit diagnosis included in the outpatient antimicrobial prescribing guidelines.^4,5 A random sample of 10% of charts of each diagnosis code was used for analysis. An Excel spreadsheet with all patient charts, separated by diagnosis code, was created. Each chart was then assigned a number, and the Excel function RAND was used to select a random number from the pool. This was continued until 10% of each category, or at least 1 chart from diagnosis code categories with less than 10 total charts available, were selected.

Inclusion criteria were patients seen in ambulatory clinics or the emergency department, an infectious disease diagnosis addressed in the facility guidelines, diagnosis and treatment occurred between August 1, 2020, and August 1, 2021, and the patient was discharged home after the visit. Exclusion criteria were patients who required inpatient admission, patient visits to the clinic established solely for COVID-19 vaccination or testing as no other care was ever provided at this location, patients who refused treatment, patients who failed empiric therapy and required treatment adjustments, or patients who were initially treated and received an antibiotic prescription at a facility outside the TCRHCC system.

After chart review and analysis were completed, a prescriber survey and educational intervention were performed from March 2, 2022, to March 31, 2022. This consisted of an anonymous survey to gather demographic data and prescribing habits pre-education, a short educational brief on the existence, location, and recommended use of the outpatient antimicrobial prescribing guidelines, and a posteducation survey to assess knowledge of the guidelines and willingness to adhere to them after the educational intervention.

Results

We reviewed 8779 patient records. A random sample of 10% of the records of each diagnosis code was taken and 876 charts were reviewed. Of the charts reviewed, 351 patients met the inclusion criteria and were included in the analysis. A goal of 90% was established as the target for prescriber adherence for the study based on author consensus for a reasonable goal. Of the 351 evaluated charts, 62 (16.1%) were pediatric patients (aged < 19 years) and 289 (83.9%) were adults (aged ≥ 19 years). Fifty-two (84%) of the pediatric charts and 249 (86%) of the adult charts demonstrated prescribers had appropriately followed guidelines for a combined total of 301 of the 351 charts and an overall adherence rate of 86%. This was 4 points below the established goal of 90%, warranting further investigation. An analysis of prescribers and locations revealed no trends or patterns of nonadherence. A prescriber survey and educational intervention were designed and disseminated to all prescribers at the facility with the approval and assistance of the chief of medicine.

Thirty-nine prescribers responded to the survey. In the pre-educational survey, clinical resources were the most common source of guidance with 36 prescribers (92%) indicating they used them to make an appropriate selection of an antimicrobial; 32 (82%) used personal knowledge, 30 (77%) used culture results, and 24 (62%) used facility guidelines. This was consistent with the posteducational questions: 12 (31%) indicated they were not aware of the facility guidelines before the educational intervention.

Discussion

This study’s objective was to evaluate prescriber adherence to the facility outpatient prescribing guidelines after they were implemented in 2019 and to plan for interventions if necessary. Overall prescriber adherence was high with 86% of the sampled charts adherent. This was below the goal of 90%, so evaluation of the nonadherent charts was warranted for the determination of any patterns to guide the planned interventions with the facility prescribers. However, no trends were identified, so the intervention was designed as a general survey and educational session for all prescribers. Overall prescriber response was positive, with a total of 34 responding prescribers (87%) indicating a willingness to use the guidelines.

Limitations

This is a retrospective observational study performed through chart review that allowed for frequency analysis but did not allow for statistical analysis, so the significance of results cannot be obtained. Additionally, this study was not able to compare rates of adherence before and after the educational intervention, so the effectiveness of the intervention cannot be assessed.

Conclusions

This retrospective observational study’s data demonstrate that prescribers are adhering at a high rate to recommended empiric antimicrobials for outpatient treatment with an 86% adherence rate. Response to educational intervention indicated a larger proportion of prescribers than previously will use the guidelines. However, the impact this will have on appropriate prescribing rates in the future could not be assessed during this study.

Tuba City Regional Health Care Corporation (TCRHCC) is located on the Navajo Reservation in northeastern Arizona and provides medical coverage to a 6000-square-mile area, serving more than 33,000 residents of the Navajo, Hopi, and San Juan Southern Paiute tribes.^1,2 In 2021, there were 334,497 outpatient visits. TCRHCC departments involved in prescribing outpatient antibiotics include the emergency, internal medicine, family medicine, pediatrics, dentistry, surgery, podiatry, obstetrics and gynecology, and midwifery.

Antimicrobial resistance is one of the largest public health threats, causing an estimated 2 million infections and 23,000 deaths every year in the United States.³ This can lead to increased health care costs, morbidity, and mortality. A large, modifiable risk factor is the inappropriate prescribing of antibiotics: An estimated half of all outpatient antibiotics prescribed may be inappropriate in some manner, such as antibiotic choice, dosing, or duration. In addition, at least 30% of US antibiotic prescriptions are unnecessary, leading to significant overuse.³ As such, antimicrobial stewardship is a cornerstone of improving antibiotic use, patient care, and safety.

The goals of antimicrobial stewardship are to measure antimicrobial prescribing, improve clinician prescribing, minimize misdiagnosis or delayed diagnoses, and ensure the right drug, dose, and duration are selected when antimicrobial therapy is appropriate.³ The Centers for Disease Control and Prevention recommends 4 core elements of outpatient antimicrobial stewardship: commitment, action for policy and practice, tracking and reporting, and education and expertise.³ This study focuses on the pillars of action for policy and practice and tracking and reporting.

Methods

The study objectives were not designed to achieve statistical power. A retrospective chart review was performed for patients of any age who were seen in an ambulatory care setting at TCRHCC from August 1, 2020, to August 1, 2021, with a visit diagnosis included in the outpatient antimicrobial prescribing guidelines.^4,5 A random sample of 10% of charts of each diagnosis code was used for analysis. An Excel spreadsheet with all patient charts, separated by diagnosis code, was created. Each chart was then assigned a number, and the Excel function RAND was used to select a random number from the pool. This was continued until 10% of each category, or at least 1 chart from diagnosis code categories with less than 10 total charts available, were selected.

Inclusion criteria were patients seen in ambulatory clinics or the emergency department, an infectious disease diagnosis addressed in the facility guidelines, diagnosis and treatment occurred between August 1, 2020, and August 1, 2021, and the patient was discharged home after the visit. Exclusion criteria were patients who required inpatient admission, patient visits to the clinic established solely for COVID-19 vaccination or testing as no other care was ever provided at this location, patients who refused treatment, patients who failed empiric therapy and required treatment adjustments, or patients who were initially treated and received an antibiotic prescription at a facility outside the TCRHCC system.

After chart review and analysis were completed, a prescriber survey and educational intervention were performed from March 2, 2022, to March 31, 2022. This consisted of an anonymous survey to gather demographic data and prescribing habits pre-education, a short educational brief on the existence, location, and recommended use of the outpatient antimicrobial prescribing guidelines, and a posteducation survey to assess knowledge of the guidelines and willingness to adhere to them after the educational intervention.

Results

We reviewed 8779 patient records. A random sample of 10% of the records of each diagnosis code was taken and 876 charts were reviewed. Of the charts reviewed, 351 patients met the inclusion criteria and were included in the analysis. A goal of 90% was established as the target for prescriber adherence for the study based on author consensus for a reasonable goal. Of the 351 evaluated charts, 62 (16.1%) were pediatric patients (aged < 19 years) and 289 (83.9%) were adults (aged ≥ 19 years). Fifty-two (84%) of the pediatric charts and 249 (86%) of the adult charts demonstrated prescribers had appropriately followed guidelines for a combined total of 301 of the 351 charts and an overall adherence rate of 86%. This was 4 points below the established goal of 90%, warranting further investigation. An analysis of prescribers and locations revealed no trends or patterns of nonadherence. A prescriber survey and educational intervention were designed and disseminated to all prescribers at the facility with the approval and assistance of the chief of medicine.

Thirty-nine prescribers responded to the survey. In the pre-educational survey, clinical resources were the most common source of guidance with 36 prescribers (92%) indicating they used them to make an appropriate selection of an antimicrobial; 32 (82%) used personal knowledge, 30 (77%) used culture results, and 24 (62%) used facility guidelines. This was consistent with the posteducational questions: 12 (31%) indicated they were not aware of the facility guidelines before the educational intervention.

Discussion

This study’s objective was to evaluate prescriber adherence to the facility outpatient prescribing guidelines after they were implemented in 2019 and to plan for interventions if necessary. Overall prescriber adherence was high with 86% of the sampled charts adherent. This was below the goal of 90%, so evaluation of the nonadherent charts was warranted for the determination of any patterns to guide the planned interventions with the facility prescribers. However, no trends were identified, so the intervention was designed as a general survey and educational session for all prescribers. Overall prescriber response was positive, with a total of 34 responding prescribers (87%) indicating a willingness to use the guidelines.

