SAN DIEGO – Researchers were able to “flip the switch” from the adult to fetal form of hemoglobin using autologous stem cells genetically modified to simultaneously induce the fetal form of hemoglobin and decrease sickle hemoglobin.

The advance was announced by investigators at the Dana-Farber Cancer Institute and Boston Children’s Hospital at the annual meeting of the American Society of Hematology. At 6 months of follow-up, one adult patient in the proof-of-concept study has experienced a reversal of the sickle cell phenotype, with no pain episodes or respiratory or neurologic events.

The fetal form of hemoglobin is known to be protective against the signs and symptoms of sickle cell disease, but apart from a few rare exceptions, people with the disorder begin to experience debilitating symptoms as levels of the fetal form begin to decline in early childhood and levels of the adult form of hemoglobin steadily rise.

In this video interview, Erica B. Esrick, MD, from the Dana-Farber/Boston Children’s Cancer and Blood Disorders Center, describes the novel approach of using RNA interference to knock down a repressor that suppresses expression of gamma globin in sickle cell disease.

SAN DIEGO – Researchers were able to “flip the switch” from the adult to fetal form of hemoglobin using autologous stem cells genetically modified to simultaneously induce the fetal form of hemoglobin and decrease sickle hemoglobin.

The advance was announced by investigators at the Dana-Farber Cancer Institute and Boston Children’s Hospital at the annual meeting of the American Society of Hematology. At 6 months of follow-up, one adult patient in the proof-of-concept study has experienced a reversal of the sickle cell phenotype, with no pain episodes or respiratory or neurologic events.

The fetal form of hemoglobin is known to be protective against the signs and symptoms of sickle cell disease, but apart from a few rare exceptions, people with the disorder begin to experience debilitating symptoms as levels of the fetal form begin to decline in early childhood and levels of the adult form of hemoglobin steadily rise.

In this video interview, Erica B. Esrick, MD, from the Dana-Farber/Boston Children’s Cancer and Blood Disorders Center, describes the novel approach of using RNA interference to knock down a repressor that suppresses expression of gamma globin in sickle cell disease.

SAN DIEGO – Researchers were able to “flip the switch” from the adult to fetal form of hemoglobin using autologous stem cells genetically modified to simultaneously induce the fetal form of hemoglobin and decrease sickle hemoglobin.

The advance was announced by investigators at the Dana-Farber Cancer Institute and Boston Children’s Hospital at the annual meeting of the American Society of Hematology. At 6 months of follow-up, one adult patient in the proof-of-concept study has experienced a reversal of the sickle cell phenotype, with no pain episodes or respiratory or neurologic events.

The fetal form of hemoglobin is known to be protective against the signs and symptoms of sickle cell disease, but apart from a few rare exceptions, people with the disorder begin to experience debilitating symptoms as levels of the fetal form begin to decline in early childhood and levels of the adult form of hemoglobin steadily rise.

In this video interview, Erica B. Esrick, MD, from the Dana-Farber/Boston Children’s Cancer and Blood Disorders Center, describes the novel approach of using RNA interference to knock down a repressor that suppresses expression of gamma globin in sickle cell disease.

CHICAGO – Apremilast was effective and well tolerated for up to 28 weeks for the treatment of oral ulcers in patients with active Behçet’s disease, based on findings from the randomized, placebo-controlled, phase 3 RELIEF trial.

At baseline, mean oral ulcer counts were 4.2 in 104 patients randomized to receive the oral phosphodiesterase-4 inhibitor and 3.9 in 103 patients in the placebo group. Mean visual analog scale (VAS) pain scores were 61.2 and 60.8 in the two groups, respectively.

The primary study endpoint of area under the curve for total number of oral ulcers over a 12-week period (AUC_Wk0-12) – a measure that reflects the number of oral ulcers that occur over time and also accounts for the recurring-remitting course of oral ulcers – was achieved. AUC_Wk0-12 was significantly lower in the apremilast group than in the placebo group (129.54 vs. 222.14, respectively; P less than .0001), Gulen Hatemi, MD, reported at the annual meeting of the American College of Rheumatology.

From baseline to week 12, apremilast treatment also resulted in a significantly lower number of oral ulcers (mean of 1.1 vs. 2.0 for placebo at 12 weeks) and significantly reduced pain from oral ulcers at every visit from week 1 through week 12 of the study, compared with placebo (mean VAS score change from baseline, –40.7 vs. –15.9), said Dr. Hatemi, a professor of medicine at Istanbul University.

“The [12-week] complete response rate ... was 53% in the apremilast group and 22.3% in the placebo group. The [12-week] partial response rate ...was 76% in the apremilast group and 48% in the placebo group,” she said, adding that the efficacy of apremilast was sustained with continued treatment through 28 weeks.

Study participants were adults (mean age, 40 years) with active Behçet’s disease and three or more oral ulcers at randomization or two or more at screening and at randomization. All had been previously treated with at least one nonbiologic medication for oral ulcers and were allowed to have received previous biologic therapies for other disease manifestations. Those with active major organ involvement were excluded.

Treatment included a 30-mg dose of apremilast twice daily for 12 weeks or placebo. After 12 weeks, all patients received apremilast through at least 28 weeks of the 64-week study.

At the 28-week analysis, patients who were initially randomized to placebo and who switched to apremilast after week 12 had benefits comparable with those seen in those randomized to apremilast at the start of the study. A complete response was seen in 59% and 62% of patients in the groups, respectively, and a partial response was seen in 90% and 85%, respectively. Additionally, the mean change in the VAS score for oral ulcer pain in the groups at that time was –40.6 and –41.9, Dr. Hatemi said.

Apremilast was well tolerated in this study; the incidence of adverse events was comparable in the treatment and placebo groups during the 12-week placebo-controlled phase of the study – 78.8% and 71.8%, respectively. The most common events were diarrhea, nausea, headache, and upper respiratory tract infection, she said.

“These were generally mild to moderate, and only two patients had to discontinue the study due to gastrointestinal adverse events,” she said, noting that no new safety signals were observed.

Behçet’s disease is a chronic, relapsing, multisystem inflammatory disorder characterized by recurrent oral ulcers that can be disabling and have a substantial effect on quality of life. These findings, which include efficacy data up to 28 weeks and safety data for at least 100 patients exposed to apremilast for at least 1 year, demonstrate the efficacy of apremilast for the treatment oral ulcers in patients with Behçet’s disease, she said, noting that “the safety findings were consistent with the known safety profile of apremilast.”

The RELIEF study was supported by Celgene. Dr. Hatemi reported receiving grant/research support from Celgene and serving as a speaker for AbbVie, Mustafa Nevzet Pharmaceuticals, and UCB.

[email protected]

SOURCE: Hatemi G et al. Arthritis Rheumatol. 2018;70(Suppl 10), Abstract 2789.

CHICAGO – Apremilast was effective and well tolerated for up to 28 weeks for the treatment of oral ulcers in patients with active Behçet’s disease, based on findings from the randomized, placebo-controlled, phase 3 RELIEF trial.

At baseline, mean oral ulcer counts were 4.2 in 104 patients randomized to receive the oral phosphodiesterase-4 inhibitor and 3.9 in 103 patients in the placebo group. Mean visual analog scale (VAS) pain scores were 61.2 and 60.8 in the two groups, respectively.

The primary study endpoint of area under the curve for total number of oral ulcers over a 12-week period (AUC_Wk0-12) – a measure that reflects the number of oral ulcers that occur over time and also accounts for the recurring-remitting course of oral ulcers – was achieved. AUC_Wk0-12 was significantly lower in the apremilast group than in the placebo group (129.54 vs. 222.14, respectively; P less than .0001), Gulen Hatemi, MD, reported at the annual meeting of the American College of Rheumatology.

From baseline to week 12, apremilast treatment also resulted in a significantly lower number of oral ulcers (mean of 1.1 vs. 2.0 for placebo at 12 weeks) and significantly reduced pain from oral ulcers at every visit from week 1 through week 12 of the study, compared with placebo (mean VAS score change from baseline, –40.7 vs. –15.9), said Dr. Hatemi, a professor of medicine at Istanbul University.

“The [12-week] complete response rate ... was 53% in the apremilast group and 22.3% in the placebo group. The [12-week] partial response rate ...was 76% in the apremilast group and 48% in the placebo group,” she said, adding that the efficacy of apremilast was sustained with continued treatment through 28 weeks.

Study participants were adults (mean age, 40 years) with active Behçet’s disease and three or more oral ulcers at randomization or two or more at screening and at randomization. All had been previously treated with at least one nonbiologic medication for oral ulcers and were allowed to have received previous biologic therapies for other disease manifestations. Those with active major organ involvement were excluded.

Treatment included a 30-mg dose of apremilast twice daily for 12 weeks or placebo. After 12 weeks, all patients received apremilast through at least 28 weeks of the 64-week study.

At the 28-week analysis, patients who were initially randomized to placebo and who switched to apremilast after week 12 had benefits comparable with those seen in those randomized to apremilast at the start of the study. A complete response was seen in 59% and 62% of patients in the groups, respectively, and a partial response was seen in 90% and 85%, respectively. Additionally, the mean change in the VAS score for oral ulcer pain in the groups at that time was –40.6 and –41.9, Dr. Hatemi said.

Apremilast was well tolerated in this study; the incidence of adverse events was comparable in the treatment and placebo groups during the 12-week placebo-controlled phase of the study – 78.8% and 71.8%, respectively. The most common events were diarrhea, nausea, headache, and upper respiratory tract infection, she said.

“These were generally mild to moderate, and only two patients had to discontinue the study due to gastrointestinal adverse events,” she said, noting that no new safety signals were observed.

