Flipping the Wards

Millennial learners prefer novel methods of delivery that are interactive and technology based.[7, 8, 9] Lectures and slide‐based presentations frequently do not feature the degree of interactive engagement that they seek, and methods such as case‐based presentations and simulation may be more suitable. The Khan Academy is a not‐for‐profit organization that has been proposed as a model for future directions for medical education.[18] The academy's global classroom houses over 4000 videos and interactive modules to allow students to progress through topics on their own time.[19] Teaching rounds can be similarly flipped such that discussion and group work take place during rounds, whereas lectures, modules, and reading are reserved for individual study.[18]

As time pressures shift the focus of rounds exclusively toward discussion of patient‐care tasks, finding time for teaching outside of rounds can be emphasized to inspire self‐directed learning. When residents need time to tend to immediate patient‐care issues, hospitalist attendings could take the time to search for articles to send to team members. Rather than distributing paper copies that may be lost, cloud‐based data management systems such as Dropbox (Dropbox, San Francisco, CA) or Google Drive (Google Inc., Mountain View, CA) can be used to disseminate articles, which can be pulled up in real time on mobile devices during rounds and later deposited in shared folders accessible to all team members.[20, 21] The advantage of this approach is that it does not require all learners to be present on rounds, which may not be possible with duty hours.

Using Documentation to Teach

Trainees report that one of the most desirable attributes of clinical teachers is when they delineate their clinical reasoning and thought process.[22] Similarly, Millennial learners specifically desire to understand the rationale behind their teachers' actions.[6] Documentation in the medical chart or electronic health record (EHR) can be used to enhance teaching and role‐model clinical reasoning in a transparent and readily available fashion.

Billing requirements necessitate daily attending documentation in the form of an attestation. Hospitalist attendings can use attestations to model thought process and clinical synthesis in the daily assessment of a patient. For example, an attestation one‐liner can be used to concisely summarize the patient's course or highlight the most pressing issue of the day, rather than simply serve as a placeholder for billing or agree with above in reference to housestaff documentation. This practice can demonstrate to residents how to write a short snapshot of a patient's care in addition to improving communication.

Additionally, the EHR can be a useful platform to guide feedback for residents on their clinical performance. Millennial learners prefer specific, immediate feedback, and trainee documentation can serve as a template to show examples of good documentation and clinical reasoning as well as areas needing improvement.[5] These tangible examples of clinical performance are specific and understandable for trainees to guide their self‐learning and improvement.

Technology‐Enabled Teaching

Using technology wisely on the wards can improve efficiency while also taking advantage of teaching methods familiar to Millennial learners. Technology can be used in a positive manner to keep the focus on the patient and enhance teaching when time is limited on rounds. Smartphones and tablets have become an omnipresent part of the clinical environment.[23] Rather than distracting from rounds, these tools can be used to answer clinical questions in real time, thus directly linking the question to the patient's care.

The EHR is a powerful technological resource that is readily available to enhance teaching during a busy ward schedule. Clinical information is electronically accessible at all hours for both trainees and attendings, rather than only at prespecified times on daily rounds, and the Millennial Generation is accustomed to receiving and sharing information in this fashion.[24] Technology platforms that enable simultaneous sharing of information among multiple members of a team can also be used to assist in sharing clinical information in this manner. Health Insurance Portability and Accountability Act‐compliant group text‐messaging applications for smartphones and tablets such as GroupMD (GroupMD, San Francisco, CA) allow members of a team to connect through 1 portal.[25] These discussions can foster communication, inspire clinical questions, and model the practice of timely response to new information.

Using Guerilla Teaching Tactics

Though time may be limited by work hours, there are opportunities embedded into clinical practice to create teaching moments. The principle of guerilla marketing uses unconventional marketing tactics in everyday locales to aggressively promote a product.[26] Similarly, guerilla teaching might be employed on rounds to make teaching points about common patient care issues that occur at nearly every room, such as Foley catheters after seeing one at the beside or hand hygiene after leaving a room. These types of topics are familiar to trainees as well as hospitalist attendings and fulfill the relevance that Millennial learners seek by easily applying them to the patient at hand.

Memory triggers or checklists are another way to systematically introduce guerilla teaching on commonplace topics. The IBCD checklist, for example, has been successfully implemented at our institution to promote adherence to 4 quality measures.[27] IBCD, which stands for immunizations, bedsores, catheters, and deep vein thrombosis prophylaxis, is easily and quickly tacked on as a checklist item at the end of the problem list during a presentation. Similar checklists can serve as teaching points on quality and safety in inpatient care, as well as reminders to consider these issues for every patient.

Rainy Day Teaching

Hospitalist teaching attendings recognize that duty hours have shifted the preferred time for teaching away from busy admission periods such as postcall rounds.[28] The limited time spent reviewing new admissions is now often focused on patient care issues, with much of the discussion eliminated. However, hospitalist attendings can be proactive and save certain teaching moments for rainy day teaching, anticipating topics to introduce during lower census times. Additionally, attending access to the EHRs allows attendings to preview cases the residents have admitted during a call period and may facilitate planning teaching topics for future opportunities.[23]

Though teaching is an essential part of the hospitalist teaching attending role, the Millennial Generation's affinity for teamwork makes it possible to utilize additional team members as teachers for the group. This type of distribution of responsibility, or outsourcing of teaching, can be done in the form of a teaching or float resident. These individuals can be directed to search the literature to answer clinical questions the team may have during rounds and report back, which may influence decision making and patient care as well as provide education.[29]

Embedding Teaching Moments Into Rounds

Dr. Francis W. Peabody may have been addressing students many generations removed from Millennial learners when he implored them to remember that the secret of the care of the patient is in caring for the patient, but his maxim still rings true today.[30] This advice provides an important insight on how the focus can be kept on the patient by emphasizing physical examination and history‐taking skills, which engages learners in hands‐on activity and grounds that education in a patient‐based experience.[31] The Stanford 25 represents a successful project that refocuses the doctorpatient encounter on the bedside.[32] Using a Web‐based platform, this initiative instructs on 25 physical examination maneuvers, utilizing teaching methods that are familiar to Millennial learners and are patient focused.

In addition to emphasizing bedside teaching, smaller moments can be used during rounds to establish an expectation for learning. Hospitalist attendings can create a routine with daily teaching moments, such as an electrocardiogram or a daily Medical Knowledge Self‐Assessment Program question, a source of internal medicine board preparation material published by the American College of Physicians.[33] These are opportunities to inject a quick educational moment that is easily relatable to the patients on the team's service. Using teaching moments that are routine, accessible, and relevant to patient care can help shape Millennial learners' expectations that teaching be a daily occurrence interwoven within clinical care provided during rounds.

There are several limitations to our work. These strategies do not represent a systematic review, and there is little evidence to support that our approach is more effective than conventional teaching methods. Though we address hospitalists specifically, these strategies are likely suitable for all inpatient educators as they have not been well studied in specific groups. With the paucity of literature regarding learning preferences of Millennial medical trainees, it is difficult to know what methods may truly be most desirable in the wards setting, as many of the needs and learning styles considered in our approach are borrowed from other more traditional learning environments. It is unclear how adoptable our strategies may be for educators from other generations; these faculty may have different approaches to teaching. Further research is necessary to identify areas for faculty development in learning new techniques as well as compare the efficacy of our approach to conventional methods with respect to standardized educational outcomes such as In‐Training Exam performance, as well as patient outcomes.

Flipping the Wards

Millennial learners prefer novel methods of delivery that are interactive and technology based.[7, 8, 9] Lectures and slide‐based presentations frequently do not feature the degree of interactive engagement that they seek, and methods such as case‐based presentations and simulation may be more suitable. The Khan Academy is a not‐for‐profit organization that has been proposed as a model for future directions for medical education.[18] The academy's global classroom houses over 4000 videos and interactive modules to allow students to progress through topics on their own time.[19] Teaching rounds can be similarly flipped such that discussion and group work take place during rounds, whereas lectures, modules, and reading are reserved for individual study.[18]

As time pressures shift the focus of rounds exclusively toward discussion of patient‐care tasks, finding time for teaching outside of rounds can be emphasized to inspire self‐directed learning. When residents need time to tend to immediate patient‐care issues, hospitalist attendings could take the time to search for articles to send to team members. Rather than distributing paper copies that may be lost, cloud‐based data management systems such as Dropbox (Dropbox, San Francisco, CA) or Google Drive (Google Inc., Mountain View, CA) can be used to disseminate articles, which can be pulled up in real time on mobile devices during rounds and later deposited in shared folders accessible to all team members.[20, 21] The advantage of this approach is that it does not require all learners to be present on rounds, which may not be possible with duty hours.

Using Documentation to Teach

Trainees report that one of the most desirable attributes of clinical teachers is when they delineate their clinical reasoning and thought process.[22] Similarly, Millennial learners specifically desire to understand the rationale behind their teachers' actions.[6] Documentation in the medical chart or electronic health record (EHR) can be used to enhance teaching and role‐model clinical reasoning in a transparent and readily available fashion.

Billing requirements necessitate daily attending documentation in the form of an attestation. Hospitalist attendings can use attestations to model thought process and clinical synthesis in the daily assessment of a patient. For example, an attestation one‐liner can be used to concisely summarize the patient's course or highlight the most pressing issue of the day, rather than simply serve as a placeholder for billing or agree with above in reference to housestaff documentation. This practice can demonstrate to residents how to write a short snapshot of a patient's care in addition to improving communication.