Limitations

This is a retrospective observational study performed through chart review that allowed for frequency analysis but did not allow for statistical analysis, so the significance of results cannot be obtained. Additionally, this study was not able to compare rates of adherence before and after the educational intervention, so the effectiveness of the intervention cannot be assessed.

Conclusions

This retrospective observational study’s data demonstrate that prescribers are adhering at a high rate to recommended empiric antimicrobials for outpatient treatment with an 86% adherence rate. Response to educational intervention indicated a larger proportion of prescribers than previously will use the guidelines. However, the impact this will have on appropriate prescribing rates in the future could not be assessed during this study.

Tuba City Regional Health Care Corporation (TCRHCC) is located on the Navajo Reservation in northeastern Arizona and provides medical coverage to a 6000-square-mile area, serving more than 33,000 residents of the Navajo, Hopi, and San Juan Southern Paiute tribes.^1,2 In 2021, there were 334,497 outpatient visits. TCRHCC departments involved in prescribing outpatient antibiotics include the emergency, internal medicine, family medicine, pediatrics, dentistry, surgery, podiatry, obstetrics and gynecology, and midwifery.

Antimicrobial resistance is one of the largest public health threats, causing an estimated 2 million infections and 23,000 deaths every year in the United States.³ This can lead to increased health care costs, morbidity, and mortality. A large, modifiable risk factor is the inappropriate prescribing of antibiotics: An estimated half of all outpatient antibiotics prescribed may be inappropriate in some manner, such as antibiotic choice, dosing, or duration. In addition, at least 30% of US antibiotic prescriptions are unnecessary, leading to significant overuse.³ As such, antimicrobial stewardship is a cornerstone of improving antibiotic use, patient care, and safety.

The goals of antimicrobial stewardship are to measure antimicrobial prescribing, improve clinician prescribing, minimize misdiagnosis or delayed diagnoses, and ensure the right drug, dose, and duration are selected when antimicrobial therapy is appropriate.³ The Centers for Disease Control and Prevention recommends 4 core elements of outpatient antimicrobial stewardship: commitment, action for policy and practice, tracking and reporting, and education and expertise.³ This study focuses on the pillars of action for policy and practice and tracking and reporting.

Methods

The study objectives were not designed to achieve statistical power. A retrospective chart review was performed for patients of any age who were seen in an ambulatory care setting at TCRHCC from August 1, 2020, to August 1, 2021, with a visit diagnosis included in the outpatient antimicrobial prescribing guidelines.^4,5 A random sample of 10% of charts of each diagnosis code was used for analysis. An Excel spreadsheet with all patient charts, separated by diagnosis code, was created. Each chart was then assigned a number, and the Excel function RAND was used to select a random number from the pool. This was continued until 10% of each category, or at least 1 chart from diagnosis code categories with less than 10 total charts available, were selected.

Inclusion criteria were patients seen in ambulatory clinics or the emergency department, an infectious disease diagnosis addressed in the facility guidelines, diagnosis and treatment occurred between August 1, 2020, and August 1, 2021, and the patient was discharged home after the visit. Exclusion criteria were patients who required inpatient admission, patient visits to the clinic established solely for COVID-19 vaccination or testing as no other care was ever provided at this location, patients who refused treatment, patients who failed empiric therapy and required treatment adjustments, or patients who were initially treated and received an antibiotic prescription at a facility outside the TCRHCC system.

After chart review and analysis were completed, a prescriber survey and educational intervention were performed from March 2, 2022, to March 31, 2022. This consisted of an anonymous survey to gather demographic data and prescribing habits pre-education, a short educational brief on the existence, location, and recommended use of the outpatient antimicrobial prescribing guidelines, and a posteducation survey to assess knowledge of the guidelines and willingness to adhere to them after the educational intervention.

Results

We reviewed 8779 patient records. A random sample of 10% of the records of each diagnosis code was taken and 876 charts were reviewed. Of the charts reviewed, 351 patients met the inclusion criteria and were included in the analysis. A goal of 90% was established as the target for prescriber adherence for the study based on author consensus for a reasonable goal. Of the 351 evaluated charts, 62 (16.1%) were pediatric patients (aged < 19 years) and 289 (83.9%) were adults (aged ≥ 19 years). Fifty-two (84%) of the pediatric charts and 249 (86%) of the adult charts demonstrated prescribers had appropriately followed guidelines for a combined total of 301 of the 351 charts and an overall adherence rate of 86%. This was 4 points below the established goal of 90%, warranting further investigation. An analysis of prescribers and locations revealed no trends or patterns of nonadherence. A prescriber survey and educational intervention were designed and disseminated to all prescribers at the facility with the approval and assistance of the chief of medicine.

Thirty-nine prescribers responded to the survey. In the pre-educational survey, clinical resources were the most common source of guidance with 36 prescribers (92%) indicating they used them to make an appropriate selection of an antimicrobial; 32 (82%) used personal knowledge, 30 (77%) used culture results, and 24 (62%) used facility guidelines. This was consistent with the posteducational questions: 12 (31%) indicated they were not aware of the facility guidelines before the educational intervention.

Discussion

This study’s objective was to evaluate prescriber adherence to the facility outpatient prescribing guidelines after they were implemented in 2019 and to plan for interventions if necessary. Overall prescriber adherence was high with 86% of the sampled charts adherent. This was below the goal of 90%, so evaluation of the nonadherent charts was warranted for the determination of any patterns to guide the planned interventions with the facility prescribers. However, no trends were identified, so the intervention was designed as a general survey and educational session for all prescribers. Overall prescriber response was positive, with a total of 34 responding prescribers (87%) indicating a willingness to use the guidelines.

Limitations

This is a retrospective observational study performed through chart review that allowed for frequency analysis but did not allow for statistical analysis, so the significance of results cannot be obtained. Additionally, this study was not able to compare rates of adherence before and after the educational intervention, so the effectiveness of the intervention cannot be assessed.

Conclusions

This retrospective observational study’s data demonstrate that prescribers are adhering at a high rate to recommended empiric antimicrobials for outpatient treatment with an 86% adherence rate. Response to educational intervention indicated a larger proportion of prescribers than previously will use the guidelines. However, the impact this will have on appropriate prescribing rates in the future could not be assessed during this study.

Building the health care system of the future requires the thoughtful development and integration of innovative technologies to positively transform care.^1-4 Extended reality (XR) represents a spectrum of emerging technologies that have the potential to enhance health care. This includes virtual reality (VR), where a computer-generated visual experience fills the screen; augmented reality (AR), which allows users to see computer-generated images superimposed into an otherwise normal real-world field of view; and mixed reality (MR), which allows users to interact and manipulate computer-generated AR images.

Clinicians and researchers have begun exploring the potential of XR to address a wide variety of health care challenges. A recent systematic review concluded that many clinical studies in this area have small sample sizes and are in the preclinical, proof-of-concept stage, but demonstrate the potential and impact of the underlying VR, AR, and MR technologies.⁵ Common emerging health care uses for XR include medical education, training, presurgical planning, surgical guidance, distraction therapy for pain and anxiety, and home health indications, including rehabilitation.^5-39

Methods

The VA Palo Alto Health Care System (VAPAHCS) consists of 3 inpatient hospitals and 7 outpatient clinics, provides a full range of care services to > 90,000 enrolled veterans with 800 hospital beds, 3 nursing homes, and a 100-bed domiciliary. The facility also runs data-driven care projects in research, innovation, and evidence-based practice group under nursing services.⁴⁸ This project was performed by the VA National Center for Collaborative Healthcare Innovation at the VAPAHCS campus.

The combined technical system used for this assessment included a wireless communication network, AR/MR hardware, and software. Medivis AnatomyX software displayed an interactive human anatomy atlas segmented into about 6000 individual interactive parts. Medivis SurgicalAR received US Food and Drug Administration clearance for presurgical planning and was used to transform and display deidentified diagnostic images (eg, magnetic resonance images and computed tomography) in 3-dimensional (3D) interactive holograms (Figures 1 and 2).

 The wireless Microsoft HoloLens 2 AR/MR headset was used for viewing and sensor-enabled collaborative interaction. Multiple participants in the same physical location simultaneously participated and interacted with 3D holograms. The interactive hologram data were enabled for 3D stereoscopic viewing and manipulation.

Setting and Participants

We reviewed published studies that used questionnaires to evaluate institutions’ level of innovation and new technology user acceptance to develop the questionnaire.^49-56 Questions and methods were modified, with a focus on understanding the impact on hospital employees. The questionnaire consisted of 2 predemonstration and 3 postdemonstration sections. The first section included background questions. The second (predemonstration) and third (postdemonstration) sections provided matched questions on feelings about the VA. The fourth section included 2 unmatched questions about how the participant felt this technology would impact veterans and whether the VA should implement similar technologies. We used a 5-point Likert scale for sections 2, 3 and 4 (1 = not at all to 5 = extremely). Two unmatched free-text questions asked how the technology could be used in the participant’s hospital service, and another open-ended question asked for any additional comments. To reduce potential reporting bias, 2 VA employees that did not work at VAPAHCS assisted with the survey distribution and collection. VAPAHCS staff were informed by all employee email and facility intranet of the opportunity to participate; the voluntary demonstration and survey took place on February 10 and 11, 2020.