Behçet’s disease is a chronic, relapsing, multisystem inflammatory disorder characterized by recurrent oral ulcers that can be disabling and have a substantial effect on quality of life. These findings, which include efficacy data up to 28 weeks and safety data for at least 100 patients exposed to apremilast for at least 1 year, demonstrate the efficacy of apremilast for the treatment oral ulcers in patients with Behçet’s disease, she said, noting that “the safety findings were consistent with the known safety profile of apremilast.”

The RELIEF study was supported by Celgene. Dr. Hatemi reported receiving grant/research support from Celgene and serving as a speaker for AbbVie, Mustafa Nevzet Pharmaceuticals, and UCB.

[email protected]

SOURCE: Hatemi G et al. Arthritis Rheumatol. 2018;70(Suppl 10), Abstract 2789.

CHICAGO – Apremilast was effective and well tolerated for up to 28 weeks for the treatment of oral ulcers in patients with active Behçet’s disease, based on findings from the randomized, placebo-controlled, phase 3 RELIEF trial.

At baseline, mean oral ulcer counts were 4.2 in 104 patients randomized to receive the oral phosphodiesterase-4 inhibitor and 3.9 in 103 patients in the placebo group. Mean visual analog scale (VAS) pain scores were 61.2 and 60.8 in the two groups, respectively.

The primary study endpoint of area under the curve for total number of oral ulcers over a 12-week period (AUC_Wk0-12) – a measure that reflects the number of oral ulcers that occur over time and also accounts for the recurring-remitting course of oral ulcers – was achieved. AUC_Wk0-12 was significantly lower in the apremilast group than in the placebo group (129.54 vs. 222.14, respectively; P less than .0001), Gulen Hatemi, MD, reported at the annual meeting of the American College of Rheumatology.

From baseline to week 12, apremilast treatment also resulted in a significantly lower number of oral ulcers (mean of 1.1 vs. 2.0 for placebo at 12 weeks) and significantly reduced pain from oral ulcers at every visit from week 1 through week 12 of the study, compared with placebo (mean VAS score change from baseline, –40.7 vs. –15.9), said Dr. Hatemi, a professor of medicine at Istanbul University.

“The [12-week] complete response rate ... was 53% in the apremilast group and 22.3% in the placebo group. The [12-week] partial response rate ...was 76% in the apremilast group and 48% in the placebo group,” she said, adding that the efficacy of apremilast was sustained with continued treatment through 28 weeks.

Study participants were adults (mean age, 40 years) with active Behçet’s disease and three or more oral ulcers at randomization or two or more at screening and at randomization. All had been previously treated with at least one nonbiologic medication for oral ulcers and were allowed to have received previous biologic therapies for other disease manifestations. Those with active major organ involvement were excluded.

Treatment included a 30-mg dose of apremilast twice daily for 12 weeks or placebo. After 12 weeks, all patients received apremilast through at least 28 weeks of the 64-week study.

At the 28-week analysis, patients who were initially randomized to placebo and who switched to apremilast after week 12 had benefits comparable with those seen in those randomized to apremilast at the start of the study. A complete response was seen in 59% and 62% of patients in the groups, respectively, and a partial response was seen in 90% and 85%, respectively. Additionally, the mean change in the VAS score for oral ulcer pain in the groups at that time was –40.6 and –41.9, Dr. Hatemi said.

Apremilast was well tolerated in this study; the incidence of adverse events was comparable in the treatment and placebo groups during the 12-week placebo-controlled phase of the study – 78.8% and 71.8%, respectively. The most common events were diarrhea, nausea, headache, and upper respiratory tract infection, she said.

“These were generally mild to moderate, and only two patients had to discontinue the study due to gastrointestinal adverse events,” she said, noting that no new safety signals were observed.

Behçet’s disease is a chronic, relapsing, multisystem inflammatory disorder characterized by recurrent oral ulcers that can be disabling and have a substantial effect on quality of life. These findings, which include efficacy data up to 28 weeks and safety data for at least 100 patients exposed to apremilast for at least 1 year, demonstrate the efficacy of apremilast for the treatment oral ulcers in patients with Behçet’s disease, she said, noting that “the safety findings were consistent with the known safety profile of apremilast.”

The RELIEF study was supported by Celgene. Dr. Hatemi reported receiving grant/research support from Celgene and serving as a speaker for AbbVie, Mustafa Nevzet Pharmaceuticals, and UCB.

[email protected]

SOURCE: Hatemi G et al. Arthritis Rheumatol. 2018;70(Suppl 10), Abstract 2789.

NEW ORLEANS – Depression severity is correlated with seizure frequency and quality of life in adults with epilepsy, according to an analysis presented at the annual meeting of the American Epilepsy Society.

The conclusion comes from a study of 120 people with epilepsy, 62 of whom had at least moderate depression based on the Patient Health Questionnaire-9 (PHQ-9). The Rapid Estimate of Adult Literacy in Medicine (REALM-R), Quality of Life in Epilepsy (QOLIE-10) and Charlson Comorbidity Index were used to assess patients’ health literacy, quality of life, and medical comorbidity, respectively

Among demographic characteristics, only inability to work was significantly associated with depression severity. Higher 30-day seizure frequency, panic disorder, and obsessive-compulsive disorder were correlated with more severe depression severity. Medical comorbidity was not associated with increased risk of depression.

Identifying and treating psychiatric comorbidities should be part of the management of patients with epilepsy, said Martha X. Sajatovic, MD, director of the Neurological and Behavioral Outcomes Center at Case Western Reserve University in Cleveland, who presented the data. “Following up to ensure they receive treatment is vital, because it can truly change patient outcomes and help them achieve their best quality of life.”

The study findings are consistent with those of previous research indicating that people with symptoms of depression are more likely to have more frequent seizures and decreased quality of life, said Dr. Sajatovic.

“Health care providers should screen their epilepsy patients for depression, but they shouldn’t stop there,” she advised. “A person may have depressive symptoms that don’t reach the level of depression but should be assessed for other types of mental health issues that could easily be overlooked.”

Patients with epilepsy should respond to the PHQ-9 annually, or more frequently, if warranted, she added.

“It’s important that people with epilepsy who have depression or other mental health issues get treatment such as cognitive behavioral therapy and medication,” said Dr. Sajatovic. “Even being in a self-management program helps, because the better they are at self management, the less likely they are to suffer negative health effects.”

This study was supported by a grant from the Centers for Disease Control and Prevention SIP 14-007 1U48DP005030.
 

[email protected]

SOURCE: Kumar N et al. AES 2018, Abstract 1.371.

NEW ORLEANS – Depression severity is correlated with seizure frequency and quality of life in adults with epilepsy, according to an analysis presented at the annual meeting of the American Epilepsy Society.

The conclusion comes from a study of 120 people with epilepsy, 62 of whom had at least moderate depression based on the Patient Health Questionnaire-9 (PHQ-9). The Rapid Estimate of Adult Literacy in Medicine (REALM-R), Quality of Life in Epilepsy (QOLIE-10) and Charlson Comorbidity Index were used to assess patients’ health literacy, quality of life, and medical comorbidity, respectively

Among demographic characteristics, only inability to work was significantly associated with depression severity. Higher 30-day seizure frequency, panic disorder, and obsessive-compulsive disorder were correlated with more severe depression severity. Medical comorbidity was not associated with increased risk of depression.

Identifying and treating psychiatric comorbidities should be part of the management of patients with epilepsy, said Martha X. Sajatovic, MD, director of the Neurological and Behavioral Outcomes Center at Case Western Reserve University in Cleveland, who presented the data. “Following up to ensure they receive treatment is vital, because it can truly change patient outcomes and help them achieve their best quality of life.”

The study findings are consistent with those of previous research indicating that people with symptoms of depression are more likely to have more frequent seizures and decreased quality of life, said Dr. Sajatovic.

“Health care providers should screen their epilepsy patients for depression, but they shouldn’t stop there,” she advised. “A person may have depressive symptoms that don’t reach the level of depression but should be assessed for other types of mental health issues that could easily be overlooked.”

Patients with epilepsy should respond to the PHQ-9 annually, or more frequently, if warranted, she added.

“It’s important that people with epilepsy who have depression or other mental health issues get treatment such as cognitive behavioral therapy and medication,” said Dr. Sajatovic. “Even being in a self-management program helps, because the better they are at self management, the less likely they are to suffer negative health effects.”

This study was supported by a grant from the Centers for Disease Control and Prevention SIP 14-007 1U48DP005030.
 

[email protected]

SOURCE: Kumar N et al. AES 2018, Abstract 1.371.

NEW ORLEANS – Depression severity is correlated with seizure frequency and quality of life in adults with epilepsy, according to an analysis presented at the annual meeting of the American Epilepsy Society.

The conclusion comes from a study of 120 people with epilepsy, 62 of whom had at least moderate depression based on the Patient Health Questionnaire-9 (PHQ-9). The Rapid Estimate of Adult Literacy in Medicine (REALM-R), Quality of Life in Epilepsy (QOLIE-10) and Charlson Comorbidity Index were used to assess patients’ health literacy, quality of life, and medical comorbidity, respectively

Among demographic characteristics, only inability to work was significantly associated with depression severity. Higher 30-day seizure frequency, panic disorder, and obsessive-compulsive disorder were correlated with more severe depression severity. Medical comorbidity was not associated with increased risk of depression.

Identifying and treating psychiatric comorbidities should be part of the management of patients with epilepsy, said Martha X. Sajatovic, MD, director of the Neurological and Behavioral Outcomes Center at Case Western Reserve University in Cleveland, who presented the data. “Following up to ensure they receive treatment is vital, because it can truly change patient outcomes and help them achieve their best quality of life.”

The study findings are consistent with those of previous research indicating that people with symptoms of depression are more likely to have more frequent seizures and decreased quality of life, said Dr. Sajatovic.