Additionally, the EHR can be a useful platform to guide feedback for residents on their clinical performance. Millennial learners prefer specific, immediate feedback, and trainee documentation can serve as a template to show examples of good documentation and clinical reasoning as well as areas needing improvement.[5] These tangible examples of clinical performance are specific and understandable for trainees to guide their self‐learning and improvement.

Technology‐Enabled Teaching

Using technology wisely on the wards can improve efficiency while also taking advantage of teaching methods familiar to Millennial learners. Technology can be used in a positive manner to keep the focus on the patient and enhance teaching when time is limited on rounds. Smartphones and tablets have become an omnipresent part of the clinical environment.[23] Rather than distracting from rounds, these tools can be used to answer clinical questions in real time, thus directly linking the question to the patient's care.

The EHR is a powerful technological resource that is readily available to enhance teaching during a busy ward schedule. Clinical information is electronically accessible at all hours for both trainees and attendings, rather than only at prespecified times on daily rounds, and the Millennial Generation is accustomed to receiving and sharing information in this fashion.[24] Technology platforms that enable simultaneous sharing of information among multiple members of a team can also be used to assist in sharing clinical information in this manner. Health Insurance Portability and Accountability Act‐compliant group text‐messaging applications for smartphones and tablets such as GroupMD (GroupMD, San Francisco, CA) allow members of a team to connect through 1 portal.[25] These discussions can foster communication, inspire clinical questions, and model the practice of timely response to new information.

Using Guerilla Teaching Tactics

Though time may be limited by work hours, there are opportunities embedded into clinical practice to create teaching moments. The principle of guerilla marketing uses unconventional marketing tactics in everyday locales to aggressively promote a product.[26] Similarly, guerilla teaching might be employed on rounds to make teaching points about common patient care issues that occur at nearly every room, such as Foley catheters after seeing one at the beside or hand hygiene after leaving a room. These types of topics are familiar to trainees as well as hospitalist attendings and fulfill the relevance that Millennial learners seek by easily applying them to the patient at hand.

Memory triggers or checklists are another way to systematically introduce guerilla teaching on commonplace topics. The IBCD checklist, for example, has been successfully implemented at our institution to promote adherence to 4 quality measures.[27] IBCD, which stands for immunizations, bedsores, catheters, and deep vein thrombosis prophylaxis, is easily and quickly tacked on as a checklist item at the end of the problem list during a presentation. Similar checklists can serve as teaching points on quality and safety in inpatient care, as well as reminders to consider these issues for every patient.

Rainy Day Teaching

Hospitalist teaching attendings recognize that duty hours have shifted the preferred time for teaching away from busy admission periods such as postcall rounds.[28] The limited time spent reviewing new admissions is now often focused on patient care issues, with much of the discussion eliminated. However, hospitalist attendings can be proactive and save certain teaching moments for rainy day teaching, anticipating topics to introduce during lower census times. Additionally, attending access to the EHRs allows attendings to preview cases the residents have admitted during a call period and may facilitate planning teaching topics for future opportunities.[23]

Though teaching is an essential part of the hospitalist teaching attending role, the Millennial Generation's affinity for teamwork makes it possible to utilize additional team members as teachers for the group. This type of distribution of responsibility, or outsourcing of teaching, can be done in the form of a teaching or float resident. These individuals can be directed to search the literature to answer clinical questions the team may have during rounds and report back, which may influence decision making and patient care as well as provide education.[29]

Embedding Teaching Moments Into Rounds

Dr. Francis W. Peabody may have been addressing students many generations removed from Millennial learners when he implored them to remember that the secret of the care of the patient is in caring for the patient, but his maxim still rings true today.[30] This advice provides an important insight on how the focus can be kept on the patient by emphasizing physical examination and history‐taking skills, which engages learners in hands‐on activity and grounds that education in a patient‐based experience.[31] The Stanford 25 represents a successful project that refocuses the doctorpatient encounter on the bedside.[32] Using a Web‐based platform, this initiative instructs on 25 physical examination maneuvers, utilizing teaching methods that are familiar to Millennial learners and are patient focused.

In addition to emphasizing bedside teaching, smaller moments can be used during rounds to establish an expectation for learning. Hospitalist attendings can create a routine with daily teaching moments, such as an electrocardiogram or a daily Medical Knowledge Self‐Assessment Program question, a source of internal medicine board preparation material published by the American College of Physicians.[33] These are opportunities to inject a quick educational moment that is easily relatable to the patients on the team's service. Using teaching moments that are routine, accessible, and relevant to patient care can help shape Millennial learners' expectations that teaching be a daily occurrence interwoven within clinical care provided during rounds.

There are several limitations to our work. These strategies do not represent a systematic review, and there is little evidence to support that our approach is more effective than conventional teaching methods. Though we address hospitalists specifically, these strategies are likely suitable for all inpatient educators as they have not been well studied in specific groups. With the paucity of literature regarding learning preferences of Millennial medical trainees, it is difficult to know what methods may truly be most desirable in the wards setting, as many of the needs and learning styles considered in our approach are borrowed from other more traditional learning environments. It is unclear how adoptable our strategies may be for educators from other generations; these faculty may have different approaches to teaching. Further research is necessary to identify areas for faculty development in learning new techniques as well as compare the efficacy of our approach to conventional methods with respect to standardized educational outcomes such as In‐Training Exam performance, as well as patient outcomes.

Flipping the Wards

Millennial learners prefer novel methods of delivery that are interactive and technology based.[7, 8, 9] Lectures and slide‐based presentations frequently do not feature the degree of interactive engagement that they seek, and methods such as case‐based presentations and simulation may be more suitable. The Khan Academy is a not‐for‐profit organization that has been proposed as a model for future directions for medical education.[18] The academy's global classroom houses over 4000 videos and interactive modules to allow students to progress through topics on their own time.[19] Teaching rounds can be similarly flipped such that discussion and group work take place during rounds, whereas lectures, modules, and reading are reserved for individual study.[18]

As time pressures shift the focus of rounds exclusively toward discussion of patient‐care tasks, finding time for teaching outside of rounds can be emphasized to inspire self‐directed learning. When residents need time to tend to immediate patient‐care issues, hospitalist attendings could take the time to search for articles to send to team members. Rather than distributing paper copies that may be lost, cloud‐based data management systems such as Dropbox (Dropbox, San Francisco, CA) or Google Drive (Google Inc., Mountain View, CA) can be used to disseminate articles, which can be pulled up in real time on mobile devices during rounds and later deposited in shared folders accessible to all team members.[20, 21] The advantage of this approach is that it does not require all learners to be present on rounds, which may not be possible with duty hours.

Using Documentation to Teach

Trainees report that one of the most desirable attributes of clinical teachers is when they delineate their clinical reasoning and thought process.[22] Similarly, Millennial learners specifically desire to understand the rationale behind their teachers' actions.[6] Documentation in the medical chart or electronic health record (EHR) can be used to enhance teaching and role‐model clinical reasoning in a transparent and readily available fashion.

Billing requirements necessitate daily attending documentation in the form of an attestation. Hospitalist attendings can use attestations to model thought process and clinical synthesis in the daily assessment of a patient. For example, an attestation one‐liner can be used to concisely summarize the patient's course or highlight the most pressing issue of the day, rather than simply serve as a placeholder for billing or agree with above in reference to housestaff documentation. This practice can demonstrate to residents how to write a short snapshot of a patient's care in addition to improving communication.

Additionally, the EHR can be a useful platform to guide feedback for residents on their clinical performance. Millennial learners prefer specific, immediate feedback, and trainee documentation can serve as a template to show examples of good documentation and clinical reasoning as well as areas needing improvement.[5] These tangible examples of clinical performance are specific and understandable for trainees to guide their self‐learning and improvement.

Technology‐Enabled Teaching

Using technology wisely on the wards can improve efficiency while also taking advantage of teaching methods familiar to Millennial learners. Technology can be used in a positive manner to keep the focus on the patient and enhance teaching when time is limited on rounds. Smartphones and tablets have become an omnipresent part of the clinical environment.[23] Rather than distracting from rounds, these tools can be used to answer clinical questions in real time, thus directly linking the question to the patient's care.

The EHR is a powerful technological resource that is readily available to enhance teaching during a busy ward schedule. Clinical information is electronically accessible at all hours for both trainees and attendings, rather than only at prespecified times on daily rounds, and the Millennial Generation is accustomed to receiving and sharing information in this fashion.[24] Technology platforms that enable simultaneous sharing of information among multiple members of a team can also be used to assist in sharing clinical information in this manner. Health Insurance Portability and Accountability Act‐compliant group text‐messaging applications for smartphones and tablets such as GroupMD (GroupMD, San Francisco, CA) allow members of a team to connect through 1 portal.[25] These discussions can foster communication, inspire clinical questions, and model the practice of timely response to new information.

Using Guerilla Teaching Tactics

Though time may be limited by work hours, there are opportunities embedded into clinical practice to create teaching moments. The principle of guerilla marketing uses unconventional marketing tactics in everyday locales to aggressively promote a product.[26] Similarly, guerilla teaching might be employed on rounds to make teaching points about common patient care issues that occur at nearly every room, such as Foley catheters after seeing one at the beside or hand hygiene after leaving a room. These types of topics are familiar to trainees as well as hospitalist attendings and fulfill the relevance that Millennial learners seek by easily applying them to the patient at hand.