Data Analysis

All matching pre/post questions were analyzed together to determine statistically significant differences using the Wilcoxon signed rank matched pairs test and pooled t test. Survey respondents were also grouped by employment type to evaluate the impact on subgroups. Results were also grouped by VA tenure into 4 categorical 10-year increments (0-10, 11-20, 21-30, 31-40). Additionally, analysis of variance (ANOVA) was performed on employment types and VA tenure to understand whether there was a statistically significant difference in responses by these subgroups. Respondents’ optional free-text answers were manually reviewed by 2 authors (ZPV and DMA), classified, coded by the common themes, and analyzed for comparison.

Results

A total of 166 participants completed the predemonstration survey, which was a requirement for participating in the AR demonstration. Of those, 159 staff members (95.8%) also completed at least part of the postdemonstration paired structured questions, and their results were included in the analysis.

Paired Questions

For questions about how innovative the VA is, 108 of 152 participants (71.1%) provided higher scores after the demonstration, 42 (27.6%) had no change, and 2 (1.3%) provided decreased scores. The mean innovative score increased from 3.4 predemonstration to 4.5 postdemonstration on a Likert scale, which is a 1.1 point increase from predemonstration to postdemonstration (95% CI, 0.9- 1.2) or a 22% increase (95% CI, 18%-24%) (P < .001). Respondents level of excitement about VA also increased with 82 of 157 participants (52.2%) providing higher scores after the demonstration, 71 (45.2%) had no change, and 4 scores (2.5%) decreased. The predemonstration mean excitement score of 3.7 increased to 4.3 postdemonstration, which is a 0.6 point increase from before to after the demonstration (95% CI, 0.5-0.7) or a 12% increase (95% CI, 10%-14%) (P < .001). In the survey, 36 of 149 participants (24.2%) had higher scores for their expectation to continue working at VA postdemonstration, 109 (73.2%) had no change, and 4 scores (2.7%) decreased. The mean employee retention score increased from 4.2 predemonstration to 4.5 postdemonstration, which is a 0.3 point increase between pre/post (95% CI, 0.2-0.4) or a 6% increase (95% CI, 4%-8%) (P < .001)

The pre/post questions were analyzed using 1-way ANOVA by hospital department and VA tenure. The responses by department were not statistically significant. Of the 159 employees assessed, 101 respondents (63.5%) had 0 to 10 years VA tenure, 44 (27.7%) had 11 to 20 years, 10 (6.3%) had 21 to 30 years, and 4 (2.5%) had > 31 to 40 years. Length of VA tenure did not impact respondent excitement. Respondents opinions on innovation in the 0 to 10 year and the 11 to 20 year groups rose from 3.2 and 3.7 predemonstration to 4.3 and 4.6 postdemonstration, respectively (P < .001 for both statistical comparisons) (Table 2). Interestingly, the 0 to 10 group saw a 9% rise from a 4.0 score predemonstration to a 4.4 score postdemonstration (P < .001), indicating that the demonstration had a positive impact on their plans to continue employment at VA (Table 3).

Sex did not play a significant role in how respondents answered questions regarding VA excitement or innovation. However, there was a statistically significant difference in how male and female respondents answered the predemonstration question about their plans to continue VA employment, according to the Wilcoxon rank sum test. Predemonstration, female respondents had a mean score of 4.1, which was 6% lower than the 4.4 score of male colleagues (P = .04). Veteran status did have an impact on how respondents felt about VA innovation, and their plans to continue employment at VA. After the demonstration, veteran staff felt the VA was more innovative compared with nonveterans: 4.7 vs 4.4, respectively, a 6% difference (P = .02) Similarly, for the continued VA employment question, veterans had a mean score of 4.8 vs 4.4 for nonveterans, an 8% difference (P = .03) These results suggest that the demonstration had more of an impact on veteran employees vs nonveteran employees.

Unpaired Questions

There were 2 structured unpaired postdemonstration questions. Respondents agreed that similar technology would impact veteran health care with mean (SD) of 4.6 (0.6) and a median score of 5 on a 5-point Likert scale. Respondents also agreed on the importance of implementing similar innovations with mean (SD) of 4.7 (0.5), and a median score of 5.

The survey asked how this technology could benefit their hospital service department and had 64 responses. Forty-six respondents saw applications for education or patient care/surgery. Other responses shared excitement about the technology and its potential to positively impact patient education. There were 37 responses to the open-ended question: 21 respondents expressed excitement for the technology, and 10 respondents reiterated that the demonstration would be of benefit to patient care/surgery and training.

Discussion

Successful development, design, and deployment of any new health care tool depends on leveraging insights from the employees that will be using and supporting these systems. Correspondingly, understanding the impact that advanced technologies have on health care employees’ satisfaction, morale, and retention is critical to our overall institutional strategy. Our findings show that a one-time experience with AR/MR technology elicited positive employee reactions. Of note, the survey revealed statistically significant improvements in staff’s view of the VA, with the greatest positive impact for questions about innovation, followed by excitement to work at the VA, and likelihood to continue work at the VA. It is very disruptive and costly when health care employees leave, and improving employee satisfaction and morale is important for better patient care and patient satisfaction, which is priority for VAPAHCS leadership.^57-62

The paired predemonstration and postdemonstration scores were similarly high, nearing the top threshold available for the Likert scale (4.3 to 4.5). Furthermore, the least incremental improvement for these responses was observed for topics that had the highest initial baseline score. Therefore, the improvements observed for the paired questions may have more to do with the high baseline values.

Of additional interest, the self-reported likelihood of continuing to work at the VA increased the most for female employees, veteran employees, and employees with the least number of years at the VA. These demographic differences have important implications for VA staff recruitment and retention strategies.⁶² The unpaired questions about the impact on veteran care and whether the VA should continue similar work demonstrated extremely high support with median scores of 5 for both questions. The free-text postdemonstration responses also demonstrate similar positive themes, with a disposition for excitement about both the training and patient care applications for this technology. In addition, respondents felt strongly that this and other similar technologies will positively impact the health care for veterans and that the VA should continue these efforts.

Strengths and Limitations

A strength of this assessment is the ability to evaluate survey responses that were systematically collected and matched from the same individual immediately before and after exposure to the new technology. The free-text responses provided additional important information that both confirmed the results and provided additional valued supplementary guidance for future implementation strategies, which is critical for our translational implementation goals. An additional strength is that the voluntary surveys were managed by non-VAPAHCS colleagues, limiting potential bias. Importantly, the number of respondents allowed a statistically significant assessment of important health care employee metrics. These results have emphasized how being part of an innovative organization, and the introduction of advanced AR/MR technology, improve employees’ satisfaction and morale about where they work as well as their intention to stay at their institution.

A limitation of this assessment was the lack of comparative data for employee acceptance of other technologies at VAPAHCS. This limits our ability to differentiate whether the strong positive results observed in this evaluation were a result of the specific technology assessed, or of new and advanced health care technology in general. Nonetheless, our unpaired questions, which received extremely high scores, also included participant questions about comparing the system with other similar technologies. This assessment was also focused on veteran care, which limits generalizability.

Conclusions

One-time exposure to advanced AR technology for health care significantly increased employee morale as measured by excitement about working at the VA as well as employee intention to continue employment at the VA. These collateral benefits of the technology are particularly important in health care because our employees are our most important asset and improving employee morale equates to better patient care. Positive impacts were most pronounced for women employees, newer VA employees, and employees who are also veterans. These more detailed insights are also positioned to have a direct impact on employee recruitment and retention strategies. Additional valuable insights regarding the most applicable use of the technology in the clinical setting were also obtained. 

Acknowledgments

We thank Andrew Spiegelman, Hyewon Kim, Jonathan Sills, and Alexander Erickson for their assistance in developing the survey questions. We also thank Jason Rhodes and Mark Bulson for traveling to our facility to assist with managing the anonymous surveys during the demonstration event.

Building the health care system of the future requires the thoughtful development and integration of innovative technologies to positively transform care.^1-4 Extended reality (XR) represents a spectrum of emerging technologies that have the potential to enhance health care. This includes virtual reality (VR), where a computer-generated visual experience fills the screen; augmented reality (AR), which allows users to see computer-generated images superimposed into an otherwise normal real-world field of view; and mixed reality (MR), which allows users to interact and manipulate computer-generated AR images.