“Health care providers should screen their epilepsy patients for depression, but they shouldn’t stop there,” she advised. “A person may have depressive symptoms that don’t reach the level of depression but should be assessed for other types of mental health issues that could easily be overlooked.”

Patients with epilepsy should respond to the PHQ-9 annually, or more frequently, if warranted, she added.

“It’s important that people with epilepsy who have depression or other mental health issues get treatment such as cognitive behavioral therapy and medication,” said Dr. Sajatovic. “Even being in a self-management program helps, because the better they are at self management, the less likely they are to suffer negative health effects.”

This study was supported by a grant from the Centers for Disease Control and Prevention SIP 14-007 1U48DP005030.
 

[email protected]

SOURCE: Kumar N et al. AES 2018, Abstract 1.371.

NEW ORLEANS – Patients taking enzyme-inducing antiepileptic drugs (AEDs) may require a clinically meaningful increase in their vitamin D doses to achieve the same 25-hydroxyvitamin D (25[OH]D) plasma levels as patients taking nonenzyme-inducing AEDs, based on a retrospective chart review presented at the annual meeting of the American Epilepsy Society.

While patients receiving either type of AED had similar average 25(OH)D levels in the study (32.0 ng/mL in the enzyme-inducing AED group and 33.2 ng/mL in the noninducing AED group), those in the enzyme-inducing group required 1,587 U/day to meet the goal – a 409-unit increase in dose, compared with the 1,108 U/day dose taken by patients in the nonenzyme-inducing group.

“Patients taking enzyme-inducing AEDs may benefit from more intensive monitoring of their vitamin D supplementation, and clinicians should anticipate this likely pharmacokinetic interaction,” said Barry E. Gidal, PharmD, professor of pharmacy and neurology at the University of Wisconsin–Madison, and his colleagues.

Researchers have suggested that enzyme-inducing AEDs may affect CYP450 isoenzymes, increase vitamin D metabolism, and reduce 25(OH)D plasma levels. “It follows … that a potential pharmacokinetic interaction could exist between enzyme-inducing AEDs and oral formulations of vitamin D used for supplementation,” the investigators said.

To test the hypothesis, Dr. Gidal and his colleagues reviewed the charts of patients with epilepsy who were on any AED regimen and were prescribed vitamin D at William S. Middleton Memorial Veterans Hospital in Madison, Wisconsin, between January 2013 and September 2017.

The researchers grouped patients by those using enzyme-inducing AEDs and those taking noninducing AEDs. Patients who were taking AEDs in both categories were placed in the enzyme-inducing AED group. Patients with malabsorptive conditions and patients using calcitriol were excluded from the analysis.

Data included AEDs used, prescription and over-the-counter vitamin D use, 25(OH)D plasma concentration, renal function, age, gender, and ethnicity. Patients’ 25(OH)D levels were measured using a chemiluminescence immunoassay, and a minimum 25(OH)D plasma level of 30 ng/mL was the therapeutic goal.

The multivariant analysis was adjusted for potentially confounding variables including 25(OH)D concentration, over-the-counter vitamin D use, chronic kidney disease, age, gender, and ethnicity.

The analysis included 1,113 observations from 315 patients, and 263 of the observations (23.6%) were in the enzyme-inducing AED group. The enzyme-inducing group and noninducing groups were mostly male (90.5% and 91.8%, respectively) and similar in average age (65.9 and 61.4 years, respectively). Variables were evenly distributed between the groups, with the exceptions of chronic kidney disease, which was less common in the enzyme-inducing group (6.1% vs. 13.8%), and ethnicity (78.7% Caucasian in the enzyme-inducing group vs. 87.7% Caucasian in the noninducing group). The most common enzyme-inducing AED was phenytoin (50.6%), followed by carbamazepine (31.9%), phenobarbital (14.1%), oxcarbazepine (6.8%), primidone (1.9%), and eslicarbazepine (0.8%).

Dr. Gidal reported honoraria from Eisai, Sunovion, Lundbeck, and GW Pharmaceuticals.
 

[email protected]

SOURCE: Gidal BE et al. AES 2018, Abstract 1.315.

NEW ORLEANS – Patients taking enzyme-inducing antiepileptic drugs (AEDs) may require a clinically meaningful increase in their vitamin D doses to achieve the same 25-hydroxyvitamin D (25[OH]D) plasma levels as patients taking nonenzyme-inducing AEDs, based on a retrospective chart review presented at the annual meeting of the American Epilepsy Society.

While patients receiving either type of AED had similar average 25(OH)D levels in the study (32.0 ng/mL in the enzyme-inducing AED group and 33.2 ng/mL in the noninducing AED group), those in the enzyme-inducing group required 1,587 U/day to meet the goal – a 409-unit increase in dose, compared with the 1,108 U/day dose taken by patients in the nonenzyme-inducing group.

“Patients taking enzyme-inducing AEDs may benefit from more intensive monitoring of their vitamin D supplementation, and clinicians should anticipate this likely pharmacokinetic interaction,” said Barry E. Gidal, PharmD, professor of pharmacy and neurology at the University of Wisconsin–Madison, and his colleagues.

Researchers have suggested that enzyme-inducing AEDs may affect CYP450 isoenzymes, increase vitamin D metabolism, and reduce 25(OH)D plasma levels. “It follows … that a potential pharmacokinetic interaction could exist between enzyme-inducing AEDs and oral formulations of vitamin D used for supplementation,” the investigators said.

To test the hypothesis, Dr. Gidal and his colleagues reviewed the charts of patients with epilepsy who were on any AED regimen and were prescribed vitamin D at William S. Middleton Memorial Veterans Hospital in Madison, Wisconsin, between January 2013 and September 2017.

The researchers grouped patients by those using enzyme-inducing AEDs and those taking noninducing AEDs. Patients who were taking AEDs in both categories were placed in the enzyme-inducing AED group. Patients with malabsorptive conditions and patients using calcitriol were excluded from the analysis.

Data included AEDs used, prescription and over-the-counter vitamin D use, 25(OH)D plasma concentration, renal function, age, gender, and ethnicity. Patients’ 25(OH)D levels were measured using a chemiluminescence immunoassay, and a minimum 25(OH)D plasma level of 30 ng/mL was the therapeutic goal.

The multivariant analysis was adjusted for potentially confounding variables including 25(OH)D concentration, over-the-counter vitamin D use, chronic kidney disease, age, gender, and ethnicity.

The analysis included 1,113 observations from 315 patients, and 263 of the observations (23.6%) were in the enzyme-inducing AED group. The enzyme-inducing group and noninducing groups were mostly male (90.5% and 91.8%, respectively) and similar in average age (65.9 and 61.4 years, respectively). Variables were evenly distributed between the groups, with the exceptions of chronic kidney disease, which was less common in the enzyme-inducing group (6.1% vs. 13.8%), and ethnicity (78.7% Caucasian in the enzyme-inducing group vs. 87.7% Caucasian in the noninducing group). The most common enzyme-inducing AED was phenytoin (50.6%), followed by carbamazepine (31.9%), phenobarbital (14.1%), oxcarbazepine (6.8%), primidone (1.9%), and eslicarbazepine (0.8%).

Dr. Gidal reported honoraria from Eisai, Sunovion, Lundbeck, and GW Pharmaceuticals.
 

[email protected]

SOURCE: Gidal BE et al. AES 2018, Abstract 1.315.

NEW ORLEANS – Patients taking enzyme-inducing antiepileptic drugs (AEDs) may require a clinically meaningful increase in their vitamin D doses to achieve the same 25-hydroxyvitamin D (25[OH]D) plasma levels as patients taking nonenzyme-inducing AEDs, based on a retrospective chart review presented at the annual meeting of the American Epilepsy Society.

While patients receiving either type of AED had similar average 25(OH)D levels in the study (32.0 ng/mL in the enzyme-inducing AED group and 33.2 ng/mL in the noninducing AED group), those in the enzyme-inducing group required 1,587 U/day to meet the goal – a 409-unit increase in dose, compared with the 1,108 U/day dose taken by patients in the nonenzyme-inducing group.

“Patients taking enzyme-inducing AEDs may benefit from more intensive monitoring of their vitamin D supplementation, and clinicians should anticipate this likely pharmacokinetic interaction,” said Barry E. Gidal, PharmD, professor of pharmacy and neurology at the University of Wisconsin–Madison, and his colleagues.

Researchers have suggested that enzyme-inducing AEDs may affect CYP450 isoenzymes, increase vitamin D metabolism, and reduce 25(OH)D plasma levels. “It follows … that a potential pharmacokinetic interaction could exist between enzyme-inducing AEDs and oral formulations of vitamin D used for supplementation,” the investigators said.

To test the hypothesis, Dr. Gidal and his colleagues reviewed the charts of patients with epilepsy who were on any AED regimen and were prescribed vitamin D at William S. Middleton Memorial Veterans Hospital in Madison, Wisconsin, between January 2013 and September 2017.

The researchers grouped patients by those using enzyme-inducing AEDs and those taking noninducing AEDs. Patients who were taking AEDs in both categories were placed in the enzyme-inducing AED group. Patients with malabsorptive conditions and patients using calcitriol were excluded from the analysis.

Data included AEDs used, prescription and over-the-counter vitamin D use, 25(OH)D plasma concentration, renal function, age, gender, and ethnicity. Patients’ 25(OH)D levels were measured using a chemiluminescence immunoassay, and a minimum 25(OH)D plasma level of 30 ng/mL was the therapeutic goal.

The multivariant analysis was adjusted for potentially confounding variables including 25(OH)D concentration, over-the-counter vitamin D use, chronic kidney disease, age, gender, and ethnicity.