Memory triggers or checklists are another way to systematically introduce guerilla teaching on commonplace topics. The IBCD checklist, for example, has been successfully implemented at our institution to promote adherence to 4 quality measures.[27] IBCD, which stands for immunizations, bedsores, catheters, and deep vein thrombosis prophylaxis, is easily and quickly tacked on as a checklist item at the end of the problem list during a presentation. Similar checklists can serve as teaching points on quality and safety in inpatient care, as well as reminders to consider these issues for every patient.

Rainy Day Teaching

Hospitalist teaching attendings recognize that duty hours have shifted the preferred time for teaching away from busy admission periods such as postcall rounds.[28] The limited time spent reviewing new admissions is now often focused on patient care issues, with much of the discussion eliminated. However, hospitalist attendings can be proactive and save certain teaching moments for rainy day teaching, anticipating topics to introduce during lower census times. Additionally, attending access to the EHRs allows attendings to preview cases the residents have admitted during a call period and may facilitate planning teaching topics for future opportunities.[23]

Though teaching is an essential part of the hospitalist teaching attending role, the Millennial Generation's affinity for teamwork makes it possible to utilize additional team members as teachers for the group. This type of distribution of responsibility, or outsourcing of teaching, can be done in the form of a teaching or float resident. These individuals can be directed to search the literature to answer clinical questions the team may have during rounds and report back, which may influence decision making and patient care as well as provide education.[29]

Embedding Teaching Moments Into Rounds

Dr. Francis W. Peabody may have been addressing students many generations removed from Millennial learners when he implored them to remember that the secret of the care of the patient is in caring for the patient, but his maxim still rings true today.[30] This advice provides an important insight on how the focus can be kept on the patient by emphasizing physical examination and history‐taking skills, which engages learners in hands‐on activity and grounds that education in a patient‐based experience.[31] The Stanford 25 represents a successful project that refocuses the doctorpatient encounter on the bedside.[32] Using a Web‐based platform, this initiative instructs on 25 physical examination maneuvers, utilizing teaching methods that are familiar to Millennial learners and are patient focused.

In addition to emphasizing bedside teaching, smaller moments can be used during rounds to establish an expectation for learning. Hospitalist attendings can create a routine with daily teaching moments, such as an electrocardiogram or a daily Medical Knowledge Self‐Assessment Program question, a source of internal medicine board preparation material published by the American College of Physicians.[33] These are opportunities to inject a quick educational moment that is easily relatable to the patients on the team's service. Using teaching moments that are routine, accessible, and relevant to patient care can help shape Millennial learners' expectations that teaching be a daily occurrence interwoven within clinical care provided during rounds.

There are several limitations to our work. These strategies do not represent a systematic review, and there is little evidence to support that our approach is more effective than conventional teaching methods. Though we address hospitalists specifically, these strategies are likely suitable for all inpatient educators as they have not been well studied in specific groups. With the paucity of literature regarding learning preferences of Millennial medical trainees, it is difficult to know what methods may truly be most desirable in the wards setting, as many of the needs and learning styles considered in our approach are borrowed from other more traditional learning environments. It is unclear how adoptable our strategies may be for educators from other generations; these faculty may have different approaches to teaching. Further research is necessary to identify areas for faculty development in learning new techniques as well as compare the efficacy of our approach to conventional methods with respect to standardized educational outcomes such as In‐Training Exam performance, as well as patient outcomes.

Ethics Statement

All data were deidentified, by Premier, Inc., at both the hospital and patient level in accordance with the Health Insurance Portability and Accountability Act. The Yale University Human Investigation Committee reviewed the protocol for this study and determined that it is not considered to be human subjects research as defined by the Office of Human Research Protections.

Data Source

We conducted a cross‐sectional study using data from hospitals that participated in the database maintained by Premier Healthcare Informatics (Charlotte, NC) in the years 2009 to 2010. The Premier database is a voluntary, fee‐supported database created to measure quality and healthcare utilization.[3, 16, 17, 18] In 2010, it included detailed billing data from 500 hospitals in the United States, with more than 130 million cumulative hospital discharges. The detailed billing data includes all elements found in hospital claims derived from the uniform billing‐04 form, as well as an itemized, date‐stamped log of all items and services charged to the patient or insurer, such as medications, laboratory tests, and diagnostic and therapeutic services. The database includes approximately 15% of all US hospitalizations. Participating hospitals are similar to the composition of acute care hospitals nationwide. They represent all regions of the United States, and represent predominantly small‐ to mid‐sized nonteaching facilities that serve a largely urban population. The database also contains hospital reported costs at the item level as well as the total cost of the hospitalization. Approximately 75% of hospitals that participate submit RVU‐based costs taken from internal cost accounting systems. Because of our focus on comparing standardized costs to reported costs, we included only data from hospitals that use RVU‐based costs in this study.

Study Subjects

We included adult patients with a hospitalization recorded in the Premier database between January 1, 2009 and December 31, 2010, and a principal discharge diagnosis of heart failure (HF) (International Classification of Diseases, Ninth Revision, Clinical Modification codes: 402.01, 402.11, 402.91, 404.01, 404.03, 404.11, 404.13, 404.91, 404.93, 428.xx). We excluded transfers, patients assigned a pediatrician as the attending of record, and those who received a heart transplant or ventricular assist device during their stay. Because cost data are prone to extreme outliers, we excluded hospitalizations that were in the top 0.1% of length of stay, number of billing records, quantity of items billed, or total standardized cost. We also excluded hospitals that admitted fewer than 25 HF patients during the study period to reduce the possibility that a single high‐cost patient affected the hospital's cost profile.

Hospital Information

For each hospital included in the study, we recorded number of beds, teaching status, geographic region, and whether it served an urban or rural population.

Assignment of Standardized Costs

We defined reported cost as the RVU‐based cost per item in the database. We then calculated the median across hospitals for each item in the database and set this as the standardized unit cost of that item at every hospital (Figure 1). Once standardized costs were assigned at the item level, we summed the costs of all items assigned to each patient and calculated the standardized cost of a hospitalization per patient at each hospital.

Figure 1

Examination of Cost Variation

We compared the standardized and reported costs of hospitalizations using medians, interquartile ranges, and interquartile ratios (Q75/Q25). To examine whether standardized costs can reduce the noise due to differences in overhead and other fixed costs, we calculated, for each hospital, the coefficients of variation (CV) for per‐day reported and standardized costs and per‐hospitalization reported and standardized costs. We used the Fligner‐Killeen test to determine whether the variance of CVs was different for reported and standardized costs.[19]

Creation of Basket of Goods

Because there can be differences in the costs of items, the number and types of items administered during hospitalizations, 2 hospitals with similar reported costs for a hospitalization might deliver different quantities and combinations of treatments (Figure 1). We wished to demonstrate that there is variation in reported costs of items when the quantity and type of item is held constant, so we created a basket of items. We chose items that are commonly administered to patients with heart failure, but could have chosen any combination of items. The basket included a day of medical room and board, a day of intensive care unit (ICU) room and board, a single dose of ‐blocker, a single dose of angiotensin‐converting enzyme inhibitor, complete blood count, a B‐natriuretic peptide level, a chest radiograph, a chest computed tomography, and an echocardiogram. We then examined the range of hospitals' reported costs for this basket of goods using percentiles, medians, and interquartile ranges.

Reported to Standardized Cost Ratio

Next, we calculated standardized costs of hospitalizations for included hospitals and examined the relationship between hospitals' mean reported costs and mean standardized costs. This ratio could help diagnose the mechanism of high reported costs for a hospital, because high reported costs with low utilization would indicate high fixed costs, while high reported costs with high utilization would indicate greater use of tests and treatments. We assigned hospitals to strata based on reported costs greater than standardized costs by more than 25%, reported costs within 25% of standardized costs, and reported costs less than standardized costs by more than 25%. We examined the association between hospital characteristics and strata using a ² test. All analyses were carried out using SAS version 9.3 (SAS Institute Inc., Cary, NC).

Ethics Statement

All data were deidentified, by Premier, Inc., at both the hospital and patient level in accordance with the Health Insurance Portability and Accountability Act. The Yale University Human Investigation Committee reviewed the protocol for this study and determined that it is not considered to be human subjects research as defined by the Office of Human Research Protections.

Data Source

We conducted a cross‐sectional study using data from hospitals that participated in the database maintained by Premier Healthcare Informatics (Charlotte, NC) in the years 2009 to 2010. The Premier database is a voluntary, fee‐supported database created to measure quality and healthcare utilization.[3, 16, 17, 18] In 2010, it included detailed billing data from 500 hospitals in the United States, with more than 130 million cumulative hospital discharges. The detailed billing data includes all elements found in hospital claims derived from the uniform billing‐04 form, as well as an itemized, date‐stamped log of all items and services charged to the patient or insurer, such as medications, laboratory tests, and diagnostic and therapeutic services. The database includes approximately 15% of all US hospitalizations. Participating hospitals are similar to the composition of acute care hospitals nationwide. They represent all regions of the United States, and represent predominantly small‐ to mid‐sized nonteaching facilities that serve a largely urban population. The database also contains hospital reported costs at the item level as well as the total cost of the hospitalization. Approximately 75% of hospitals that participate submit RVU‐based costs taken from internal cost accounting systems. Because of our focus on comparing standardized costs to reported costs, we included only data from hospitals that use RVU‐based costs in this study.