Clinicians and researchers have begun exploring the potential of XR to address a wide variety of health care challenges. A recent systematic review concluded that many clinical studies in this area have small sample sizes and are in the preclinical, proof-of-concept stage, but demonstrate the potential and impact of the underlying VR, AR, and MR technologies.⁵ Common emerging health care uses for XR include medical education, training, presurgical planning, surgical guidance, distraction therapy for pain and anxiety, and home health indications, including rehabilitation.^5-39

Methods

The VA Palo Alto Health Care System (VAPAHCS) consists of 3 inpatient hospitals and 7 outpatient clinics, provides a full range of care services to > 90,000 enrolled veterans with 800 hospital beds, 3 nursing homes, and a 100-bed domiciliary. The facility also runs data-driven care projects in research, innovation, and evidence-based practice group under nursing services.⁴⁸ This project was performed by the VA National Center for Collaborative Healthcare Innovation at the VAPAHCS campus.

The combined technical system used for this assessment included a wireless communication network, AR/MR hardware, and software. Medivis AnatomyX software displayed an interactive human anatomy atlas segmented into about 6000 individual interactive parts. Medivis SurgicalAR received US Food and Drug Administration clearance for presurgical planning and was used to transform and display deidentified diagnostic images (eg, magnetic resonance images and computed tomography) in 3-dimensional (3D) interactive holograms (Figures 1 and 2).

 The wireless Microsoft HoloLens 2 AR/MR headset was used for viewing and sensor-enabled collaborative interaction. Multiple participants in the same physical location simultaneously participated and interacted with 3D holograms. The interactive hologram data were enabled for 3D stereoscopic viewing and manipulation.

Setting and Participants

We reviewed published studies that used questionnaires to evaluate institutions’ level of innovation and new technology user acceptance to develop the questionnaire.^49-56 Questions and methods were modified, with a focus on understanding the impact on hospital employees. The questionnaire consisted of 2 predemonstration and 3 postdemonstration sections. The first section included background questions. The second (predemonstration) and third (postdemonstration) sections provided matched questions on feelings about the VA. The fourth section included 2 unmatched questions about how the participant felt this technology would impact veterans and whether the VA should implement similar technologies. We used a 5-point Likert scale for sections 2, 3 and 4 (1 = not at all to 5 = extremely). Two unmatched free-text questions asked how the technology could be used in the participant’s hospital service, and another open-ended question asked for any additional comments. To reduce potential reporting bias, 2 VA employees that did not work at VAPAHCS assisted with the survey distribution and collection. VAPAHCS staff were informed by all employee email and facility intranet of the opportunity to participate; the voluntary demonstration and survey took place on February 10 and 11, 2020.

Data Analysis

All matching pre/post questions were analyzed together to determine statistically significant differences using the Wilcoxon signed rank matched pairs test and pooled t test. Survey respondents were also grouped by employment type to evaluate the impact on subgroups. Results were also grouped by VA tenure into 4 categorical 10-year increments (0-10, 11-20, 21-30, 31-40). Additionally, analysis of variance (ANOVA) was performed on employment types and VA tenure to understand whether there was a statistically significant difference in responses by these subgroups. Respondents’ optional free-text answers were manually reviewed by 2 authors (ZPV and DMA), classified, coded by the common themes, and analyzed for comparison.

Results

A total of 166 participants completed the predemonstration survey, which was a requirement for participating in the AR demonstration. Of those, 159 staff members (95.8%) also completed at least part of the postdemonstration paired structured questions, and their results were included in the analysis.

Paired Questions

For questions about how innovative the VA is, 108 of 152 participants (71.1%) provided higher scores after the demonstration, 42 (27.6%) had no change, and 2 (1.3%) provided decreased scores. The mean innovative score increased from 3.4 predemonstration to 4.5 postdemonstration on a Likert scale, which is a 1.1 point increase from predemonstration to postdemonstration (95% CI, 0.9- 1.2) or a 22% increase (95% CI, 18%-24%) (P < .001). Respondents level of excitement about VA also increased with 82 of 157 participants (52.2%) providing higher scores after the demonstration, 71 (45.2%) had no change, and 4 scores (2.5%) decreased. The predemonstration mean excitement score of 3.7 increased to 4.3 postdemonstration, which is a 0.6 point increase from before to after the demonstration (95% CI, 0.5-0.7) or a 12% increase (95% CI, 10%-14%) (P < .001). In the survey, 36 of 149 participants (24.2%) had higher scores for their expectation to continue working at VA postdemonstration, 109 (73.2%) had no change, and 4 scores (2.7%) decreased. The mean employee retention score increased from 4.2 predemonstration to 4.5 postdemonstration, which is a 0.3 point increase between pre/post (95% CI, 0.2-0.4) or a 6% increase (95% CI, 4%-8%) (P < .001)

The pre/post questions were analyzed using 1-way ANOVA by hospital department and VA tenure. The responses by department were not statistically significant. Of the 159 employees assessed, 101 respondents (63.5%) had 0 to 10 years VA tenure, 44 (27.7%) had 11 to 20 years, 10 (6.3%) had 21 to 30 years, and 4 (2.5%) had > 31 to 40 years. Length of VA tenure did not impact respondent excitement. Respondents opinions on innovation in the 0 to 10 year and the 11 to 20 year groups rose from 3.2 and 3.7 predemonstration to 4.3 and 4.6 postdemonstration, respectively (P < .001 for both statistical comparisons) (Table 2). Interestingly, the 0 to 10 group saw a 9% rise from a 4.0 score predemonstration to a 4.4 score postdemonstration (P < .001), indicating that the demonstration had a positive impact on their plans to continue employment at VA (Table 3).

Sex did not play a significant role in how respondents answered questions regarding VA excitement or innovation. However, there was a statistically significant difference in how male and female respondents answered the predemonstration question about their plans to continue VA employment, according to the Wilcoxon rank sum test. Predemonstration, female respondents had a mean score of 4.1, which was 6% lower than the 4.4 score of male colleagues (P = .04). Veteran status did have an impact on how respondents felt about VA innovation, and their plans to continue employment at VA. After the demonstration, veteran staff felt the VA was more innovative compared with nonveterans: 4.7 vs 4.4, respectively, a 6% difference (P = .02) Similarly, for the continued VA employment question, veterans had a mean score of 4.8 vs 4.4 for nonveterans, an 8% difference (P = .03) These results suggest that the demonstration had more of an impact on veteran employees vs nonveteran employees.

Unpaired Questions

There were 2 structured unpaired postdemonstration questions. Respondents agreed that similar technology would impact veteran health care with mean (SD) of 4.6 (0.6) and a median score of 5 on a 5-point Likert scale. Respondents also agreed on the importance of implementing similar innovations with mean (SD) of 4.7 (0.5), and a median score of 5.

The survey asked how this technology could benefit their hospital service department and had 64 responses. Forty-six respondents saw applications for education or patient care/surgery. Other responses shared excitement about the technology and its potential to positively impact patient education. There were 37 responses to the open-ended question: 21 respondents expressed excitement for the technology, and 10 respondents reiterated that the demonstration would be of benefit to patient care/surgery and training.

Discussion

Successful development, design, and deployment of any new health care tool depends on leveraging insights from the employees that will be using and supporting these systems. Correspondingly, understanding the impact that advanced technologies have on health care employees’ satisfaction, morale, and retention is critical to our overall institutional strategy. Our findings show that a one-time experience with AR/MR technology elicited positive employee reactions. Of note, the survey revealed statistically significant improvements in staff’s view of the VA, with the greatest positive impact for questions about innovation, followed by excitement to work at the VA, and likelihood to continue work at the VA. It is very disruptive and costly when health care employees leave, and improving employee satisfaction and morale is important for better patient care and patient satisfaction, which is priority for VAPAHCS leadership.^57-62

The paired predemonstration and postdemonstration scores were similarly high, nearing the top threshold available for the Likert scale (4.3 to 4.5). Furthermore, the least incremental improvement for these responses was observed for topics that had the highest initial baseline score. Therefore, the improvements observed for the paired questions may have more to do with the high baseline values.

Of additional interest, the self-reported likelihood of continuing to work at the VA increased the most for female employees, veteran employees, and employees with the least number of years at the VA. These demographic differences have important implications for VA staff recruitment and retention strategies.⁶² The unpaired questions about the impact on veteran care and whether the VA should continue similar work demonstrated extremely high support with median scores of 5 for both questions. The free-text postdemonstration responses also demonstrate similar positive themes, with a disposition for excitement about both the training and patient care applications for this technology. In addition, respondents felt strongly that this and other similar technologies will positively impact the health care for veterans and that the VA should continue these efforts.