The analysis included 1,113 observations from 315 patients, and 263 of the observations (23.6%) were in the enzyme-inducing AED group. The enzyme-inducing group and noninducing groups were mostly male (90.5% and 91.8%, respectively) and similar in average age (65.9 and 61.4 years, respectively). Variables were evenly distributed between the groups, with the exceptions of chronic kidney disease, which was less common in the enzyme-inducing group (6.1% vs. 13.8%), and ethnicity (78.7% Caucasian in the enzyme-inducing group vs. 87.7% Caucasian in the noninducing group). The most common enzyme-inducing AED was phenytoin (50.6%), followed by carbamazepine (31.9%), phenobarbital (14.1%), oxcarbazepine (6.8%), primidone (1.9%), and eslicarbazepine (0.8%).

Dr. Gidal reported honoraria from Eisai, Sunovion, Lundbeck, and GW Pharmaceuticals.
 

[email protected]

SOURCE: Gidal BE et al. AES 2018, Abstract 1.315.

NEW ORLEANS – Clinicians should screen patients with pediatric epilepsy for depression regularly, according to research described at the annual meeting of the American Epilepsy Society. Referral to a mental health provider is adequate for most patients with moderately severe symptoms of depression, but some patients may require active intervention during the clinical visit, said the researchers.

“We know that depression is more common in people with epilepsy, compared to the general population, but there is less information about depression in children and teens than adults, and little is known about the factors that increase the likelihood of depressive symptoms,” said Hillary Thomas, PhD, a pediatric psychologist at Children’s Medical Center in Dallas. “Depression screening should be routine at epilepsy treatment centers and can identify children and teens who would benefit from intervention.”

Following 2015 guidelines from the American Academy of Neurology, the Comprehensive Epilepsy Center at Children’s Health System in Dallas developed a behavioral health screening protocol for teens with epilepsy. The center aims to identify patients with depressive symptoms and ensure that they are referred to appropriate behavioral health practitioners. Clinicians also review the screening data and seizure variables for their potential implications for clinical care. Researchers at the center also seek to elucidate the relationship between depressive symptoms and seizure diagnosis and treatment.

As part of the protocol, Dr. Thomas and her colleagues administer the Patient Health Questionnaire-9 (adolescent version) to all patients aged 15-18 years during their visit to the epilepsy clinic. Patients with intellectual disability or other factors that prevent them from providing valid responses are excluded. If a patient’s PHQ-9 score indicates at least moderately severe depressive symptoms, or if he or she reports suicidal ideation, clinicians follow a specific response protocol that includes providing referrals, encouraging follow-up with the patient’s current mental health provider, and obtaining a suicide risk assessment from a psychologist or social worker. After the screener is completed, clinicians retrieve demographic and clinical data (e.g., seizure diagnosis, medication, number of clinic or emergency department visits) from the patient’s medical record and include them in a database for subsequent analysis.

Dr. Thomas and her colleagues presented data from 394 youth with epilepsy whom they had screened. Patients’ mean age was 16 years, and half of the population was female. The study population had rates of depression similar to those identified in previous studies, said Dr. Thomas. Approximately 87% of patients had minimal or mild depressive symptoms, and 8% had moderately severe depressive symptoms. Furthermore, 5% of the patients reported suicidal ideation or previous suicide attempt. Several of the patients with suicidal ideation had a current mental health provider, and the others required an in-clinic risk assessment. Overall, 13% of the population required behavioral health referral or intervention. When the researchers conducted chi-squared analysis, they found no significant association between seizure type and depression severity.

“Our results don’t mean that only 13% of the teens with epilepsy had depressive symptoms,” said Susan Arnold, MD, director of the Comprehensive Epilepsy Center and a coauthor of the study. “They indicate the significant percentage of teens whose level of depressive symptoms warranted behavioral health referrals or further evaluation or even intervention during a clinic visit. Health care providers need to be vigilant about continually screening children and teens for depression.” As part of each patient’s comprehensive care, epilepsy treatment centers should provide psychosocial teams that include social workers or psychologists, she added.

The investigators plan to continue analyzing the data for specific depression symptoms that are most common in teens. These symptoms could be the basis for developing additional resources for families, such as lists of warning signs and guides to symptom management, as well as group therapy and support groups.
 

[email protected]

SOURCE: Thomas HM et al. Abstract 1.388.

NEW ORLEANS – Clinicians should screen patients with pediatric epilepsy for depression regularly, according to research described at the annual meeting of the American Epilepsy Society. Referral to a mental health provider is adequate for most patients with moderately severe symptoms of depression, but some patients may require active intervention during the clinical visit, said the researchers.

“We know that depression is more common in people with epilepsy, compared to the general population, but there is less information about depression in children and teens than adults, and little is known about the factors that increase the likelihood of depressive symptoms,” said Hillary Thomas, PhD, a pediatric psychologist at Children’s Medical Center in Dallas. “Depression screening should be routine at epilepsy treatment centers and can identify children and teens who would benefit from intervention.”

Following 2015 guidelines from the American Academy of Neurology, the Comprehensive Epilepsy Center at Children’s Health System in Dallas developed a behavioral health screening protocol for teens with epilepsy. The center aims to identify patients with depressive symptoms and ensure that they are referred to appropriate behavioral health practitioners. Clinicians also review the screening data and seizure variables for their potential implications for clinical care. Researchers at the center also seek to elucidate the relationship between depressive symptoms and seizure diagnosis and treatment.

As part of the protocol, Dr. Thomas and her colleagues administer the Patient Health Questionnaire-9 (adolescent version) to all patients aged 15-18 years during their visit to the epilepsy clinic. Patients with intellectual disability or other factors that prevent them from providing valid responses are excluded. If a patient’s PHQ-9 score indicates at least moderately severe depressive symptoms, or if he or she reports suicidal ideation, clinicians follow a specific response protocol that includes providing referrals, encouraging follow-up with the patient’s current mental health provider, and obtaining a suicide risk assessment from a psychologist or social worker. After the screener is completed, clinicians retrieve demographic and clinical data (e.g., seizure diagnosis, medication, number of clinic or emergency department visits) from the patient’s medical record and include them in a database for subsequent analysis.

Dr. Thomas and her colleagues presented data from 394 youth with epilepsy whom they had screened. Patients’ mean age was 16 years, and half of the population was female. The study population had rates of depression similar to those identified in previous studies, said Dr. Thomas. Approximately 87% of patients had minimal or mild depressive symptoms, and 8% had moderately severe depressive symptoms. Furthermore, 5% of the patients reported suicidal ideation or previous suicide attempt. Several of the patients with suicidal ideation had a current mental health provider, and the others required an in-clinic risk assessment. Overall, 13% of the population required behavioral health referral or intervention. When the researchers conducted chi-squared analysis, they found no significant association between seizure type and depression severity.

“Our results don’t mean that only 13% of the teens with epilepsy had depressive symptoms,” said Susan Arnold, MD, director of the Comprehensive Epilepsy Center and a coauthor of the study. “They indicate the significant percentage of teens whose level of depressive symptoms warranted behavioral health referrals or further evaluation or even intervention during a clinic visit. Health care providers need to be vigilant about continually screening children and teens for depression.” As part of each patient’s comprehensive care, epilepsy treatment centers should provide psychosocial teams that include social workers or psychologists, she added.

The investigators plan to continue analyzing the data for specific depression symptoms that are most common in teens. These symptoms could be the basis for developing additional resources for families, such as lists of warning signs and guides to symptom management, as well as group therapy and support groups.
 

[email protected]

SOURCE: Thomas HM et al. Abstract 1.388.

NEW ORLEANS – Clinicians should screen patients with pediatric epilepsy for depression regularly, according to research described at the annual meeting of the American Epilepsy Society. Referral to a mental health provider is adequate for most patients with moderately severe symptoms of depression, but some patients may require active intervention during the clinical visit, said the researchers.

“We know that depression is more common in people with epilepsy, compared to the general population, but there is less information about depression in children and teens than adults, and little is known about the factors that increase the likelihood of depressive symptoms,” said Hillary Thomas, PhD, a pediatric psychologist at Children’s Medical Center in Dallas. “Depression screening should be routine at epilepsy treatment centers and can identify children and teens who would benefit from intervention.”

Following 2015 guidelines from the American Academy of Neurology, the Comprehensive Epilepsy Center at Children’s Health System in Dallas developed a behavioral health screening protocol for teens with epilepsy. The center aims to identify patients with depressive symptoms and ensure that they are referred to appropriate behavioral health practitioners. Clinicians also review the screening data and seizure variables for their potential implications for clinical care. Researchers at the center also seek to elucidate the relationship between depressive symptoms and seizure diagnosis and treatment.

As part of the protocol, Dr. Thomas and her colleagues administer the Patient Health Questionnaire-9 (adolescent version) to all patients aged 15-18 years during their visit to the epilepsy clinic. Patients with intellectual disability or other factors that prevent them from providing valid responses are excluded. If a patient’s PHQ-9 score indicates at least moderately severe depressive symptoms, or if he or she reports suicidal ideation, clinicians follow a specific response protocol that includes providing referrals, encouraging follow-up with the patient’s current mental health provider, and obtaining a suicide risk assessment from a psychologist or social worker. After the screener is completed, clinicians retrieve demographic and clinical data (e.g., seizure diagnosis, medication, number of clinic or emergency department visits) from the patient’s medical record and include them in a database for subsequent analysis.

Dr. Thomas and her colleagues presented data from 394 youth with epilepsy whom they had screened. Patients’ mean age was 16 years, and half of the population was female. The study population had rates of depression similar to those identified in previous studies, said Dr. Thomas. Approximately 87% of patients had minimal or mild depressive symptoms, and 8% had moderately severe depressive symptoms. Furthermore, 5% of the patients reported suicidal ideation or previous suicide attempt. Several of the patients with suicidal ideation had a current mental health provider, and the others required an in-clinic risk assessment. Overall, 13% of the population required behavioral health referral or intervention. When the researchers conducted chi-squared analysis, they found no significant association between seizure type and depression severity.