Study Subjects

We included adult patients with a hospitalization recorded in the Premier database between January 1, 2009 and December 31, 2010, and a principal discharge diagnosis of heart failure (HF) (International Classification of Diseases, Ninth Revision, Clinical Modification codes: 402.01, 402.11, 402.91, 404.01, 404.03, 404.11, 404.13, 404.91, 404.93, 428.xx). We excluded transfers, patients assigned a pediatrician as the attending of record, and those who received a heart transplant or ventricular assist device during their stay. Because cost data are prone to extreme outliers, we excluded hospitalizations that were in the top 0.1% of length of stay, number of billing records, quantity of items billed, or total standardized cost. We also excluded hospitals that admitted fewer than 25 HF patients during the study period to reduce the possibility that a single high‐cost patient affected the hospital's cost profile.

Hospital Information

For each hospital included in the study, we recorded number of beds, teaching status, geographic region, and whether it served an urban or rural population.

Assignment of Standardized Costs

We defined reported cost as the RVU‐based cost per item in the database. We then calculated the median across hospitals for each item in the database and set this as the standardized unit cost of that item at every hospital (Figure 1). Once standardized costs were assigned at the item level, we summed the costs of all items assigned to each patient and calculated the standardized cost of a hospitalization per patient at each hospital.

Figure 1

Examination of Cost Variation

We compared the standardized and reported costs of hospitalizations using medians, interquartile ranges, and interquartile ratios (Q75/Q25). To examine whether standardized costs can reduce the noise due to differences in overhead and other fixed costs, we calculated, for each hospital, the coefficients of variation (CV) for per‐day reported and standardized costs and per‐hospitalization reported and standardized costs. We used the Fligner‐Killeen test to determine whether the variance of CVs was different for reported and standardized costs.[19]

Creation of Basket of Goods

Because there can be differences in the costs of items, the number and types of items administered during hospitalizations, 2 hospitals with similar reported costs for a hospitalization might deliver different quantities and combinations of treatments (Figure 1). We wished to demonstrate that there is variation in reported costs of items when the quantity and type of item is held constant, so we created a basket of items. We chose items that are commonly administered to patients with heart failure, but could have chosen any combination of items. The basket included a day of medical room and board, a day of intensive care unit (ICU) room and board, a single dose of ‐blocker, a single dose of angiotensin‐converting enzyme inhibitor, complete blood count, a B‐natriuretic peptide level, a chest radiograph, a chest computed tomography, and an echocardiogram. We then examined the range of hospitals' reported costs for this basket of goods using percentiles, medians, and interquartile ranges.

Reported to Standardized Cost Ratio

Next, we calculated standardized costs of hospitalizations for included hospitals and examined the relationship between hospitals' mean reported costs and mean standardized costs. This ratio could help diagnose the mechanism of high reported costs for a hospital, because high reported costs with low utilization would indicate high fixed costs, while high reported costs with high utilization would indicate greater use of tests and treatments. We assigned hospitals to strata based on reported costs greater than standardized costs by more than 25%, reported costs within 25% of standardized costs, and reported costs less than standardized costs by more than 25%. We examined the association between hospital characteristics and strata using a ² test. All analyses were carried out using SAS version 9.3 (SAS Institute Inc., Cary, NC).

Ethics Statement

All data were deidentified, by Premier, Inc., at both the hospital and patient level in accordance with the Health Insurance Portability and Accountability Act. The Yale University Human Investigation Committee reviewed the protocol for this study and determined that it is not considered to be human subjects research as defined by the Office of Human Research Protections.

Data Source

We conducted a cross‐sectional study using data from hospitals that participated in the database maintained by Premier Healthcare Informatics (Charlotte, NC) in the years 2009 to 2010. The Premier database is a voluntary, fee‐supported database created to measure quality and healthcare utilization.[3, 16, 17, 18] In 2010, it included detailed billing data from 500 hospitals in the United States, with more than 130 million cumulative hospital discharges. The detailed billing data includes all elements found in hospital claims derived from the uniform billing‐04 form, as well as an itemized, date‐stamped log of all items and services charged to the patient or insurer, such as medications, laboratory tests, and diagnostic and therapeutic services. The database includes approximately 15% of all US hospitalizations. Participating hospitals are similar to the composition of acute care hospitals nationwide. They represent all regions of the United States, and represent predominantly small‐ to mid‐sized nonteaching facilities that serve a largely urban population. The database also contains hospital reported costs at the item level as well as the total cost of the hospitalization. Approximately 75% of hospitals that participate submit RVU‐based costs taken from internal cost accounting systems. Because of our focus on comparing standardized costs to reported costs, we included only data from hospitals that use RVU‐based costs in this study.

Study Subjects

We included adult patients with a hospitalization recorded in the Premier database between January 1, 2009 and December 31, 2010, and a principal discharge diagnosis of heart failure (HF) (International Classification of Diseases, Ninth Revision, Clinical Modification codes: 402.01, 402.11, 402.91, 404.01, 404.03, 404.11, 404.13, 404.91, 404.93, 428.xx). We excluded transfers, patients assigned a pediatrician as the attending of record, and those who received a heart transplant or ventricular assist device during their stay. Because cost data are prone to extreme outliers, we excluded hospitalizations that were in the top 0.1% of length of stay, number of billing records, quantity of items billed, or total standardized cost. We also excluded hospitals that admitted fewer than 25 HF patients during the study period to reduce the possibility that a single high‐cost patient affected the hospital's cost profile.

Hospital Information

For each hospital included in the study, we recorded number of beds, teaching status, geographic region, and whether it served an urban or rural population.

Assignment of Standardized Costs

We defined reported cost as the RVU‐based cost per item in the database. We then calculated the median across hospitals for each item in the database and set this as the standardized unit cost of that item at every hospital (Figure 1). Once standardized costs were assigned at the item level, we summed the costs of all items assigned to each patient and calculated the standardized cost of a hospitalization per patient at each hospital.

Figure 1

Examination of Cost Variation

We compared the standardized and reported costs of hospitalizations using medians, interquartile ranges, and interquartile ratios (Q75/Q25). To examine whether standardized costs can reduce the noise due to differences in overhead and other fixed costs, we calculated, for each hospital, the coefficients of variation (CV) for per‐day reported and standardized costs and per‐hospitalization reported and standardized costs. We used the Fligner‐Killeen test to determine whether the variance of CVs was different for reported and standardized costs.[19]

Creation of Basket of Goods

Because there can be differences in the costs of items, the number and types of items administered during hospitalizations, 2 hospitals with similar reported costs for a hospitalization might deliver different quantities and combinations of treatments (Figure 1). We wished to demonstrate that there is variation in reported costs of items when the quantity and type of item is held constant, so we created a basket of items. We chose items that are commonly administered to patients with heart failure, but could have chosen any combination of items. The basket included a day of medical room and board, a day of intensive care unit (ICU) room and board, a single dose of ‐blocker, a single dose of angiotensin‐converting enzyme inhibitor, complete blood count, a B‐natriuretic peptide level, a chest radiograph, a chest computed tomography, and an echocardiogram. We then examined the range of hospitals' reported costs for this basket of goods using percentiles, medians, and interquartile ranges.

Reported to Standardized Cost Ratio

Next, we calculated standardized costs of hospitalizations for included hospitals and examined the relationship between hospitals' mean reported costs and mean standardized costs. This ratio could help diagnose the mechanism of high reported costs for a hospital, because high reported costs with low utilization would indicate high fixed costs, while high reported costs with high utilization would indicate greater use of tests and treatments. We assigned hospitals to strata based on reported costs greater than standardized costs by more than 25%, reported costs within 25% of standardized costs, and reported costs less than standardized costs by more than 25%. We examined the association between hospital characteristics and strata using a ² test. All analyses were carried out using SAS version 9.3 (SAS Institute Inc., Cary, NC).

Search Strategy

We searched MEDLINE, EMBASE, and SCOPUS through August 2011, CINAHL, International Pharmaceutical Abstracts, clinicaltrial.gov, and the Cochrane Library through August 2012. We developed a search strategy based on medical subject headings (MeSH) terms and text words of key articles that we identified a priori[9] (see the Appendix for search strategy details).

Study Selection

We reviewed titles followed by abstracts to identify randomized controlled trials (RCTs) or observational studies with comparison groups reporting on the effectiveness or safety of VTE prevention in our populations. Two investigators independently reviewed abstracts, and we excluded the abstracts if both investigators agreed that the article met 1 or more of the exclusion criteria. We included only English‐language articles that evaluated the effectiveness of pharmacological or mechanical interventions that have been approved for clinical use in the United States. To be eligible, the studies must have addressed relevant key questions in the population of our interest. We resolved disagreements by consensus. We used DistillerSR (Evidence Partners Inc., Ottawa, Ontario, Canada), a Web‐based database management program to manage the review process. Two investigators assessed the risk of bias in each study independently, using the Downs and Black instrument for observational studies and trials.[10]

Data Synthesis

For each select population, we created detailed evidence tables containing the information abstracted from the eligible studies. After synthesizing the evidence, we graded the quantity, quality, and consistency of the best available evidence for each select population by adapting an evidence‐grading scheme recommended in the Methods Guide for Conducting Comparative Effectiveness Reviews.[7]

Obese Patients

We found 1 subgroup analysis of an RCT (total 3706 patients, 2563 nonobese and 1118 obese patients) that reported on the comparative effectiveness and safety of fixed low‐dose dalteparin 5000 IU/day compared to placebo among 1118 hospitalized medically ill patients with body mass indices (BMI) greater than 30 kg/m².¹¹ Neither group received additional concurrent prophylactic therapies. The 3 most prevalent medical diagnoses prompting hospitalization were congestive heart failure, respiratory failure, and infectious diseases. Compression ultrasound was performed in all patients by day 21 of hospitalization. The primary end point was the composite of VTE, fatal PE, and sudden death, and secondary end points included DVT, bleeding, and thrombocytopenia by day 21 (Table 1). In obese patients, the primary end point occurred in 2.8% (95% confidence interval [CI]: 1.34.3) of the dalteparin group and in 4.3% (95% CI: 2.56.2) of the placebo group (relative risk [RR]: 0.64; 95% CI: 0.32‐1.28). In nonobese patients, the primary end point occurred in 2.8% (95% CI: 1.8‐3.8) and 5.2% (95% CI: 3.9‐6.6) of the dalteparin and placebo groups, respectively (RR: 0.53; 95% CI: 0.34‐0.82). When weight was modeled as a continuous variable, no statistically significant interaction between weight and dalteparin efficacy was observed (P=0.97). The authors calculated the RR in predefined BMI subgroups and found that dalteparin was effective in reducing VTE in patients with BMIs up to 40, with RRs of <1.0 for all (approximate range, 0.20.8). However, a fixed dose of dalteparin 5000 IU/day was not better than placebo for individuals with BMI >40 kg/m². There was no significant difference in mortality or major hemorrhage by day 21 between treatment and placebo groups.