Strengths and Limitations

A strength of this assessment is the ability to evaluate survey responses that were systematically collected and matched from the same individual immediately before and after exposure to the new technology. The free-text responses provided additional important information that both confirmed the results and provided additional valued supplementary guidance for future implementation strategies, which is critical for our translational implementation goals. An additional strength is that the voluntary surveys were managed by non-VAPAHCS colleagues, limiting potential bias. Importantly, the number of respondents allowed a statistically significant assessment of important health care employee metrics. These results have emphasized how being part of an innovative organization, and the introduction of advanced AR/MR technology, improve employees’ satisfaction and morale about where they work as well as their intention to stay at their institution.

A limitation of this assessment was the lack of comparative data for employee acceptance of other technologies at VAPAHCS. This limits our ability to differentiate whether the strong positive results observed in this evaluation were a result of the specific technology assessed, or of new and advanced health care technology in general. Nonetheless, our unpaired questions, which received extremely high scores, also included participant questions about comparing the system with other similar technologies. This assessment was also focused on veteran care, which limits generalizability.

Conclusions

One-time exposure to advanced AR technology for health care significantly increased employee morale as measured by excitement about working at the VA as well as employee intention to continue employment at the VA. These collateral benefits of the technology are particularly important in health care because our employees are our most important asset and improving employee morale equates to better patient care. Positive impacts were most pronounced for women employees, newer VA employees, and employees who are also veterans. These more detailed insights are also positioned to have a direct impact on employee recruitment and retention strategies. Additional valuable insights regarding the most applicable use of the technology in the clinical setting were also obtained. 

Acknowledgments

We thank Andrew Spiegelman, Hyewon Kim, Jonathan Sills, and Alexander Erickson for their assistance in developing the survey questions. We also thank Jason Rhodes and Mark Bulson for traveling to our facility to assist with managing the anonymous surveys during the demonstration event.

Building the health care system of the future requires the thoughtful development and integration of innovative technologies to positively transform care.^1-4 Extended reality (XR) represents a spectrum of emerging technologies that have the potential to enhance health care. This includes virtual reality (VR), where a computer-generated visual experience fills the screen; augmented reality (AR), which allows users to see computer-generated images superimposed into an otherwise normal real-world field of view; and mixed reality (MR), which allows users to interact and manipulate computer-generated AR images.

Clinicians and researchers have begun exploring the potential of XR to address a wide variety of health care challenges. A recent systematic review concluded that many clinical studies in this area have small sample sizes and are in the preclinical, proof-of-concept stage, but demonstrate the potential and impact of the underlying VR, AR, and MR technologies.⁵ Common emerging health care uses for XR include medical education, training, presurgical planning, surgical guidance, distraction therapy for pain and anxiety, and home health indications, including rehabilitation.^5-39

Methods

The VA Palo Alto Health Care System (VAPAHCS) consists of 3 inpatient hospitals and 7 outpatient clinics, provides a full range of care services to > 90,000 enrolled veterans with 800 hospital beds, 3 nursing homes, and a 100-bed domiciliary. The facility also runs data-driven care projects in research, innovation, and evidence-based practice group under nursing services.⁴⁸ This project was performed by the VA National Center for Collaborative Healthcare Innovation at the VAPAHCS campus.

The combined technical system used for this assessment included a wireless communication network, AR/MR hardware, and software. Medivis AnatomyX software displayed an interactive human anatomy atlas segmented into about 6000 individual interactive parts. Medivis SurgicalAR received US Food and Drug Administration clearance for presurgical planning and was used to transform and display deidentified diagnostic images (eg, magnetic resonance images and computed tomography) in 3-dimensional (3D) interactive holograms (Figures 1 and 2).

 The wireless Microsoft HoloLens 2 AR/MR headset was used for viewing and sensor-enabled collaborative interaction. Multiple participants in the same physical location simultaneously participated and interacted with 3D holograms. The interactive hologram data were enabled for 3D stereoscopic viewing and manipulation.

Setting and Participants

We reviewed published studies that used questionnaires to evaluate institutions’ level of innovation and new technology user acceptance to develop the questionnaire.^49-56 Questions and methods were modified, with a focus on understanding the impact on hospital employees. The questionnaire consisted of 2 predemonstration and 3 postdemonstration sections. The first section included background questions. The second (predemonstration) and third (postdemonstration) sections provided matched questions on feelings about the VA. The fourth section included 2 unmatched questions about how the participant felt this technology would impact veterans and whether the VA should implement similar technologies. We used a 5-point Likert scale for sections 2, 3 and 4 (1 = not at all to 5 = extremely). Two unmatched free-text questions asked how the technology could be used in the participant’s hospital service, and another open-ended question asked for any additional comments. To reduce potential reporting bias, 2 VA employees that did not work at VAPAHCS assisted with the survey distribution and collection. VAPAHCS staff were informed by all employee email and facility intranet of the opportunity to participate; the voluntary demonstration and survey took place on February 10 and 11, 2020.

Data Analysis

All matching pre/post questions were analyzed together to determine statistically significant differences using the Wilcoxon signed rank matched pairs test and pooled t test. Survey respondents were also grouped by employment type to evaluate the impact on subgroups. Results were also grouped by VA tenure into 4 categorical 10-year increments (0-10, 11-20, 21-30, 31-40). Additionally, analysis of variance (ANOVA) was performed on employment types and VA tenure to understand whether there was a statistically significant difference in responses by these subgroups. Respondents’ optional free-text answers were manually reviewed by 2 authors (ZPV and DMA), classified, coded by the common themes, and analyzed for comparison.

Results

A total of 166 participants completed the predemonstration survey, which was a requirement for participating in the AR demonstration. Of those, 159 staff members (95.8%) also completed at least part of the postdemonstration paired structured questions, and their results were included in the analysis.

Paired Questions

For questions about how innovative the VA is, 108 of 152 participants (71.1%) provided higher scores after the demonstration, 42 (27.6%) had no change, and 2 (1.3%) provided decreased scores. The mean innovative score increased from 3.4 predemonstration to 4.5 postdemonstration on a Likert scale, which is a 1.1 point increase from predemonstration to postdemonstration (95% CI, 0.9- 1.2) or a 22% increase (95% CI, 18%-24%) (P < .001). Respondents level of excitement about VA also increased with 82 of 157 participants (52.2%) providing higher scores after the demonstration, 71 (45.2%) had no change, and 4 scores (2.5%) decreased. The predemonstration mean excitement score of 3.7 increased to 4.3 postdemonstration, which is a 0.6 point increase from before to after the demonstration (95% CI, 0.5-0.7) or a 12% increase (95% CI, 10%-14%) (P < .001). In the survey, 36 of 149 participants (24.2%) had higher scores for their expectation to continue working at VA postdemonstration, 109 (73.2%) had no change, and 4 scores (2.7%) decreased. The mean employee retention score increased from 4.2 predemonstration to 4.5 postdemonstration, which is a 0.3 point increase between pre/post (95% CI, 0.2-0.4) or a 6% increase (95% CI, 4%-8%) (P < .001)

The pre/post questions were analyzed using 1-way ANOVA by hospital department and VA tenure. The responses by department were not statistically significant. Of the 159 employees assessed, 101 respondents (63.5%) had 0 to 10 years VA tenure, 44 (27.7%) had 11 to 20 years, 10 (6.3%) had 21 to 30 years, and 4 (2.5%) had > 31 to 40 years. Length of VA tenure did not impact respondent excitement. Respondents opinions on innovation in the 0 to 10 year and the 11 to 20 year groups rose from 3.2 and 3.7 predemonstration to 4.3 and 4.6 postdemonstration, respectively (P < .001 for both statistical comparisons) (Table 2). Interestingly, the 0 to 10 group saw a 9% rise from a 4.0 score predemonstration to a 4.4 score postdemonstration (P < .001), indicating that the demonstration had a positive impact on their plans to continue employment at VA (Table 3).

Sex did not play a significant role in how respondents answered questions regarding VA excitement or innovation. However, there was a statistically significant difference in how male and female respondents answered the predemonstration question about their plans to continue VA employment, according to the Wilcoxon rank sum test. Predemonstration, female respondents had a mean score of 4.1, which was 6% lower than the 4.4 score of male colleagues (P = .04). Veteran status did have an impact on how respondents felt about VA innovation, and their plans to continue employment at VA. After the demonstration, veteran staff felt the VA was more innovative compared with nonveterans: 4.7 vs 4.4, respectively, a 6% difference (P = .02) Similarly, for the continued VA employment question, veterans had a mean score of 4.8 vs 4.4 for nonveterans, an 8% difference (P = .03) These results suggest that the demonstration had more of an impact on veteran employees vs nonveteran employees.

Unpaired Questions

There were 2 structured unpaired postdemonstration questions. Respondents agreed that similar technology would impact veteran health care with mean (SD) of 4.6 (0.6) and a median score of 5 on a 5-point Likert scale. Respondents also agreed on the importance of implementing similar innovations with mean (SD) of 4.7 (0.5), and a median score of 5.