“Our results don’t mean that only 13% of the teens with epilepsy had depressive symptoms,” said Susan Arnold, MD, director of the Comprehensive Epilepsy Center and a coauthor of the study. “They indicate the significant percentage of teens whose level of depressive symptoms warranted behavioral health referrals or further evaluation or even intervention during a clinic visit. Health care providers need to be vigilant about continually screening children and teens for depression.” As part of each patient’s comprehensive care, epilepsy treatment centers should provide psychosocial teams that include social workers or psychologists, she added.

The investigators plan to continue analyzing the data for specific depression symptoms that are most common in teens. These symptoms could be the basis for developing additional resources for families, such as lists of warning signs and guides to symptom management, as well as group therapy and support groups.
 

[email protected]

SOURCE: Thomas HM et al. Abstract 1.388.

One evening as the oncology fellow on call, I received a phone call from the ICU fellow.

“Can you meet me in the emergency room?” he asked. “I want to make sure we’re on the same page.”

A patient we had discharged from the hospital 2 days before was back. He had metastatic stomach cancer that had spread into his lungs and the lymph nodes in his chest. While he was in the hospital, he had required several liters of oxygen to maintain a normal work of breathing.

But now, he was in the emergency room, he was requiring a full face mask to help him breathe – and his oxygen levels were still dropping.

The ICU had been called. The next step along the algorithm of worsening breathing would be intubation. They would have to sedate him, put a breathing tube down his throat, and connect him to a ventilator to keep him alive.

But they didn’t want to do that if he was dying from his cancer.

Hence the call to me. My job, as the oncologist on call, was to answer the question: Is he dying?

Specifically, that meant weigh in on his cancer prognosis. Put his disease into context. Does he have any more options, chemotherapy or otherwise?

As an oncology fellow, I’ve found this to be one of the most common calls I get. Someone is critically ill and they need something to survive – maybe it’s intubation; maybe it’s surgery. The patient also happens to have metastatic cancer. The question posed to me is: Should we proceed?

It’s also one of the most difficult calls. Because doctors are historically bad at prognosticating. Because often I’m meeting the patient for the first time. Because the decision is huge and often final, and because both options are bad.

Suppose I say he has a good year or 2 ahead of him, and we intubate him – and then he never comes off the ventilator. We are eventually forced to withdraw care, and to the family it’s as though they are killing their father. It’s traumatic; it’s painful; and it deprives someone of a comfortable passing. Suppose I say he is dying from his cancer and we decide against a breathing tube. If I am wrong in that direction, a person’s life is cut short. It’s a perfect storm of high risk and low certainty.

Many people with metastatic cancer say they wouldn’t want invasive treatment near the end of life. But how do we know when it’s the end? There is still a moment when you must determine: Is this it? The truth is it’s not always clear.

Dr. Yurkiewicz is a fellow in hematology and oncology at Stanford (Calif.) University. Follow her on Twitter @ilanayurkiewicz.

One evening as the oncology fellow on call, I received a phone call from the ICU fellow.

“Can you meet me in the emergency room?” he asked. “I want to make sure we’re on the same page.”

A patient we had discharged from the hospital 2 days before was back. He had metastatic stomach cancer that had spread into his lungs and the lymph nodes in his chest. While he was in the hospital, he had required several liters of oxygen to maintain a normal work of breathing.

But now, he was in the emergency room, he was requiring a full face mask to help him breathe – and his oxygen levels were still dropping.

The ICU had been called. The next step along the algorithm of worsening breathing would be intubation. They would have to sedate him, put a breathing tube down his throat, and connect him to a ventilator to keep him alive.

But they didn’t want to do that if he was dying from his cancer.

Hence the call to me. My job, as the oncologist on call, was to answer the question: Is he dying?

Specifically, that meant weigh in on his cancer prognosis. Put his disease into context. Does he have any more options, chemotherapy or otherwise?

As an oncology fellow, I’ve found this to be one of the most common calls I get. Someone is critically ill and they need something to survive – maybe it’s intubation; maybe it’s surgery. The patient also happens to have metastatic cancer. The question posed to me is: Should we proceed?

It’s also one of the most difficult calls. Because doctors are historically bad at prognosticating. Because often I’m meeting the patient for the first time. Because the decision is huge and often final, and because both options are bad.

Suppose I say he has a good year or 2 ahead of him, and we intubate him – and then he never comes off the ventilator. We are eventually forced to withdraw care, and to the family it’s as though they are killing their father. It’s traumatic; it’s painful; and it deprives someone of a comfortable passing. Suppose I say he is dying from his cancer and we decide against a breathing tube. If I am wrong in that direction, a person’s life is cut short. It’s a perfect storm of high risk and low certainty.

Many people with metastatic cancer say they wouldn’t want invasive treatment near the end of life. But how do we know when it’s the end? There is still a moment when you must determine: Is this it? The truth is it’s not always clear.

Dr. Yurkiewicz is a fellow in hematology and oncology at Stanford (Calif.) University. Follow her on Twitter @ilanayurkiewicz.

One evening as the oncology fellow on call, I received a phone call from the ICU fellow.

“Can you meet me in the emergency room?” he asked. “I want to make sure we’re on the same page.”

A patient we had discharged from the hospital 2 days before was back. He had metastatic stomach cancer that had spread into his lungs and the lymph nodes in his chest. While he was in the hospital, he had required several liters of oxygen to maintain a normal work of breathing.

But now, he was in the emergency room, he was requiring a full face mask to help him breathe – and his oxygen levels were still dropping.

The ICU had been called. The next step along the algorithm of worsening breathing would be intubation. They would have to sedate him, put a breathing tube down his throat, and connect him to a ventilator to keep him alive.

But they didn’t want to do that if he was dying from his cancer.

Hence the call to me. My job, as the oncologist on call, was to answer the question: Is he dying?

Specifically, that meant weigh in on his cancer prognosis. Put his disease into context. Does he have any more options, chemotherapy or otherwise?

As an oncology fellow, I’ve found this to be one of the most common calls I get. Someone is critically ill and they need something to survive – maybe it’s intubation; maybe it’s surgery. The patient also happens to have metastatic cancer. The question posed to me is: Should we proceed?

It’s also one of the most difficult calls. Because doctors are historically bad at prognosticating. Because often I’m meeting the patient for the first time. Because the decision is huge and often final, and because both options are bad.

Suppose I say he has a good year or 2 ahead of him, and we intubate him – and then he never comes off the ventilator. We are eventually forced to withdraw care, and to the family it’s as though they are killing their father. It’s traumatic; it’s painful; and it deprives someone of a comfortable passing. Suppose I say he is dying from his cancer and we decide against a breathing tube. If I am wrong in that direction, a person’s life is cut short. It’s a perfect storm of high risk and low certainty.

Many people with metastatic cancer say they wouldn’t want invasive treatment near the end of life. But how do we know when it’s the end? There is still a moment when you must determine: Is this it? The truth is it’s not always clear.

Dr. Yurkiewicz is a fellow in hematology and oncology at Stanford (Calif.) University. Follow her on Twitter @ilanayurkiewicz.

After seven years at the helm of the Journal of Hospital Medicine, I am both pleased to hand over the reins and sad to let them go. My time as Editor in Chief has been wonderful, challenging, and fulfilling.

When I began my tenure, JHM managed approximately 350 papers annually, and published 10 times per year. We had no social media presence, a developing editorial sense (and developing Editor in Chief), and a pool of hard-working and passionate Editors. As of this year, we have handled more than 700 papers and are publishing content monthly, online only, and online first. Our dedicated team is deeply passionate about making every paper better through interaction with the authors—whether we accept it for publication or not.

JHM has added a presence on Facebook and Twitter, launched a Twitter Journal Club as a regular offering (#JHMChat), added visual abstracts to our Tweets and Facebook postings, and researched how these novel approaches increase not only the Journal’s social media presence but also its public face. Our efforts in social media were trendsetting in peer-reviewed literature, and the Editors who lead those efforts—Vineet Arora and Charlie Wray—are asked to consult for other journals regularly.

We launched two new series— Choosing Wisely®: Next Steps, and Choosing Wisely®: Things We Do For No Reason—with help from the ABIM Foundation and visionary Editors, Andy Masica, Ann Sheehy, and Lenny Feldman. These papers have pushed Hospitalists and Hospital Medicine to think carefully about the simple things we do every day, to think broadly about how to move past the initial ‘low-hanging fruit’ of value improvement, and point us towards policy and intervention approaches that are disruptive rather than incremental.

A special thank you to Som Mookherjee, Brian Harte, Dan Hunt, and Read Pierce who ably developed the Clinical Care Conundrums and Review series. They are assisted by teams of national correspondents and many contributors who’ve submitted work for those series.

I have been blessed by a team of more than a dozen Associate Editors who have ably, expeditiously, and collegially managed more than 2,000 papers. These Editors work out of a sense of altruism and commitment to Hospital Medicine and have made huge individual contributions to JHM through their reviewing expertise and ensuring that the editorial sense for JHM is as broad and innovative as our field.

Finally, I must thank my core team of Senior Deputy Editors who have shouldered the majority of editorial work, mentored Editors (including me) and Peer Reviewers, and provided strategic guidance.

How peer-reviewed journals are published is changing rapidly. Setting aside the questions of how we consume our medical literature and the transition from paper to digital, old financial models depending on subscriptions and advertising are either dying or evolving into something very different. The challenge is that the new model is very unclear and the old model based on ads and subscriptions is clearly nonviable but is the primary way to support the work of producing a journal. Moving from the current model to one based on clicks, views, or downloads will come down to who will derive benefit from those clicks/downloads, who will be willing to pay to read and learn from the work of authors, or who views that activity as being worthy enough advertise somewhere in that process or to monetize the data garnered from readers’ activities. In addition, many journals, including JHM, are supported by professional societies. While professional societies have a goal to serve their members, the goal of the peer-reviewed journal is to independently and broadly represent the field. One must reflect the other, but space between the two will always be required.