Freeman and colleagues prospectively assigned 31 medically ill patients with extreme obesity (BMI >40 kg/m²) to 1 of 3 dosing regimens of enoxaparin: a fixed dose of 40 mg daily enoxaparin (control group, n=11), enoxaparin at 0.4 mg/kg (n=9), or enoxaparin at 0.5 mg/kg (n=11).[12] The average BMI of the entire cohort was 62.1 kg/m² (range, 40.582.4). All patients had anti‐factor Xa levels drawn on the day of enrollment and daily for 3 days (Table 2). The relationship between anti‐factor Xa levels and clinical efficacy of low‐molecular weight heparin (LMWH) in VTE prophylaxis is still unclear; however, an anti‐factor Xa level of 0.2 to 0.5 IU/mL, measured 4 hours after the fourth dose of LMWH, is the target level recommended for VTE prophylaxis.[13] Patients who received weight‐based enoxaparin at 0.5mg/kg achieved target anti‐factor Xa level 86% of the time compared to 32% of the time in those receiving 0.4 mg/kg and 19% of the time for those in the fixed‐dose group (P<0.001). No clinical outcomes were reported in this study.

Patients on Antiplatelet Drugs

We did not find studies that directly looked at the comparative effectiveness of VTE prophylaxis in patients who were on antiplatelet drugs including aspirin. However, there were 2 studies that looked at the risk of bleeding in patients who received VTE pharmacologic prophylaxis while concurrently taking antiplatelet agents including aspirin. Both studies used pooled data from large phase III trials.

The study by Eriksson et al. used data from the RECORD (Regulation of Coagulation in Orthopedic Surgery to Prevent Deep Venous Thrombosis and Pulmonary Embolism) trial where over 12,000 patients undergoing elective total knee or hip replacement were randomized to receive VTE prophylaxis with oral rivaroxaban or subcutaneous enoxaparin.[14] Nine percent of participants in each arm (563 in rivaroxaban and 526 in enoxaparin/placebo) were concomitantly using antiplatelet agents or aspirin at least once during the at risk period, defined as starting at day 1 of surgery up to 2 days after the last intake of the study drug. The only end point evaluated was bleeding, and the authors found no statistically significant bleeding difference among the 2 arms (Table 1). Any bleeding event in the rivaroxaban with antiplatelets or aspirin arm was found in 20 (3.6%) patients, whereas in those on enoxaparin/placebo with antiplatelets or aspirin arm it was 17 (3.2%). The relative rate of bleeding among users versus nonusers of antiplatelet drugs or aspirin was 1.32 (95% CI: 0.85‐2.05) in the rivaroxaban group and 1.40 (95% CI: 0.87‐2.25) in the enoxaparin arm (Table 1).

Friedman et al. used pooled data from the RE‐MODEL, RENOVATE, and REMOBILIZE trials, where patients who were undergoing hip or knee arthroplasty were randomized to 220 mg of dabigatran once daily, 150 mg of dabigatran once daily (we focused on this lower dosage as this is the only available dose used in the US), 40 mg of enoxaparin once daily, or 30 mg of enoxaparin twice a day.[15] Of the 8135 patients, 4.7% were on concomitant aspirin. The baseline characteristics of those on aspirin were similar to the other enrollees. The primary outcome was major bleeding events requiring transfusion, symptomatic internal bleeding, or bleeding requiring surgery. Among patients receiving 150 mg of dabigatran, bleeding events with and without concomitant aspirin occurred in 1.6% and 1.0%, respectively (odds ratio [OR]: 1.64; 95% CI: 0.36‐7.49; P=0.523). The percentages of participants with bleeding who received enoxaparin, with and without aspirin, were 3.0% and 1.2%, respectively (OR: 2.57; 95% CI: 0.83‐7.94; P=0.101). The RR of bleeding on dabigatran compared to enoxaparin with and without aspirin therapy was 0.55 (95% CI: 0.11‐2.78) and 0.82 (95% CI: 0.37‐1.84), respectively (Table 1).

Patients With Renal Insufficiency

We found 5 studies that evaluated the comparative effectiveness and safety of pharmacologic prophylaxis for prevention of VTE in patients with acute kidney injury, moderate renal insufficiency, severe renal insufficiency not undergoing dialysis, or patients receiving dialysis. Four studies were RCTs,[16, 17, 18, 19] and 1 used a cohort design assessing separate cohorts before and after a quality improvement intervention.[20] Bauersachs and colleagues conducted an RCT comparing unfractionated heparin at 5000 IU, 3 times daily to certoparin, which is not approved in the United States and is not further discussed here.[16] The rate of DVT among patients treated with unfractionated heparin in patients with a glomerular filtration rate >30 mL/min was marginally lower than those with severe renal dysfunction (10.3 vs 11.1%) (Table 1).

Patients with severe renal dysfunction who received 5000 IU of unfractionated heparin 3 times a day were at increased risk of all bleeds (RR: 3.4; 95% CI: 2.05.9), major bleeds (RR: 7.3; 95% CI: 3.316), and minor bleeds (RR: 2.6; 95% CI: 1.4‐4.9) compared to patients treated with unfractionated heparin without severe renal dysfunction.[16]

A randomized trial by Mah and colleagues compared drug accumulation and anti‐Xa activity in elderly patients with renal dysfunction (defined as a glomerular filtration rate of 20 to 50 mL/min) who received either tinzaparin at 4500 IU once daily or enoxaparin at 4000 IU once daily.[17] Enoxaparin accumulated to a greater extent from day 1 to day 8 than did tinzaparin; the ratio of maximum concentration on day 8 compared to day 1 was 1.22 for enoxaparin and 1.05 for tinzaparin (P=0.016). No VTE events were reported in patients who received tinzaparin or enoxaparin. There was no statistical difference in the incidence of bleeding events between patients receiving tinzaparin (5, including 2 major events) and enoxaparin (4, including 3 major events, P=0.67) (Table 1).

The trial by Dahl and colleagues randomly assigned patients who were over 75 years of age and/or who had moderate renal dysfunction (defined as creatinine clearance between 30 and 49 mL/min) to receive enoxaparin 40 mg daily or dabigatran 150 mg daily.[18] There was no significant difference in the rate of major VTE events between patients receiving dabigatran (4.3%) and enoxaparin (9%) (OR: 0.48; 95% CI: 0.13‐1.73; P=0.271) (Table 1). The rate of major bleeding was significantly higher among patients randomly assigned to receive enoxaparin (4.7%) versus dabigatran (0%) (P=0.039).[18]

Shorr and colleagues published a post hoc subgroup analysis of a multicenter trial in which orthopedic patients were randomly assigned to receive desirudin 15 mg twice daily or enoxaparin 40 mg once daily.[19] Evaluable patients (1565 of the 2079 patients randomized in the trial) receiving desirudin experienced a significantly lower rate of major VTE compared with patients receiving enoxaparin (4.9% vs 7.6%, P=0.019). This relationship was particularly pronounced for evaluable patients whose creatinine clearance was between 30 and 44 mL/min. In evaluable patients with this degree of renal dysfunction, 11% of patients taking enoxaparin compared to 3.4% of those taking desirudin had a major VTE (OR: 3.52; 95% CI: 1.48‐8.4; P=0.004). There was no significant difference in the rates of major bleeding among a subset of patients assessed for safety outcomes (2078 of the 2079 patients randomized in the trial) who received desirudin (0.8%) or enoxaparin (0.2%) (Table 1).

Elsaid and Collins assessed VTE and bleeding events associated with the use of unfractionated heparin 5000U either 2 or 3 times daily and enoxaparin 30 mg once or twice daily across patients stratified by renal function (creatinine clearance <30, 3059, and 60 mL/min). The investigators made assessments before and after a quality improvement intervention that was designed to eliminate the use of enoxaparin in patients whose creatinine clearance was <30 mL/min. No VTE events were reported. Patients receiving enoxaparin were significantly more likely to experience a major bleeding episode compared with patients receiving unfractionated heparin (overall rates for all levels of renal function: 13.5% vs 4. 2%; RR: 3.2; 95% CI: 1.47.3) (Table 2). This association was largely driven by the subgroup of patients with a creatinine clearance <30 mL/min. For this subgroup with severe renal insufficiency, patients receiving enoxaparin were significantly more likely to have a bleed compared with patients receiving unfractionated heparin (18.9% vs 4.1%; RR: 4.68; 95% CI: 1.120.6) (Tables 1 and 2). There was no difference in the bleeding rates for patients whose creatinine clearances were >60 mL/min.[20]

Strength of Evidence

Search Strategy

We searched MEDLINE, EMBASE, and SCOPUS through August 2011, CINAHL, International Pharmaceutical Abstracts, clinicaltrial.gov, and the Cochrane Library through August 2012. We developed a search strategy based on medical subject headings (MeSH) terms and text words of key articles that we identified a priori[9] (see the Appendix for search strategy details).