The survey asked how this technology could benefit their hospital service department and had 64 responses. Forty-six respondents saw applications for education or patient care/surgery. Other responses shared excitement about the technology and its potential to positively impact patient education. There were 37 responses to the open-ended question: 21 respondents expressed excitement for the technology, and 10 respondents reiterated that the demonstration would be of benefit to patient care/surgery and training.

Discussion

Successful development, design, and deployment of any new health care tool depends on leveraging insights from the employees that will be using and supporting these systems. Correspondingly, understanding the impact that advanced technologies have on health care employees’ satisfaction, morale, and retention is critical to our overall institutional strategy. Our findings show that a one-time experience with AR/MR technology elicited positive employee reactions. Of note, the survey revealed statistically significant improvements in staff’s view of the VA, with the greatest positive impact for questions about innovation, followed by excitement to work at the VA, and likelihood to continue work at the VA. It is very disruptive and costly when health care employees leave, and improving employee satisfaction and morale is important for better patient care and patient satisfaction, which is priority for VAPAHCS leadership.^57-62

The paired predemonstration and postdemonstration scores were similarly high, nearing the top threshold available for the Likert scale (4.3 to 4.5). Furthermore, the least incremental improvement for these responses was observed for topics that had the highest initial baseline score. Therefore, the improvements observed for the paired questions may have more to do with the high baseline values.

Of additional interest, the self-reported likelihood of continuing to work at the VA increased the most for female employees, veteran employees, and employees with the least number of years at the VA. These demographic differences have important implications for VA staff recruitment and retention strategies.⁶² The unpaired questions about the impact on veteran care and whether the VA should continue similar work demonstrated extremely high support with median scores of 5 for both questions. The free-text postdemonstration responses also demonstrate similar positive themes, with a disposition for excitement about both the training and patient care applications for this technology. In addition, respondents felt strongly that this and other similar technologies will positively impact the health care for veterans and that the VA should continue these efforts.

Strengths and Limitations

A strength of this assessment is the ability to evaluate survey responses that were systematically collected and matched from the same individual immediately before and after exposure to the new technology. The free-text responses provided additional important information that both confirmed the results and provided additional valued supplementary guidance for future implementation strategies, which is critical for our translational implementation goals. An additional strength is that the voluntary surveys were managed by non-VAPAHCS colleagues, limiting potential bias. Importantly, the number of respondents allowed a statistically significant assessment of important health care employee metrics. These results have emphasized how being part of an innovative organization, and the introduction of advanced AR/MR technology, improve employees’ satisfaction and morale about where they work as well as their intention to stay at their institution.

A limitation of this assessment was the lack of comparative data for employee acceptance of other technologies at VAPAHCS. This limits our ability to differentiate whether the strong positive results observed in this evaluation were a result of the specific technology assessed, or of new and advanced health care technology in general. Nonetheless, our unpaired questions, which received extremely high scores, also included participant questions about comparing the system with other similar technologies. This assessment was also focused on veteran care, which limits generalizability.

Conclusions

One-time exposure to advanced AR technology for health care significantly increased employee morale as measured by excitement about working at the VA as well as employee intention to continue employment at the VA. These collateral benefits of the technology are particularly important in health care because our employees are our most important asset and improving employee morale equates to better patient care. Positive impacts were most pronounced for women employees, newer VA employees, and employees who are also veterans. These more detailed insights are also positioned to have a direct impact on employee recruitment and retention strategies. Additional valuable insights regarding the most applicable use of the technology in the clinical setting were also obtained. 

Acknowledgments

We thank Andrew Spiegelman, Hyewon Kim, Jonathan Sills, and Alexander Erickson for their assistance in developing the survey questions. We also thank Jason Rhodes and Mark Bulson for traveling to our facility to assist with managing the anonymous surveys during the demonstration event.

Dermatology providers are at an increased risk for blood-borne pathogen (BBP) exposures during procedures in clinical practice.^1-3 Current data regarding the characterization of these exposures are limited. Prior studies are based on surveys that result in low response rates and potential for selection bias. Donnelly et al¹ reported a 26% response rate in a national survey-based study evaluating BBP exposures in resident physicians, fellows, and practicing dermatologists, with 85% of respondents reporting at least 1 injury. Similarly, Goulart et al² reported a 35% response rate in a survey evaluating sharps injuries in residents and medical students, with 85% reporting a sharps injury. In addition, there are conflicting data regarding characteristics of these exposures, including common implicated instruments and procedures.^1-3 Prior studies also have not evaluated exposures in all members of dermatologic staff, including resident physicians, practicing dermatologists, and ancillary staff.

To make appropriate quality improvements in dermatologic procedures, a more comprehensive understanding of BBP exposures is needed. We conducted a retrospective review of BBP incidence reports to identify the incidence of BBP events among all dermatologic staff, including resident physicians, practicing dermatologists, and ancillary staff. We further investigated the type of exposure, the type of procedure associated with each exposure, anatomic locations of exposures, and instruments involved in each exposure.

Methods

Data on BBP exposures in the dermatology departments were obtained from the occupational health departments at each of 3 Mayo Clinic sites—Scottsdale, Arizona; Jacksonville, Florida; and Rochester, Minnesota—from March 2010 through January 2021. The institutional review board at Mayo Clinic, Scottsdale, Arizona, granted approval of this study (IRB #20-012625). A retrospective review of each exposure was conducted to identify the incidence of BBP exposures. Occupational BBP exposure was defined as any percutaneous injury or mucosal exposure with foreign blood, tissue, or other bodily fluids that placed the health care worker at risk for communicable infections. Secondary aims included identification of the type of exposure, type of procedure associated with each exposure, common anatomic locations of exposures, and common instruments involved in each exposure.

Statistical Analysis—Variables were summarized using counts and percentages. The 3 most common categories for each variable were then compared among occupational groups using the Fisher exact test. All other categories were grouped for analysis purposes. Medical staff were categorized into 3 occupational groups: practicing dermatologists; resident physicians; and ancillary staff, including nurse/medical assistants, physician assistants, and clinical laboratory technologists. All analyses were 2 sided and considered statistically significant at P<.05. Analyses were performed using SAS 9.4 (SAS Institute Inc).

Results

Type of Exposure—A total of 222 BBP exposures were identified through the trisite retrospective review from March 2010 through January 2021. One hundred ninety-nine (89.6%) of 222 exposures were attributed to needlesticks and medical sharps, while 23 (10.4%) of 222 exposures were attributed to splash incidents (Table).

Anatomic Sites Affected—The anatomic location most frequently involved was the thumb (130/217 events [59.9%]), followed by the hand (39/217 events [18.0%]) and finger (22/217 events [10.1%]). The arm, face, and knee were affected with the lowest frequency, with only 1 event reported at each anatomic site (0.5%)(eTable). Five incidents were excluded from the analysis of anatomic location because of insufficient details of events.

Incident Tasks and Tools—Most BBP exposures occurred during suturing or assisting with suturing (64/210 events [30.5%]), followed by handling of sharps, wires, or instruments (40/210 events [19.0%]) and medication administration (37/210 events [17.6%])(eTable). Twelve incidents were excluded from the analysis of implicated tasks because of insufficient details of events.

The tools involved in exposure events with the greatest prevalence included the suture needle (76/201 events [37.8%]), injection syringe/needle (43/201 events [21.4%]), and shave biopsy razor (24/201 events [11.9%])(eTable). Twenty-one incidents were excluded from the analysis of implicated instruments because of insufficient details of events.

Providers Affected by BBP Exposures—Resident physicians experienced the greatest number of BBP exposures (105/222 events [47.3%]), followed by ancillary providers (84/222 events [37.8%]) and practicing dermatologists (33/222 events [14.9%]). All occupational groups experienced more BBP exposures through needlesticks/medical sharps compared with splash incidents (resident physicians, 88.6%; ancillary staff, 91.7%; practicing dermatologists, 87.9%; P=.725)(Table).

Among resident physicians, practicing dermatologists, and ancillary staff, the most frequent site of injury was the thumb. Suturing/assisting with suturing was the most common task leading to injury, and the suture needle was the most common instrument of injury for both resident physicians and practicing dermatologists. Handling of sharps, wires, or instruments was the most common task leading to injury for ancillary staff, and the injection syringe/needle was the most common instrument of injury in this cohort.

Resident physicians experienced the lowest rate of BBP exposures during administration of medications (12.7%; P=.003). Ancillary staff experienced the highest rate of BBP exposures with an injection needle (35.5%; P=.001). There were no statistically significant differences among occupational groups for the anatomic location of injury (P=.074)(eTable).