The speed with which research takes place is too slow, and the process of getting evidence into print (much less adopted) is even slower. But, this too is changing; the role of peer review and the publication process is evolving. In order to speed the potential discovery of new innovations, prepublication repositories (such as BioRxViv) are gaining popularity; well-publicized scandals around peer reviewing rings ¹ have not gone unnoticed, and have produced greater interest in using prepublication comments and online discussions as early forms of review. As a result, the disintermediation between scientist and ‘evidence’ is paralleling the disintermediation between events and messengers elsewhere. One need only review Twitter for a moment to get a sense for how crowdsourcing can lead to evidence (or news) generation for good or ill. I agree that while the end of journals (as we understand them now) is upon us, these are also opportunities for JHM as it enters its new phase and a place for leadership. ²

I am proud of what we have done at JHM in the last seven years. We have grown substantially. We have innovated and provided great service to our authors and the field of Hospital Medicine. Our growth and forward-looking approaches to social media and our digital footprint put the journal on a great path towards adapting to the trends in Hospital Medicine research and peer-reviewed publishing. Our focus on being doctors who care for patients and our teams—not just doctors who care for hospitals—is supporting the field and our practice. I look forward to seeing where JHM goes next.

After seven years at the helm of the Journal of Hospital Medicine, I am both pleased to hand over the reins and sad to let them go. My time as Editor in Chief has been wonderful, challenging, and fulfilling.

When I began my tenure, JHM managed approximately 350 papers annually, and published 10 times per year. We had no social media presence, a developing editorial sense (and developing Editor in Chief), and a pool of hard-working and passionate Editors. As of this year, we have handled more than 700 papers and are publishing content monthly, online only, and online first. Our dedicated team is deeply passionate about making every paper better through interaction with the authors—whether we accept it for publication or not.

JHM has added a presence on Facebook and Twitter, launched a Twitter Journal Club as a regular offering (#JHMChat), added visual abstracts to our Tweets and Facebook postings, and researched how these novel approaches increase not only the Journal’s social media presence but also its public face. Our efforts in social media were trendsetting in peer-reviewed literature, and the Editors who lead those efforts—Vineet Arora and Charlie Wray—are asked to consult for other journals regularly.

We launched two new series— Choosing Wisely®: Next Steps, and Choosing Wisely®: Things We Do For No Reason—with help from the ABIM Foundation and visionary Editors, Andy Masica, Ann Sheehy, and Lenny Feldman. These papers have pushed Hospitalists and Hospital Medicine to think carefully about the simple things we do every day, to think broadly about how to move past the initial ‘low-hanging fruit’ of value improvement, and point us towards policy and intervention approaches that are disruptive rather than incremental.

A special thank you to Som Mookherjee, Brian Harte, Dan Hunt, and Read Pierce who ably developed the Clinical Care Conundrums and Review series. They are assisted by teams of national correspondents and many contributors who’ve submitted work for those series.

I have been blessed by a team of more than a dozen Associate Editors who have ably, expeditiously, and collegially managed more than 2,000 papers. These Editors work out of a sense of altruism and commitment to Hospital Medicine and have made huge individual contributions to JHM through their reviewing expertise and ensuring that the editorial sense for JHM is as broad and innovative as our field.

Finally, I must thank my core team of Senior Deputy Editors who have shouldered the majority of editorial work, mentored Editors (including me) and Peer Reviewers, and provided strategic guidance.

How peer-reviewed journals are published is changing rapidly. Setting aside the questions of how we consume our medical literature and the transition from paper to digital, old financial models depending on subscriptions and advertising are either dying or evolving into something very different. The challenge is that the new model is very unclear and the old model based on ads and subscriptions is clearly nonviable but is the primary way to support the work of producing a journal. Moving from the current model to one based on clicks, views, or downloads will come down to who will derive benefit from those clicks/downloads, who will be willing to pay to read and learn from the work of authors, or who views that activity as being worthy enough advertise somewhere in that process or to monetize the data garnered from readers’ activities. In addition, many journals, including JHM, are supported by professional societies. While professional societies have a goal to serve their members, the goal of the peer-reviewed journal is to independently and broadly represent the field. One must reflect the other, but space between the two will always be required.

The speed with which research takes place is too slow, and the process of getting evidence into print (much less adopted) is even slower. But, this too is changing; the role of peer review and the publication process is evolving. In order to speed the potential discovery of new innovations, prepublication repositories (such as BioRxViv) are gaining popularity; well-publicized scandals around peer reviewing rings ¹ have not gone unnoticed, and have produced greater interest in using prepublication comments and online discussions as early forms of review. As a result, the disintermediation between scientist and ‘evidence’ is paralleling the disintermediation between events and messengers elsewhere. One need only review Twitter for a moment to get a sense for how crowdsourcing can lead to evidence (or news) generation for good or ill. I agree that while the end of journals (as we understand them now) is upon us, these are also opportunities for JHM as it enters its new phase and a place for leadership. ²

I am proud of what we have done at JHM in the last seven years. We have grown substantially. We have innovated and provided great service to our authors and the field of Hospital Medicine. Our growth and forward-looking approaches to social media and our digital footprint put the journal on a great path towards adapting to the trends in Hospital Medicine research and peer-reviewed publishing. Our focus on being doctors who care for patients and our teams—not just doctors who care for hospitals—is supporting the field and our practice. I look forward to seeing where JHM goes next.

After seven years at the helm of the Journal of Hospital Medicine, I am both pleased to hand over the reins and sad to let them go. My time as Editor in Chief has been wonderful, challenging, and fulfilling.

When I began my tenure, JHM managed approximately 350 papers annually, and published 10 times per year. We had no social media presence, a developing editorial sense (and developing Editor in Chief), and a pool of hard-working and passionate Editors. As of this year, we have handled more than 700 papers and are publishing content monthly, online only, and online first. Our dedicated team is deeply passionate about making every paper better through interaction with the authors—whether we accept it for publication or not.

JHM has added a presence on Facebook and Twitter, launched a Twitter Journal Club as a regular offering (#JHMChat), added visual abstracts to our Tweets and Facebook postings, and researched how these novel approaches increase not only the Journal’s social media presence but also its public face. Our efforts in social media were trendsetting in peer-reviewed literature, and the Editors who lead those efforts—Vineet Arora and Charlie Wray—are asked to consult for other journals regularly.

We launched two new series— Choosing Wisely®: Next Steps, and Choosing Wisely®: Things We Do For No Reason—with help from the ABIM Foundation and visionary Editors, Andy Masica, Ann Sheehy, and Lenny Feldman. These papers have pushed Hospitalists and Hospital Medicine to think carefully about the simple things we do every day, to think broadly about how to move past the initial ‘low-hanging fruit’ of value improvement, and point us towards policy and intervention approaches that are disruptive rather than incremental.

A special thank you to Som Mookherjee, Brian Harte, Dan Hunt, and Read Pierce who ably developed the Clinical Care Conundrums and Review series. They are assisted by teams of national correspondents and many contributors who’ve submitted work for those series.

I have been blessed by a team of more than a dozen Associate Editors who have ably, expeditiously, and collegially managed more than 2,000 papers. These Editors work out of a sense of altruism and commitment to Hospital Medicine and have made huge individual contributions to JHM through their reviewing expertise and ensuring that the editorial sense for JHM is as broad and innovative as our field.

Finally, I must thank my core team of Senior Deputy Editors who have shouldered the majority of editorial work, mentored Editors (including me) and Peer Reviewers, and provided strategic guidance.

How peer-reviewed journals are published is changing rapidly. Setting aside the questions of how we consume our medical literature and the transition from paper to digital, old financial models depending on subscriptions and advertising are either dying or evolving into something very different. The challenge is that the new model is very unclear and the old model based on ads and subscriptions is clearly nonviable but is the primary way to support the work of producing a journal. Moving from the current model to one based on clicks, views, or downloads will come down to who will derive benefit from those clicks/downloads, who will be willing to pay to read and learn from the work of authors, or who views that activity as being worthy enough advertise somewhere in that process or to monetize the data garnered from readers’ activities. In addition, many journals, including JHM, are supported by professional societies. While professional societies have a goal to serve their members, the goal of the peer-reviewed journal is to independently and broadly represent the field. One must reflect the other, but space between the two will always be required.

The speed with which research takes place is too slow, and the process of getting evidence into print (much less adopted) is even slower. But, this too is changing; the role of peer review and the publication process is evolving. In order to speed the potential discovery of new innovations, prepublication repositories (such as BioRxViv) are gaining popularity; well-publicized scandals around peer reviewing rings ¹ have not gone unnoticed, and have produced greater interest in using prepublication comments and online discussions as early forms of review. As a result, the disintermediation between scientist and ‘evidence’ is paralleling the disintermediation between events and messengers elsewhere. One need only review Twitter for a moment to get a sense for how crowdsourcing can lead to evidence (or news) generation for good or ill. I agree that while the end of journals (as we understand them now) is upon us, these are also opportunities for JHM as it enters its new phase and a place for leadership. ²

I am proud of what we have done at JHM in the last seven years. We have grown substantially. We have innovated and provided great service to our authors and the field of Hospital Medicine. Our growth and forward-looking approaches to social media and our digital footprint put the journal on a great path towards adapting to the trends in Hospital Medicine research and peer-reviewed publishing. Our focus on being doctors who care for patients and our teams—not just doctors who care for hospitals—is supporting the field and our practice. I look forward to seeing where JHM goes next.