Study Selection

We reviewed titles followed by abstracts to identify randomized controlled trials (RCTs) or observational studies with comparison groups reporting on the effectiveness or safety of VTE prevention in our populations. Two investigators independently reviewed abstracts, and we excluded the abstracts if both investigators agreed that the article met 1 or more of the exclusion criteria. We included only English‐language articles that evaluated the effectiveness of pharmacological or mechanical interventions that have been approved for clinical use in the United States. To be eligible, the studies must have addressed relevant key questions in the population of our interest. We resolved disagreements by consensus. We used DistillerSR (Evidence Partners Inc., Ottawa, Ontario, Canada), a Web‐based database management program to manage the review process. Two investigators assessed the risk of bias in each study independently, using the Downs and Black instrument for observational studies and trials.[10]

Data Synthesis

For each select population, we created detailed evidence tables containing the information abstracted from the eligible studies. After synthesizing the evidence, we graded the quantity, quality, and consistency of the best available evidence for each select population by adapting an evidence‐grading scheme recommended in the Methods Guide for Conducting Comparative Effectiveness Reviews.[7]

Obese Patients

We found 1 subgroup analysis of an RCT (total 3706 patients, 2563 nonobese and 1118 obese patients) that reported on the comparative effectiveness and safety of fixed low‐dose dalteparin 5000 IU/day compared to placebo among 1118 hospitalized medically ill patients with body mass indices (BMI) greater than 30 kg/m².¹¹ Neither group received additional concurrent prophylactic therapies. The 3 most prevalent medical diagnoses prompting hospitalization were congestive heart failure, respiratory failure, and infectious diseases. Compression ultrasound was performed in all patients by day 21 of hospitalization. The primary end point was the composite of VTE, fatal PE, and sudden death, and secondary end points included DVT, bleeding, and thrombocytopenia by day 21 (Table 1). In obese patients, the primary end point occurred in 2.8% (95% confidence interval [CI]: 1.34.3) of the dalteparin group and in 4.3% (95% CI: 2.56.2) of the placebo group (relative risk [RR]: 0.64; 95% CI: 0.32‐1.28). In nonobese patients, the primary end point occurred in 2.8% (95% CI: 1.8‐3.8) and 5.2% (95% CI: 3.9‐6.6) of the dalteparin and placebo groups, respectively (RR: 0.53; 95% CI: 0.34‐0.82). When weight was modeled as a continuous variable, no statistically significant interaction between weight and dalteparin efficacy was observed (P=0.97). The authors calculated the RR in predefined BMI subgroups and found that dalteparin was effective in reducing VTE in patients with BMIs up to 40, with RRs of <1.0 for all (approximate range, 0.20.8). However, a fixed dose of dalteparin 5000 IU/day was not better than placebo for individuals with BMI >40 kg/m². There was no significant difference in mortality or major hemorrhage by day 21 between treatment and placebo groups.

Freeman and colleagues prospectively assigned 31 medically ill patients with extreme obesity (BMI >40 kg/m²) to 1 of 3 dosing regimens of enoxaparin: a fixed dose of 40 mg daily enoxaparin (control group, n=11), enoxaparin at 0.4 mg/kg (n=9), or enoxaparin at 0.5 mg/kg (n=11).[12] The average BMI of the entire cohort was 62.1 kg/m² (range, 40.582.4). All patients had anti‐factor Xa levels drawn on the day of enrollment and daily for 3 days (Table 2). The relationship between anti‐factor Xa levels and clinical efficacy of low‐molecular weight heparin (LMWH) in VTE prophylaxis is still unclear; however, an anti‐factor Xa level of 0.2 to 0.5 IU/mL, measured 4 hours after the fourth dose of LMWH, is the target level recommended for VTE prophylaxis.[13] Patients who received weight‐based enoxaparin at 0.5mg/kg achieved target anti‐factor Xa level 86% of the time compared to 32% of the time in those receiving 0.4 mg/kg and 19% of the time for those in the fixed‐dose group (P<0.001). No clinical outcomes were reported in this study.

Patients on Antiplatelet Drugs

We did not find studies that directly looked at the comparative effectiveness of VTE prophylaxis in patients who were on antiplatelet drugs including aspirin. However, there were 2 studies that looked at the risk of bleeding in patients who received VTE pharmacologic prophylaxis while concurrently taking antiplatelet agents including aspirin. Both studies used pooled data from large phase III trials.

The study by Eriksson et al. used data from the RECORD (Regulation of Coagulation in Orthopedic Surgery to Prevent Deep Venous Thrombosis and Pulmonary Embolism) trial where over 12,000 patients undergoing elective total knee or hip replacement were randomized to receive VTE prophylaxis with oral rivaroxaban or subcutaneous enoxaparin.[14] Nine percent of participants in each arm (563 in rivaroxaban and 526 in enoxaparin/placebo) were concomitantly using antiplatelet agents or aspirin at least once during the at risk period, defined as starting at day 1 of surgery up to 2 days after the last intake of the study drug. The only end point evaluated was bleeding, and the authors found no statistically significant bleeding difference among the 2 arms (Table 1). Any bleeding event in the rivaroxaban with antiplatelets or aspirin arm was found in 20 (3.6%) patients, whereas in those on enoxaparin/placebo with antiplatelets or aspirin arm it was 17 (3.2%). The relative rate of bleeding among users versus nonusers of antiplatelet drugs or aspirin was 1.32 (95% CI: 0.85‐2.05) in the rivaroxaban group and 1.40 (95% CI: 0.87‐2.25) in the enoxaparin arm (Table 1).

Friedman et al. used pooled data from the RE‐MODEL, RENOVATE, and REMOBILIZE trials, where patients who were undergoing hip or knee arthroplasty were randomized to 220 mg of dabigatran once daily, 150 mg of dabigatran once daily (we focused on this lower dosage as this is the only available dose used in the US), 40 mg of enoxaparin once daily, or 30 mg of enoxaparin twice a day.[15] Of the 8135 patients, 4.7% were on concomitant aspirin. The baseline characteristics of those on aspirin were similar to the other enrollees. The primary outcome was major bleeding events requiring transfusion, symptomatic internal bleeding, or bleeding requiring surgery. Among patients receiving 150 mg of dabigatran, bleeding events with and without concomitant aspirin occurred in 1.6% and 1.0%, respectively (odds ratio [OR]: 1.64; 95% CI: 0.36‐7.49; P=0.523). The percentages of participants with bleeding who received enoxaparin, with and without aspirin, were 3.0% and 1.2%, respectively (OR: 2.57; 95% CI: 0.83‐7.94; P=0.101). The RR of bleeding on dabigatran compared to enoxaparin with and without aspirin therapy was 0.55 (95% CI: 0.11‐2.78) and 0.82 (95% CI: 0.37‐1.84), respectively (Table 1).

Patients With Renal Insufficiency

We found 5 studies that evaluated the comparative effectiveness and safety of pharmacologic prophylaxis for prevention of VTE in patients with acute kidney injury, moderate renal insufficiency, severe renal insufficiency not undergoing dialysis, or patients receiving dialysis. Four studies were RCTs,[16, 17, 18, 19] and 1 used a cohort design assessing separate cohorts before and after a quality improvement intervention.[20] Bauersachs and colleagues conducted an RCT comparing unfractionated heparin at 5000 IU, 3 times daily to certoparin, which is not approved in the United States and is not further discussed here.[16] The rate of DVT among patients treated with unfractionated heparin in patients with a glomerular filtration rate >30 mL/min was marginally lower than those with severe renal dysfunction (10.3 vs 11.1%) (Table 1).

Patients with severe renal dysfunction who received 5000 IU of unfractionated heparin 3 times a day were at increased risk of all bleeds (RR: 3.4; 95% CI: 2.05.9), major bleeds (RR: 7.3; 95% CI: 3.316), and minor bleeds (RR: 2.6; 95% CI: 1.4‐4.9) compared to patients treated with unfractionated heparin without severe renal dysfunction.[16]

A randomized trial by Mah and colleagues compared drug accumulation and anti‐Xa activity in elderly patients with renal dysfunction (defined as a glomerular filtration rate of 20 to 50 mL/min) who received either tinzaparin at 4500 IU once daily or enoxaparin at 4000 IU once daily.[17] Enoxaparin accumulated to a greater extent from day 1 to day 8 than did tinzaparin; the ratio of maximum concentration on day 8 compared to day 1 was 1.22 for enoxaparin and 1.05 for tinzaparin (P=0.016). No VTE events were reported in patients who received tinzaparin or enoxaparin. There was no statistical difference in the incidence of bleeding events between patients receiving tinzaparin (5, including 2 major events) and enoxaparin (4, including 3 major events, P=0.67) (Table 1).