Comment

In the year 2000, the annual global incidence of occupational BBP exposures among health care workers worldwide for hepatitis B virus, hepatitis C virus, and HIV was estimated at 2.1 million, 926,000, and 327,000, respectively. Most of these exposures were due to sharps injuries.⁴ Dermatologists are particularly at risk for BBP exposures given their reliance on frequent procedures in practice. During an 11-year period, 222 BBP exposures were documented in the dermatology departments at 3 Mayo Clinic institutions. Most exposures were due to needlestick/sharps across all occupational groups compared with splash injuries. Prior survey studies confirm that sharps injuries are frequently implicated, with 75% to 94% of residents and practicing dermatologists reporting at least 1 sharps injury.¹

Among occupational groups, resident physicians had the highest rate of BBP exposures, followed by nurse/medical assistants and practicing dermatologists, which may be secondary to lack of training or experience. Data from other surgical fields, including general surgery, support that resident physicians have the highest rate of sharps injuries.⁵ In a survey study (N=452), 51% of residents reported that extra training in safe techniques would be beneficial.² Safety training may be beneficial in reducing the incidence of BBP exposures in residency programs.

The most common implicated task in resident physicians and practicing dermatologists was suturing or assisting with suturing, and the most common implicated instrument was the suture needle. Prior studies showed conflicting data regarding common implicated tasks and instruments in this cohort.^1,2 The task of suturing and the suture needle also were the most implicated means of injury among other surgical specialties.⁶ Ancillary staff experienced most BBP exposures during handling of sharps, wires, or instruments, as well as the use of an injection needle. The designation of tasks among dermatologic staff likely explains the difference among occupational groups. This new information may provide the opportunity to improve safety measures among all members of the dermatologic team.

Limitations—There are several limitations to this study. This retrospective review was conducted at a single health system at 3 institutions. Hence, similar safety protocols likely were in place across all sites, which may reduce the generalizability of the results. In addition, there is risk of nonreporting bias among staff, as only documented incidence reports were evaluated. Prior studies demonstrated a nonreporting prevalence of 33% to 64% among dermatology staff.^1-3 We also did not evaluate whether injuries resulted in BBP exposure or transmission. The rates of postexposure prophylaxis also were not studied. This information was not available for review because of concerns for privacy. Demographic features, such as gender or years of training, also were not evaluated.

Conclusion

This study provides additional insight on the incidence of BBP exposures in dermatology, as well as the implicated tasks, instruments, and anatomic locations of injury. Studies show that implementing formal education regarding the risks of BBP exposure may result in reduction of sharps injuries.⁷ Formal education in residency programs may be needed in the field of dermatology to reduce BBP exposures. Quality improvement measures should focus on identified risk factors among occupational groups to reduce BBP exposures in the workplace.

Dermatology providers are at an increased risk for blood-borne pathogen (BBP) exposures during procedures in clinical practice.^1-3 Current data regarding the characterization of these exposures are limited. Prior studies are based on surveys that result in low response rates and potential for selection bias. Donnelly et al¹ reported a 26% response rate in a national survey-based study evaluating BBP exposures in resident physicians, fellows, and practicing dermatologists, with 85% of respondents reporting at least 1 injury. Similarly, Goulart et al² reported a 35% response rate in a survey evaluating sharps injuries in residents and medical students, with 85% reporting a sharps injury. In addition, there are conflicting data regarding characteristics of these exposures, including common implicated instruments and procedures.^1-3 Prior studies also have not evaluated exposures in all members of dermatologic staff, including resident physicians, practicing dermatologists, and ancillary staff.

To make appropriate quality improvements in dermatologic procedures, a more comprehensive understanding of BBP exposures is needed. We conducted a retrospective review of BBP incidence reports to identify the incidence of BBP events among all dermatologic staff, including resident physicians, practicing dermatologists, and ancillary staff. We further investigated the type of exposure, the type of procedure associated with each exposure, anatomic locations of exposures, and instruments involved in each exposure.

Methods

Data on BBP exposures in the dermatology departments were obtained from the occupational health departments at each of 3 Mayo Clinic sites—Scottsdale, Arizona; Jacksonville, Florida; and Rochester, Minnesota—from March 2010 through January 2021. The institutional review board at Mayo Clinic, Scottsdale, Arizona, granted approval of this study (IRB #20-012625). A retrospective review of each exposure was conducted to identify the incidence of BBP exposures. Occupational BBP exposure was defined as any percutaneous injury or mucosal exposure with foreign blood, tissue, or other bodily fluids that placed the health care worker at risk for communicable infections. Secondary aims included identification of the type of exposure, type of procedure associated with each exposure, common anatomic locations of exposures, and common instruments involved in each exposure.

Statistical Analysis—Variables were summarized using counts and percentages. The 3 most common categories for each variable were then compared among occupational groups using the Fisher exact test. All other categories were grouped for analysis purposes. Medical staff were categorized into 3 occupational groups: practicing dermatologists; resident physicians; and ancillary staff, including nurse/medical assistants, physician assistants, and clinical laboratory technologists. All analyses were 2 sided and considered statistically significant at P<.05. Analyses were performed using SAS 9.4 (SAS Institute Inc).

Results

Type of Exposure—A total of 222 BBP exposures were identified through the trisite retrospective review from March 2010 through January 2021. One hundred ninety-nine (89.6%) of 222 exposures were attributed to needlesticks and medical sharps, while 23 (10.4%) of 222 exposures were attributed to splash incidents (Table).

Anatomic Sites Affected—The anatomic location most frequently involved was the thumb (130/217 events [59.9%]), followed by the hand (39/217 events [18.0%]) and finger (22/217 events [10.1%]). The arm, face, and knee were affected with the lowest frequency, with only 1 event reported at each anatomic site (0.5%)(eTable). Five incidents were excluded from the analysis of anatomic location because of insufficient details of events.

Incident Tasks and Tools—Most BBP exposures occurred during suturing or assisting with suturing (64/210 events [30.5%]), followed by handling of sharps, wires, or instruments (40/210 events [19.0%]) and medication administration (37/210 events [17.6%])(eTable). Twelve incidents were excluded from the analysis of implicated tasks because of insufficient details of events.

The tools involved in exposure events with the greatest prevalence included the suture needle (76/201 events [37.8%]), injection syringe/needle (43/201 events [21.4%]), and shave biopsy razor (24/201 events [11.9%])(eTable). Twenty-one incidents were excluded from the analysis of implicated instruments because of insufficient details of events.

Providers Affected by BBP Exposures—Resident physicians experienced the greatest number of BBP exposures (105/222 events [47.3%]), followed by ancillary providers (84/222 events [37.8%]) and practicing dermatologists (33/222 events [14.9%]). All occupational groups experienced more BBP exposures through needlesticks/medical sharps compared with splash incidents (resident physicians, 88.6%; ancillary staff, 91.7%; practicing dermatologists, 87.9%; P=.725)(Table).

Among resident physicians, practicing dermatologists, and ancillary staff, the most frequent site of injury was the thumb. Suturing/assisting with suturing was the most common task leading to injury, and the suture needle was the most common instrument of injury for both resident physicians and practicing dermatologists. Handling of sharps, wires, or instruments was the most common task leading to injury for ancillary staff, and the injection syringe/needle was the most common instrument of injury in this cohort.

Resident physicians experienced the lowest rate of BBP exposures during administration of medications (12.7%; P=.003). Ancillary staff experienced the highest rate of BBP exposures with an injection needle (35.5%; P=.001). There were no statistically significant differences among occupational groups for the anatomic location of injury (P=.074)(eTable).

Comment

In the year 2000, the annual global incidence of occupational BBP exposures among health care workers worldwide for hepatitis B virus, hepatitis C virus, and HIV was estimated at 2.1 million, 926,000, and 327,000, respectively. Most of these exposures were due to sharps injuries.⁴ Dermatologists are particularly at risk for BBP exposures given their reliance on frequent procedures in practice. During an 11-year period, 222 BBP exposures were documented in the dermatology departments at 3 Mayo Clinic institutions. Most exposures were due to needlestick/sharps across all occupational groups compared with splash injuries. Prior survey studies confirm that sharps injuries are frequently implicated, with 75% to 94% of residents and practicing dermatologists reporting at least 1 sharps injury.¹

Among occupational groups, resident physicians had the highest rate of BBP exposures, followed by nurse/medical assistants and practicing dermatologists, which may be secondary to lack of training or experience. Data from other surgical fields, including general surgery, support that resident physicians have the highest rate of sharps injuries.⁵ In a survey study (N=452), 51% of residents reported that extra training in safe techniques would be beneficial.² Safety training may be beneficial in reducing the incidence of BBP exposures in residency programs.

The most common implicated task in resident physicians and practicing dermatologists was suturing or assisting with suturing, and the most common implicated instrument was the suture needle. Prior studies showed conflicting data regarding common implicated tasks and instruments in this cohort.^1,2 The task of suturing and the suture needle also were the most implicated means of injury among other surgical specialties.⁶ Ancillary staff experienced most BBP exposures during handling of sharps, wires, or instruments, as well as the use of an injection needle. The designation of tasks among dermatologic staff likely explains the difference among occupational groups. This new information may provide the opportunity to improve safety measures among all members of the dermatologic team.