Hospital discharges frequently occur in the afternoon or evening hours.^1-5 Late discharges can adversely affect patient flow throughout the hospital,^3,6-9 which, in turn, can result in delays in care,^10-16 more medication errors,¹⁷ increased mortality,^18-20 longer lengths of stay,^20-22 higher costs,²³ and lower patient satisfaction.²⁴

Various interventions have been employed in the attempts to find ways of moving discharge times to earlier in the day, including preparing the discharge paperwork and medications the previous night,²⁵ using checklists,^1,25 team huddles,² providing real-time feedback to unit staff,¹ and employing multidisciplinary teamwork.^1,2,6,25,26

The purpose of this study was to identify and determine the relative frequency of barriers to writing discharge orders in the hopes of identifying issues that might be addressed by targeted interventions. We also assessed the effects of daily team census, patients being on teaching versus nonteaching services, and how daily rounds were structured at the time that the discharge orders were written.

Study Design, Setting, and Participants

We conducted a prospective, cross-sectional survey of house-staff and attending physicians on general medicine teaching and nonteaching services from November 13, 2014, through May 31, 2016. The study was conducted at the following five hospitals: Denver Health Medical Center (DHMC) and Presbyterian/Saint Luke’s Medical Center (PSL) in Denver, Colorado; Ronald Reagan University (UCLA) and Los Angeles County/University of Southern California Medical Center (LAC+USC) in Los Angeles, California; and Harborview Medical Center (HMC) in Seattle, Washington. The study was approved by the Colorado Multi-Institutional Review Board as well as by the review boards of the other participating sites.

Data Collection

The results of the focus groups composed of attending physicians at DHMC were used to develop our initial data collection template. Additional sites joining the study provided feedback, leading to modifications (Appendix 1).

Physicians were surveyed at three different time points on study days that were selected according to the convenience of the investigators. The sampling occurred only on weekdays and was done based on the investigators’ availability. Investigators would attempt to survey as many teams as they were able to but, secondary to feasibility, not all teams could be surveyed on study days. The specific time points varied as a function of physician workflows but were standardized as much as possible to occur in the early morning, around noon, and midafternoon on weekdays. Physicians were contacted either in person or by telephone for verbal consent prior to administering the first survey. All general medicine teams were eligible. For teaching teams, the order of contact was resident, intern, and then attending based on which physician was available at the time of the survey and on which member of the team was thought to know the patients the best. For the nonteaching services, the attending physicians were contacted.

During the initial survey, the investigators assessed the provider role (ie, attending or housestaff), whether the service was a teaching or a nonteaching service, and the starting patient census on that service primarily based on interviewing the provider of record for the team and looking at team census lists. Physicians were asked about their rounding style (ie, sickest patients first, patients likely to be discharged first, room-by-room, most recently admitted patients first, patients on the team the longest, or other) and then to identify all patients they thought would be definite discharges sometime during the day of the survey. Definite discharges were defined as patients whom the provider thought were either currently ready for discharge or who had only minor barriers that, if unresolved, would not prevent same-day discharge. They were asked if the discharge order had been entered and, if not, what was preventing them from doing so, if the discharge could in their opinion have occurred the day prior and, if so, why this did not occur. We also obtained the date and time of the admission and discharge orders, the actual discharge time, as well as the length of stay either through chart review (majority of sites) or from data warehouses (Denver Health and Presbyterian St. Lukes had length of stay data retrieved from their data warehouse).

Physicians were also asked to identify all patients whom they thought might possibly be discharged that day. Possible discharges were defined as patients with barriers to discharge that, if unresolved, would prevent same-day discharge. For each of these, the physicians were asked to list whatever issues needed to be resolved prior to placing the discharge order (Appendix 1).

The second survey was administered late morning on the same day, typically between 11 am and 12 pm. In this survey, the physicians were asked to reassess the patients previously classified as definite and possible discharges for changes in status and/or barriers and to identify patients who had become definite or possible discharges since the earlier survey. Newly identified possible or definite discharges were evaluated in a similar manner as the initial survey.

The third survey was administered midafternoon, typically around 3 PM similar to the first two surveys, with the exception that the third survey did not attempt to identify new definite or possible discharges.

Sample Size

We stopped collecting data after obtaining a convenience sample of 5% of total discharges at each study site or on the study end date, which was May 31, 2016, whichever came first.

Data Analysis

Data were collected and managed using a secure, web-based application electronic data capture tool (REDCap), hosted at Denver Health. REDCap (Research Electronic Data Capture, Nashville, Tennessee) is designed to support data collection for research studies.²⁷ Data were then analyzed using SAS Enterprise Guide 5.1 (SAS Institute, Inc., Cary, North Carolina). All data entered into REDCap were reviewed by the principal investigator to ensure that data were not missing, and when there were missing data, a query was sent to verify if the data were retrievable. If retrievable, then the data would be entered. The volume of missing data that remained is described in our results.

Continuous variables were described using means and standard deviations (SD) or medians and interquartile ranges (IQR) based on tests of normality. Differences in the time that the discharge orders were placed in the electronic medical record according to morning patient census, teaching versus nonteaching service, and rounding style were compared using the Wilcoxon rank sum test. Linear regression was used to evaluate the effect of patient census on discharge order time. P < .05 was considered as significant.

We conducted 1,584 patient evaluations through surveys of 254 physicians over 156 days. Given surveys coincided with the existing work we had full participation (ie, 100% participation) and no dropout during the study days. Median (IQR) survey time points were 8:30 am (7:51 am, 9:12 am), 11:45 am (11:30 am, 12:17 pm), and 3:20 pm (3:00 pm, 4:06 pm).

The characteristics of the five hospitals participating in the study, the patients’ final discharge status, the types of physicians surveyed, the services on which they were working, the rounding styles employed, and the median starting daily census are summarized in Table 1. The majority of the physicians surveyed were housestaff working on teaching services, and only a small minority structured rounds such that patients ready for discharge were seen first.

The important findings of this study are that (1) the large majority of issues thought to delay discharging patients identified as definite discharges were related to physicians caring for other patients on their team, (2) although 91% of patients ever identified as being definite discharges were discharged on the day of the survey, only 48% of those identified as possible discharges became definite discharges by the afternoon time point, largely because the anticipated clinical improvement did not occur or care being provided by ancillary services had not been completed, and (3) discharge orders on patients identified as definite discharges were written on average 50 minutes earlier by physicians on teams with a smaller starting patient census, on nonteaching services, or when the rounding style was to see patients ready for discharges first.

Previous research has reported that physician-perceived barriers to discharge were extrinsic to providers and even extrinsic to the hospital setting (eg, awaiting subacute nursing placement and transportation).^28,29 However, many of the barriers that we identified were related directly to the providers’ workload and rounding styles and whether the patients were on teaching versus nonteaching services. We also found that delays in the ability of hospital services to complete care also contributed to delayed discharges.

Our observational data suggest that delays resulting from caring for other patients might be reduced by changing rounding styles such that patients ready for discharge are seen first and are discharged prior to seeing other patients on the team, as previously reported by Beck et al.³⁰ Intuitively, this would seem to be a straightforward way of freeing up beds earlier in the day, but such a change will, of necessity, lead to delaying care for other patients, which, in turn, could increase their length of stays. Durvasula et al. suggested that discharges could be moved to earlier in the day by completing orders and paperwork the day prior to discharge.²⁵ Such an approach might be effective on an Obstetrical or elective Orthopedic service on which patients predictably are hospitalized for a fixed number of days (or even hours) but may be less relevant to patients on internal medicine services where lengths of stay are less predictable. Interventions to improve discharge times have resulted in earlier discharge times in some studies,^2,4 but the overall length of stay either did not decrease²⁵ or increased³¹ in others. Werthheimer et al.¹ did find earlier discharge times, but other interventions also occurred during the study period (eg, extending social work services to include weekends).^1,32

We found that discharge times were approximately 50 minutes earlier on teams with a smaller starting census, on nonteaching compared with teaching services, or when the attending’s rounding style was to see patients ready for discharges first. Although 50 minutes may seem like a small change in discharge time, Khanna et al.³³ found that when discharges occur even 1 hour earlier, hospital overcrowding is reduced. To have a lower team census would require having more teams and more providers to staff these teams, raising cost-effectiveness concerns. Moving to more nonteaching services could represent a conflict with respect to one of the missions of teaching hospitals and raises a cost-benefit issue as several teaching hospitals receive substantial funding in support of their teaching activities and housestaff would have to be replaced with more expensive providers.

Delays attributable to ancillary services indicate imbalances between demand and availability of these services. Inappropriate demand and inefficiencies could be reduced by systems redesign, but in at least some instances, additional resources will be needed to add staff, increase space, or add additional equipment.

Our study has several limitations. First, we surveyed only physicians working in university-affiliated hospitals, and three of these were public safety-net hospitals. Accordingly, our results may not be generalizable to different patient populations. Second, we surveyed only physicians, and Minichiello et al.²⁹ found that barriers to discharge perceived by physicians were different from those of other staff. Third, our data were observational and were collected only on weekdays. Fourth, we did not differentiate interns from residents, and thus, potentially the level of training could have affected these results. Similarly, the decision for a “possible” and a “definite” discharge is likely dependent on the knowledge base of the participant, such that less experienced participants may have had differing perspectives than someone with more experience. Fifth, the sites did vary based on the infrastructure and support but also had several similarities. All sites had social work and case management involved in care, although at some sites, they were assigned according to team and at others according to geographic location. Similarly, rounding times varied. Most of the services surveyed did not utilize advanced practice providers (the exception was the nonteaching services at Denver Health, and their presence was variable). These differences in staffing models could also have affected these results.