The trial by Dahl and colleagues randomly assigned patients who were over 75 years of age and/or who had moderate renal dysfunction (defined as creatinine clearance between 30 and 49 mL/min) to receive enoxaparin 40 mg daily or dabigatran 150 mg daily.[18] There was no significant difference in the rate of major VTE events between patients receiving dabigatran (4.3%) and enoxaparin (9%) (OR: 0.48; 95% CI: 0.13‐1.73; P=0.271) (Table 1). The rate of major bleeding was significantly higher among patients randomly assigned to receive enoxaparin (4.7%) versus dabigatran (0%) (P=0.039).[18]

Shorr and colleagues published a post hoc subgroup analysis of a multicenter trial in which orthopedic patients were randomly assigned to receive desirudin 15 mg twice daily or enoxaparin 40 mg once daily.[19] Evaluable patients (1565 of the 2079 patients randomized in the trial) receiving desirudin experienced a significantly lower rate of major VTE compared with patients receiving enoxaparin (4.9% vs 7.6%, P=0.019). This relationship was particularly pronounced for evaluable patients whose creatinine clearance was between 30 and 44 mL/min. In evaluable patients with this degree of renal dysfunction, 11% of patients taking enoxaparin compared to 3.4% of those taking desirudin had a major VTE (OR: 3.52; 95% CI: 1.48‐8.4; P=0.004). There was no significant difference in the rates of major bleeding among a subset of patients assessed for safety outcomes (2078 of the 2079 patients randomized in the trial) who received desirudin (0.8%) or enoxaparin (0.2%) (Table 1).

Elsaid and Collins assessed VTE and bleeding events associated with the use of unfractionated heparin 5000U either 2 or 3 times daily and enoxaparin 30 mg once or twice daily across patients stratified by renal function (creatinine clearance <30, 3059, and 60 mL/min). The investigators made assessments before and after a quality improvement intervention that was designed to eliminate the use of enoxaparin in patients whose creatinine clearance was <30 mL/min. No VTE events were reported. Patients receiving enoxaparin were significantly more likely to experience a major bleeding episode compared with patients receiving unfractionated heparin (overall rates for all levels of renal function: 13.5% vs 4. 2%; RR: 3.2; 95% CI: 1.47.3) (Table 2). This association was largely driven by the subgroup of patients with a creatinine clearance <30 mL/min. For this subgroup with severe renal insufficiency, patients receiving enoxaparin were significantly more likely to have a bleed compared with patients receiving unfractionated heparin (18.9% vs 4.1%; RR: 4.68; 95% CI: 1.120.6) (Tables 1 and 2). There was no difference in the bleeding rates for patients whose creatinine clearances were >60 mL/min.[20]

Strength of Evidence

Search Strategy

We searched MEDLINE, EMBASE, and SCOPUS through August 2011, CINAHL, International Pharmaceutical Abstracts, clinicaltrial.gov, and the Cochrane Library through August 2012. We developed a search strategy based on medical subject headings (MeSH) terms and text words of key articles that we identified a priori[9] (see the Appendix for search strategy details).

Study Selection

We reviewed titles followed by abstracts to identify randomized controlled trials (RCTs) or observational studies with comparison groups reporting on the effectiveness or safety of VTE prevention in our populations. Two investigators independently reviewed abstracts, and we excluded the abstracts if both investigators agreed that the article met 1 or more of the exclusion criteria. We included only English‐language articles that evaluated the effectiveness of pharmacological or mechanical interventions that have been approved for clinical use in the United States. To be eligible, the studies must have addressed relevant key questions in the population of our interest. We resolved disagreements by consensus. We used DistillerSR (Evidence Partners Inc., Ottawa, Ontario, Canada), a Web‐based database management program to manage the review process. Two investigators assessed the risk of bias in each study independently, using the Downs and Black instrument for observational studies and trials.[10]

Data Synthesis

For each select population, we created detailed evidence tables containing the information abstracted from the eligible studies. After synthesizing the evidence, we graded the quantity, quality, and consistency of the best available evidence for each select population by adapting an evidence‐grading scheme recommended in the Methods Guide for Conducting Comparative Effectiveness Reviews.[7]

Obese Patients

We found 1 subgroup analysis of an RCT (total 3706 patients, 2563 nonobese and 1118 obese patients) that reported on the comparative effectiveness and safety of fixed low‐dose dalteparin 5000 IU/day compared to placebo among 1118 hospitalized medically ill patients with body mass indices (BMI) greater than 30 kg/m².¹¹ Neither group received additional concurrent prophylactic therapies. The 3 most prevalent medical diagnoses prompting hospitalization were congestive heart failure, respiratory failure, and infectious diseases. Compression ultrasound was performed in all patients by day 21 of hospitalization. The primary end point was the composite of VTE, fatal PE, and sudden death, and secondary end points included DVT, bleeding, and thrombocytopenia by day 21 (Table 1). In obese patients, the primary end point occurred in 2.8% (95% confidence interval [CI]: 1.34.3) of the dalteparin group and in 4.3% (95% CI: 2.56.2) of the placebo group (relative risk [RR]: 0.64; 95% CI: 0.32‐1.28). In nonobese patients, the primary end point occurred in 2.8% (95% CI: 1.8‐3.8) and 5.2% (95% CI: 3.9‐6.6) of the dalteparin and placebo groups, respectively (RR: 0.53; 95% CI: 0.34‐0.82). When weight was modeled as a continuous variable, no statistically significant interaction between weight and dalteparin efficacy was observed (P=0.97). The authors calculated the RR in predefined BMI subgroups and found that dalteparin was effective in reducing VTE in patients with BMIs up to 40, with RRs of <1.0 for all (approximate range, 0.20.8). However, a fixed dose of dalteparin 5000 IU/day was not better than placebo for individuals with BMI >40 kg/m². There was no significant difference in mortality or major hemorrhage by day 21 between treatment and placebo groups.

Freeman and colleagues prospectively assigned 31 medically ill patients with extreme obesity (BMI >40 kg/m²) to 1 of 3 dosing regimens of enoxaparin: a fixed dose of 40 mg daily enoxaparin (control group, n=11), enoxaparin at 0.4 mg/kg (n=9), or enoxaparin at 0.5 mg/kg (n=11).[12] The average BMI of the entire cohort was 62.1 kg/m² (range, 40.582.4). All patients had anti‐factor Xa levels drawn on the day of enrollment and daily for 3 days (Table 2). The relationship between anti‐factor Xa levels and clinical efficacy of low‐molecular weight heparin (LMWH) in VTE prophylaxis is still unclear; however, an anti‐factor Xa level of 0.2 to 0.5 IU/mL, measured 4 hours after the fourth dose of LMWH, is the target level recommended for VTE prophylaxis.[13] Patients who received weight‐based enoxaparin at 0.5mg/kg achieved target anti‐factor Xa level 86% of the time compared to 32% of the time in those receiving 0.4 mg/kg and 19% of the time for those in the fixed‐dose group (P<0.001). No clinical outcomes were reported in this study.

Patients on Antiplatelet Drugs

We did not find studies that directly looked at the comparative effectiveness of VTE prophylaxis in patients who were on antiplatelet drugs including aspirin. However, there were 2 studies that looked at the risk of bleeding in patients who received VTE pharmacologic prophylaxis while concurrently taking antiplatelet agents including aspirin. Both studies used pooled data from large phase III trials.

The study by Eriksson et al. used data from the RECORD (Regulation of Coagulation in Orthopedic Surgery to Prevent Deep Venous Thrombosis and Pulmonary Embolism) trial where over 12,000 patients undergoing elective total knee or hip replacement were randomized to receive VTE prophylaxis with oral rivaroxaban or subcutaneous enoxaparin.[14] Nine percent of participants in each arm (563 in rivaroxaban and 526 in enoxaparin/placebo) were concomitantly using antiplatelet agents or aspirin at least once during the at risk period, defined as starting at day 1 of surgery up to 2 days after the last intake of the study drug. The only end point evaluated was bleeding, and the authors found no statistically significant bleeding difference among the 2 arms (Table 1). Any bleeding event in the rivaroxaban with antiplatelets or aspirin arm was found in 20 (3.6%) patients, whereas in those on enoxaparin/placebo with antiplatelets or aspirin arm it was 17 (3.2%). The relative rate of bleeding among users versus nonusers of antiplatelet drugs or aspirin was 1.32 (95% CI: 0.85‐2.05) in the rivaroxaban group and 1.40 (95% CI: 0.87‐2.25) in the enoxaparin arm (Table 1).

Friedman et al. used pooled data from the RE‐MODEL, RENOVATE, and REMOBILIZE trials, where patients who were undergoing hip or knee arthroplasty were randomized to 220 mg of dabigatran once daily, 150 mg of dabigatran once daily (we focused on this lower dosage as this is the only available dose used in the US), 40 mg of enoxaparin once daily, or 30 mg of enoxaparin twice a day.[15] Of the 8135 patients, 4.7% were on concomitant aspirin. The baseline characteristics of those on aspirin were similar to the other enrollees. The primary outcome was major bleeding events requiring transfusion, symptomatic internal bleeding, or bleeding requiring surgery. Among patients receiving 150 mg of dabigatran, bleeding events with and without concomitant aspirin occurred in 1.6% and 1.0%, respectively (odds ratio [OR]: 1.64; 95% CI: 0.36‐7.49; P=0.523). The percentages of participants with bleeding who received enoxaparin, with and without aspirin, were 3.0% and 1.2%, respectively (OR: 2.57; 95% CI: 0.83‐7.94; P=0.101). The RR of bleeding on dabigatran compared to enoxaparin with and without aspirin therapy was 0.55 (95% CI: 0.11‐2.78) and 0.82 (95% CI: 0.37‐1.84), respectively (Table 1).