Limitations—There are several limitations to this study. This retrospective review was conducted at a single health system at 3 institutions. Hence, similar safety protocols likely were in place across all sites, which may reduce the generalizability of the results. In addition, there is risk of nonreporting bias among staff, as only documented incidence reports were evaluated. Prior studies demonstrated a nonreporting prevalence of 33% to 64% among dermatology staff.^1-3 We also did not evaluate whether injuries resulted in BBP exposure or transmission. The rates of postexposure prophylaxis also were not studied. This information was not available for review because of concerns for privacy. Demographic features, such as gender or years of training, also were not evaluated.

Conclusion

This study provides additional insight on the incidence of BBP exposures in dermatology, as well as the implicated tasks, instruments, and anatomic locations of injury. Studies show that implementing formal education regarding the risks of BBP exposure may result in reduction of sharps injuries.⁷ Formal education in residency programs may be needed in the field of dermatology to reduce BBP exposures. Quality improvement measures should focus on identified risk factors among occupational groups to reduce BBP exposures in the workplace.

Dermatology providers are at an increased risk for blood-borne pathogen (BBP) exposures during procedures in clinical practice.^1-3 Current data regarding the characterization of these exposures are limited. Prior studies are based on surveys that result in low response rates and potential for selection bias. Donnelly et al¹ reported a 26% response rate in a national survey-based study evaluating BBP exposures in resident physicians, fellows, and practicing dermatologists, with 85% of respondents reporting at least 1 injury. Similarly, Goulart et al² reported a 35% response rate in a survey evaluating sharps injuries in residents and medical students, with 85% reporting a sharps injury. In addition, there are conflicting data regarding characteristics of these exposures, including common implicated instruments and procedures.^1-3 Prior studies also have not evaluated exposures in all members of dermatologic staff, including resident physicians, practicing dermatologists, and ancillary staff.

To make appropriate quality improvements in dermatologic procedures, a more comprehensive understanding of BBP exposures is needed. We conducted a retrospective review of BBP incidence reports to identify the incidence of BBP events among all dermatologic staff, including resident physicians, practicing dermatologists, and ancillary staff. We further investigated the type of exposure, the type of procedure associated with each exposure, anatomic locations of exposures, and instruments involved in each exposure.

Methods

Data on BBP exposures in the dermatology departments were obtained from the occupational health departments at each of 3 Mayo Clinic sites—Scottsdale, Arizona; Jacksonville, Florida; and Rochester, Minnesota—from March 2010 through January 2021. The institutional review board at Mayo Clinic, Scottsdale, Arizona, granted approval of this study (IRB #20-012625). A retrospective review of each exposure was conducted to identify the incidence of BBP exposures. Occupational BBP exposure was defined as any percutaneous injury or mucosal exposure with foreign blood, tissue, or other bodily fluids that placed the health care worker at risk for communicable infections. Secondary aims included identification of the type of exposure, type of procedure associated with each exposure, common anatomic locations of exposures, and common instruments involved in each exposure.

Statistical Analysis—Variables were summarized using counts and percentages. The 3 most common categories for each variable were then compared among occupational groups using the Fisher exact test. All other categories were grouped for analysis purposes. Medical staff were categorized into 3 occupational groups: practicing dermatologists; resident physicians; and ancillary staff, including nurse/medical assistants, physician assistants, and clinical laboratory technologists. All analyses were 2 sided and considered statistically significant at P<.05. Analyses were performed using SAS 9.4 (SAS Institute Inc).

Results

Type of Exposure—A total of 222 BBP exposures were identified through the trisite retrospective review from March 2010 through January 2021. One hundred ninety-nine (89.6%) of 222 exposures were attributed to needlesticks and medical sharps, while 23 (10.4%) of 222 exposures were attributed to splash incidents (Table).

Anatomic Sites Affected—The anatomic location most frequently involved was the thumb (130/217 events [59.9%]), followed by the hand (39/217 events [18.0%]) and finger (22/217 events [10.1%]). The arm, face, and knee were affected with the lowest frequency, with only 1 event reported at each anatomic site (0.5%)(eTable). Five incidents were excluded from the analysis of anatomic location because of insufficient details of events.

Incident Tasks and Tools—Most BBP exposures occurred during suturing or assisting with suturing (64/210 events [30.5%]), followed by handling of sharps, wires, or instruments (40/210 events [19.0%]) and medication administration (37/210 events [17.6%])(eTable). Twelve incidents were excluded from the analysis of implicated tasks because of insufficient details of events.

The tools involved in exposure events with the greatest prevalence included the suture needle (76/201 events [37.8%]), injection syringe/needle (43/201 events [21.4%]), and shave biopsy razor (24/201 events [11.9%])(eTable). Twenty-one incidents were excluded from the analysis of implicated instruments because of insufficient details of events.

Providers Affected by BBP Exposures—Resident physicians experienced the greatest number of BBP exposures (105/222 events [47.3%]), followed by ancillary providers (84/222 events [37.8%]) and practicing dermatologists (33/222 events [14.9%]). All occupational groups experienced more BBP exposures through needlesticks/medical sharps compared with splash incidents (resident physicians, 88.6%; ancillary staff, 91.7%; practicing dermatologists, 87.9%; P=.725)(Table).

Among resident physicians, practicing dermatologists, and ancillary staff, the most frequent site of injury was the thumb. Suturing/assisting with suturing was the most common task leading to injury, and the suture needle was the most common instrument of injury for both resident physicians and practicing dermatologists. Handling of sharps, wires, or instruments was the most common task leading to injury for ancillary staff, and the injection syringe/needle was the most common instrument of injury in this cohort.

Resident physicians experienced the lowest rate of BBP exposures during administration of medications (12.7%; P=.003). Ancillary staff experienced the highest rate of BBP exposures with an injection needle (35.5%; P=.001). There were no statistically significant differences among occupational groups for the anatomic location of injury (P=.074)(eTable).

Comment

In the year 2000, the annual global incidence of occupational BBP exposures among health care workers worldwide for hepatitis B virus, hepatitis C virus, and HIV was estimated at 2.1 million, 926,000, and 327,000, respectively. Most of these exposures were due to sharps injuries.⁴ Dermatologists are particularly at risk for BBP exposures given their reliance on frequent procedures in practice. During an 11-year period, 222 BBP exposures were documented in the dermatology departments at 3 Mayo Clinic institutions. Most exposures were due to needlestick/sharps across all occupational groups compared with splash injuries. Prior survey studies confirm that sharps injuries are frequently implicated, with 75% to 94% of residents and practicing dermatologists reporting at least 1 sharps injury.¹

Among occupational groups, resident physicians had the highest rate of BBP exposures, followed by nurse/medical assistants and practicing dermatologists, which may be secondary to lack of training or experience. Data from other surgical fields, including general surgery, support that resident physicians have the highest rate of sharps injuries.⁵ In a survey study (N=452), 51% of residents reported that extra training in safe techniques would be beneficial.² Safety training may be beneficial in reducing the incidence of BBP exposures in residency programs.

The most common implicated task in resident physicians and practicing dermatologists was suturing or assisting with suturing, and the most common implicated instrument was the suture needle. Prior studies showed conflicting data regarding common implicated tasks and instruments in this cohort.^1,2 The task of suturing and the suture needle also were the most implicated means of injury among other surgical specialties.⁶ Ancillary staff experienced most BBP exposures during handling of sharps, wires, or instruments, as well as the use of an injection needle. The designation of tasks among dermatologic staff likely explains the difference among occupational groups. This new information may provide the opportunity to improve safety measures among all members of the dermatologic team.

Limitations—There are several limitations to this study. This retrospective review was conducted at a single health system at 3 institutions. Hence, similar safety protocols likely were in place across all sites, which may reduce the generalizability of the results. In addition, there is risk of nonreporting bias among staff, as only documented incidence reports were evaluated. Prior studies demonstrated a nonreporting prevalence of 33% to 64% among dermatology staff.^1-3 We also did not evaluate whether injuries resulted in BBP exposure or transmission. The rates of postexposure prophylaxis also were not studied. This information was not available for review because of concerns for privacy. Demographic features, such as gender or years of training, also were not evaluated.

Conclusion

This study provides additional insight on the incidence of BBP exposures in dermatology, as well as the implicated tasks, instruments, and anatomic locations of injury. Studies show that implementing formal education regarding the risks of BBP exposure may result in reduction of sharps injuries.⁷ Formal education in residency programs may be needed in the field of dermatology to reduce BBP exposures. Quality improvement measures should focus on identified risk factors among occupational groups to reduce BBP exposures in the workplace.