Patients With Renal Insufficiency

We found 5 studies that evaluated the comparative effectiveness and safety of pharmacologic prophylaxis for prevention of VTE in patients with acute kidney injury, moderate renal insufficiency, severe renal insufficiency not undergoing dialysis, or patients receiving dialysis. Four studies were RCTs,[16, 17, 18, 19] and 1 used a cohort design assessing separate cohorts before and after a quality improvement intervention.[20] Bauersachs and colleagues conducted an RCT comparing unfractionated heparin at 5000 IU, 3 times daily to certoparin, which is not approved in the United States and is not further discussed here.[16] The rate of DVT among patients treated with unfractionated heparin in patients with a glomerular filtration rate >30 mL/min was marginally lower than those with severe renal dysfunction (10.3 vs 11.1%) (Table 1).

Patients with severe renal dysfunction who received 5000 IU of unfractionated heparin 3 times a day were at increased risk of all bleeds (RR: 3.4; 95% CI: 2.05.9), major bleeds (RR: 7.3; 95% CI: 3.316), and minor bleeds (RR: 2.6; 95% CI: 1.4‐4.9) compared to patients treated with unfractionated heparin without severe renal dysfunction.[16]

A randomized trial by Mah and colleagues compared drug accumulation and anti‐Xa activity in elderly patients with renal dysfunction (defined as a glomerular filtration rate of 20 to 50 mL/min) who received either tinzaparin at 4500 IU once daily or enoxaparin at 4000 IU once daily.[17] Enoxaparin accumulated to a greater extent from day 1 to day 8 than did tinzaparin; the ratio of maximum concentration on day 8 compared to day 1 was 1.22 for enoxaparin and 1.05 for tinzaparin (P=0.016). No VTE events were reported in patients who received tinzaparin or enoxaparin. There was no statistical difference in the incidence of bleeding events between patients receiving tinzaparin (5, including 2 major events) and enoxaparin (4, including 3 major events, P=0.67) (Table 1).

The trial by Dahl and colleagues randomly assigned patients who were over 75 years of age and/or who had moderate renal dysfunction (defined as creatinine clearance between 30 and 49 mL/min) to receive enoxaparin 40 mg daily or dabigatran 150 mg daily.[18] There was no significant difference in the rate of major VTE events between patients receiving dabigatran (4.3%) and enoxaparin (9%) (OR: 0.48; 95% CI: 0.13‐1.73; P=0.271) (Table 1). The rate of major bleeding was significantly higher among patients randomly assigned to receive enoxaparin (4.7%) versus dabigatran (0%) (P=0.039).[18]

Shorr and colleagues published a post hoc subgroup analysis of a multicenter trial in which orthopedic patients were randomly assigned to receive desirudin 15 mg twice daily or enoxaparin 40 mg once daily.[19] Evaluable patients (1565 of the 2079 patients randomized in the trial) receiving desirudin experienced a significantly lower rate of major VTE compared with patients receiving enoxaparin (4.9% vs 7.6%, P=0.019). This relationship was particularly pronounced for evaluable patients whose creatinine clearance was between 30 and 44 mL/min. In evaluable patients with this degree of renal dysfunction, 11% of patients taking enoxaparin compared to 3.4% of those taking desirudin had a major VTE (OR: 3.52; 95% CI: 1.48‐8.4; P=0.004). There was no significant difference in the rates of major bleeding among a subset of patients assessed for safety outcomes (2078 of the 2079 patients randomized in the trial) who received desirudin (0.8%) or enoxaparin (0.2%) (Table 1).

Elsaid and Collins assessed VTE and bleeding events associated with the use of unfractionated heparin 5000U either 2 or 3 times daily and enoxaparin 30 mg once or twice daily across patients stratified by renal function (creatinine clearance <30, 3059, and 60 mL/min). The investigators made assessments before and after a quality improvement intervention that was designed to eliminate the use of enoxaparin in patients whose creatinine clearance was <30 mL/min. No VTE events were reported. Patients receiving enoxaparin were significantly more likely to experience a major bleeding episode compared with patients receiving unfractionated heparin (overall rates for all levels of renal function: 13.5% vs 4. 2%; RR: 3.2; 95% CI: 1.47.3) (Table 2). This association was largely driven by the subgroup of patients with a creatinine clearance <30 mL/min. For this subgroup with severe renal insufficiency, patients receiving enoxaparin were significantly more likely to have a bleed compared with patients receiving unfractionated heparin (18.9% vs 4.1%; RR: 4.68; 95% CI: 1.120.6) (Tables 1 and 2). There was no difference in the bleeding rates for patients whose creatinine clearances were >60 mL/min.[20]

Strength of Evidence

Design

The study was a prospective, unblinded, randomized, controlled trial that was approved by the Colorado Multiple Institutional Review Board and conducted at Denver Health, a university‐affiliated public safety net hospital. No protocol changes occurred during the study.

Participants

Participants included hospitalist physicians, internal medicine residents, physician assistants, nurse practitioners, and nurses who directly cared for patients hospitalized on internal medicine units between March 12, 2012 and August 28, 2012. Participants known to be pregnant or those who refused to participate in the study were excluded.

Intervention

Standard scrubs issued by the hospital were tested along with 2 different antimicrobial scrubs (scrub A and scrub B). Scrub A was made with a polyester microfiber material embedded with a proprietary antimicrobial chemical. Scrub B was a polyestercotton blend scrub that included 2 proprietary antimicrobial chemicals and silver embedded into the fabric. The standard scrub was made of a polyestercotton blend with no antimicrobial properties. All scrubs consisted of pants and a short‐sleeved shirt, with either a pocket at the left breast or lower front surface, and all were tested new prior to any washing or wear. Preliminary cultures were done on 2 scrubs in each group to assess the extent of preuse contamination. All providers were instructed not to wear white coats at any time during the day that they were wearing the scrubs. Providers were not told the type of scrub they received, but the antimicrobial scrubs had a different appearance and texture than the standard scrubs, so blinding was not possible.

Outcomes

The primary end point was the total bacterial colony count of samples obtained from the breast or lower front pocket, the sleeve cuff of the dominant hand, and the pant leg at the midthigh of the dominant leg on all scrubs after an 8‐hour workday. Secondary outcomes were the bacterial colony counts of cultures obtained from the volar surface of the wrists of the HCWs' dominant arm, and the colony counts of methicillin‐resistant Staphylococcus aureus (MRSA), vancomycin‐resistant enterococci (VRE), and resistant Gram‐negative bacteria on the 3 scrub types, all obtained after the 8‐hour workday.

Cultures were collected using a standardized RODAC imprint method[21] with BBL RODAC plates containing blood agar (Becton Dickinson, Sparks, MD). Cultures were incubated in ambient air at 35 to 37C for 18 to 22 hours. After incubation, visible colonies were counted using a dissecting microscope to a maximum of 200 colonies as recommended by the manufacturer. Colonies morphologically consistent with Staphylococcus species were subsequently tested for coagulase using a BactiStaph rapid latex agglutination test (Remel, Lenexa, KS). If positive, these colonies were subcultured to sheep blood agar (Remel) and BBL MRSA CHROMagar (Becton Dickinson) and incubated for an additional 18 to 24 hours. Characteristic growth on blood agar that also produced mauve‐colored colonies on CHROMagar was taken to indicate MRSA. Colonies morphologically suspicious for being VRE were identified and confirmed as VRE using a positive identification and susceptibility panel (Microscan; Siemens, Deerfield, IL). A negative combination panel (Microscan, Siemens) was also used to identify and confirm resistant Gram‐negative rods.

Each participant completed a survey that included questions that identified their occupation, whether they had had contact with patients who were known to be colonized or infected with MRSA, VRE, or resistant Gram‐negative rods during the testing period, and whether they experienced any adverse events that might relate to wearing the uniform.

Sample Size

We assumed that cultures taken from the sleeve of the control scrubs would have a mean ( standard deviation) colony count of 69 (67) based on data from our previous study.[12] Although the companies making the antimicrobial scrubs indicated that their respective products provided between 80.9% at 8 hours and >99% reduction in bacterial colony counts in laboratory settings, we assumed that a 70% decrease in colony count compared with standard scrubs could be clinically important. After adjusting for multiple comparisons and accounting for using nonparametric analyses with an unknown distribution, we estimated a need to recruit 35 subjects in each of 3 groups.

Randomization

The principal investigator and coinvestigators enrolled and consented participants. After obtaining consent, block randomization, stratified by occupation, occurred 1 day prior to the study using a computer‐generated table of random numbers.

Statistics

Data were collected and managed using REDCap (Research Electronic Data Capture; Vanderbilt UniversityThe Institute for Medicine and Public Health, Nashville, TN) electronic data capture tools hosted at Denver Health. REDCap is a secure Web‐based application designed to support data collection for research studies, providing: (1) an intuitive interface for validated data entry, (2) audit trails for tracking data manipulation and export procedures, (3) automated export procedures for seamless data downloads to common statistical packages, and (4) procedures for importing data from external sources.[22]

Colony counts were compared using a Kruskal‐Wallis 1‐way analysis of variance by ranks. Bonferroni's correction for multiple comparisons resulted in a P<0.01 as indicating statistical significance. Proportions were compared using [2] analysis. All data are presented as medians with interquartile range (IQR) or proportions.

Adverse Events

Six subjects (5.7%) reported adverse events, all of whom were wearing antimicrobial scrubs (P=0.18). For participants wearing antimicrobial scrub A, 1 (3%) reported itchiness and 2 (6%) reported heaviness or poor breathability. For participants wearing antimicrobial scrub B, 1 (3%) reported redness, 1 (3%) reported itchiness, and 1 (3%) reported heaviness or poor breathability.