Our study also has a number of strengths. First, we assessed the barriers at five different hospitals. Second, we collected real-time data related to specific barriers at multiple time points throughout the day, allowing us to assess the dynamic nature of identifying patients as being ready or nearly ready for discharge. Third, we assessed the perceptions of barriers to discharge from physicians working on teaching as well as nonteaching services and from physicians utilizing a variety of rounding styles. Fourth, we had a very high participation rate (100%), probably due to the fact that our study was strategically aligned with participants’ daily work activities.

In conclusion, we found two distinct categories of issues that physicians perceived as most commonly delaying writing discharge orders on their patients. The first pertained to patients thought to definitely be ready for discharge and was related to the physicians having to care for other patients on their team. The second pertained to patients identified as possibly ready for discharge and was related to the need for care to be completed by a variety of ancillary services. Addressing each of these barriers would require different interventions and a need to weigh the potential improvements that could be achieved against the increased costs and/or delays in care for other patients that may result.

Disclosures

The authors report no conflicts of interest relevant to this work.

Hospital discharges frequently occur in the afternoon or evening hours.^1-5 Late discharges can adversely affect patient flow throughout the hospital,^3,6-9 which, in turn, can result in delays in care,^10-16 more medication errors,¹⁷ increased mortality,^18-20 longer lengths of stay,^20-22 higher costs,²³ and lower patient satisfaction.²⁴

Various interventions have been employed in the attempts to find ways of moving discharge times to earlier in the day, including preparing the discharge paperwork and medications the previous night,²⁵ using checklists,^1,25 team huddles,² providing real-time feedback to unit staff,¹ and employing multidisciplinary teamwork.^1,2,6,25,26

The purpose of this study was to identify and determine the relative frequency of barriers to writing discharge orders in the hopes of identifying issues that might be addressed by targeted interventions. We also assessed the effects of daily team census, patients being on teaching versus nonteaching services, and how daily rounds were structured at the time that the discharge orders were written.

Study Design, Setting, and Participants

We conducted a prospective, cross-sectional survey of house-staff and attending physicians on general medicine teaching and nonteaching services from November 13, 2014, through May 31, 2016. The study was conducted at the following five hospitals: Denver Health Medical Center (DHMC) and Presbyterian/Saint Luke’s Medical Center (PSL) in Denver, Colorado; Ronald Reagan University (UCLA) and Los Angeles County/University of Southern California Medical Center (LAC+USC) in Los Angeles, California; and Harborview Medical Center (HMC) in Seattle, Washington. The study was approved by the Colorado Multi-Institutional Review Board as well as by the review boards of the other participating sites.

Data Collection

The results of the focus groups composed of attending physicians at DHMC were used to develop our initial data collection template. Additional sites joining the study provided feedback, leading to modifications (Appendix 1).

Physicians were surveyed at three different time points on study days that were selected according to the convenience of the investigators. The sampling occurred only on weekdays and was done based on the investigators’ availability. Investigators would attempt to survey as many teams as they were able to but, secondary to feasibility, not all teams could be surveyed on study days. The specific time points varied as a function of physician workflows but were standardized as much as possible to occur in the early morning, around noon, and midafternoon on weekdays. Physicians were contacted either in person or by telephone for verbal consent prior to administering the first survey. All general medicine teams were eligible. For teaching teams, the order of contact was resident, intern, and then attending based on which physician was available at the time of the survey and on which member of the team was thought to know the patients the best. For the nonteaching services, the attending physicians were contacted.

During the initial survey, the investigators assessed the provider role (ie, attending or housestaff), whether the service was a teaching or a nonteaching service, and the starting patient census on that service primarily based on interviewing the provider of record for the team and looking at team census lists. Physicians were asked about their rounding style (ie, sickest patients first, patients likely to be discharged first, room-by-room, most recently admitted patients first, patients on the team the longest, or other) and then to identify all patients they thought would be definite discharges sometime during the day of the survey. Definite discharges were defined as patients whom the provider thought were either currently ready for discharge or who had only minor barriers that, if unresolved, would not prevent same-day discharge. They were asked if the discharge order had been entered and, if not, what was preventing them from doing so, if the discharge could in their opinion have occurred the day prior and, if so, why this did not occur. We also obtained the date and time of the admission and discharge orders, the actual discharge time, as well as the length of stay either through chart review (majority of sites) or from data warehouses (Denver Health and Presbyterian St. Lukes had length of stay data retrieved from their data warehouse).

Physicians were also asked to identify all patients whom they thought might possibly be discharged that day. Possible discharges were defined as patients with barriers to discharge that, if unresolved, would prevent same-day discharge. For each of these, the physicians were asked to list whatever issues needed to be resolved prior to placing the discharge order (Appendix 1).

The second survey was administered late morning on the same day, typically between 11 am and 12 pm. In this survey, the physicians were asked to reassess the patients previously classified as definite and possible discharges for changes in status and/or barriers and to identify patients who had become definite or possible discharges since the earlier survey. Newly identified possible or definite discharges were evaluated in a similar manner as the initial survey.

The third survey was administered midafternoon, typically around 3 PM similar to the first two surveys, with the exception that the third survey did not attempt to identify new definite or possible discharges.

Sample Size

We stopped collecting data after obtaining a convenience sample of 5% of total discharges at each study site or on the study end date, which was May 31, 2016, whichever came first.

Data Analysis

Data were collected and managed using a secure, web-based application electronic data capture tool (REDCap), hosted at Denver Health. REDCap (Research Electronic Data Capture, Nashville, Tennessee) is designed to support data collection for research studies.²⁷ Data were then analyzed using SAS Enterprise Guide 5.1 (SAS Institute, Inc., Cary, North Carolina). All data entered into REDCap were reviewed by the principal investigator to ensure that data were not missing, and when there were missing data, a query was sent to verify if the data were retrievable. If retrievable, then the data would be entered. The volume of missing data that remained is described in our results.

Continuous variables were described using means and standard deviations (SD) or medians and interquartile ranges (IQR) based on tests of normality. Differences in the time that the discharge orders were placed in the electronic medical record according to morning patient census, teaching versus nonteaching service, and rounding style were compared using the Wilcoxon rank sum test. Linear regression was used to evaluate the effect of patient census on discharge order time. P < .05 was considered as significant.

We conducted 1,584 patient evaluations through surveys of 254 physicians over 156 days. Given surveys coincided with the existing work we had full participation (ie, 100% participation) and no dropout during the study days. Median (IQR) survey time points were 8:30 am (7:51 am, 9:12 am), 11:45 am (11:30 am, 12:17 pm), and 3:20 pm (3:00 pm, 4:06 pm).

The characteristics of the five hospitals participating in the study, the patients’ final discharge status, the types of physicians surveyed, the services on which they were working, the rounding styles employed, and the median starting daily census are summarized in Table 1. The majority of the physicians surveyed were housestaff working on teaching services, and only a small minority structured rounds such that patients ready for discharge were seen first.

The important findings of this study are that (1) the large majority of issues thought to delay discharging patients identified as definite discharges were related to physicians caring for other patients on their team, (2) although 91% of patients ever identified as being definite discharges were discharged on the day of the survey, only 48% of those identified as possible discharges became definite discharges by the afternoon time point, largely because the anticipated clinical improvement did not occur or care being provided by ancillary services had not been completed, and (3) discharge orders on patients identified as definite discharges were written on average 50 minutes earlier by physicians on teams with a smaller starting patient census, on nonteaching services, or when the rounding style was to see patients ready for discharges first.

Previous research has reported that physician-perceived barriers to discharge were extrinsic to providers and even extrinsic to the hospital setting (eg, awaiting subacute nursing placement and transportation).^28,29 However, many of the barriers that we identified were related directly to the providers’ workload and rounding styles and whether the patients were on teaching versus nonteaching services. We also found that delays in the ability of hospital services to complete care also contributed to delayed discharges.

Our observational data suggest that delays resulting from caring for other patients might be reduced by changing rounding styles such that patients ready for discharge are seen first and are discharged prior to seeing other patients on the team, as previously reported by Beck et al.³⁰ Intuitively, this would seem to be a straightforward way of freeing up beds earlier in the day, but such a change will, of necessity, lead to delaying care for other patients, which, in turn, could increase their length of stays. Durvasula et al. suggested that discharges could be moved to earlier in the day by completing orders and paperwork the day prior to discharge.²⁵ Such an approach might be effective on an Obstetrical or elective Orthopedic service on which patients predictably are hospitalized for a fixed number of days (or even hours) but may be less relevant to patients on internal medicine services where lengths of stay are less predictable. Interventions to improve discharge times have resulted in earlier discharge times in some studies,^2,4 but the overall length of stay either did not decrease²⁵ or increased³¹ in others. Werthheimer et al.¹ did find earlier discharge times, but other interventions also occurred during the study period (eg, extending social work services to include weekends).^1,32

We found that discharge times were approximately 50 minutes earlier on teams with a smaller starting census, on nonteaching compared with teaching services, or when the attending’s rounding style was to see patients ready for discharges first. Although 50 minutes may seem like a small change in discharge time, Khanna et al.³³ found that when discharges occur even 1 hour earlier, hospital overcrowding is reduced. To have a lower team census would require having more teams and more providers to staff these teams, raising cost-effectiveness concerns. Moving to more nonteaching services could represent a conflict with respect to one of the missions of teaching hospitals and raises a cost-benefit issue as several teaching hospitals receive substantial funding in support of their teaching activities and housestaff would have to be replaced with more expensive providers.

Delays attributable to ancillary services indicate imbalances between demand and availability of these services. Inappropriate demand and inefficiencies could be reduced by systems redesign, but in at least some instances, additional resources will be needed to add staff, increase space, or add additional equipment.

Our study has several limitations. First, we surveyed only physicians working in university-affiliated hospitals, and three of these were public safety-net hospitals. Accordingly, our results may not be generalizable to different patient populations. Second, we surveyed only physicians, and Minichiello et al.²⁹ found that barriers to discharge perceived by physicians were different from those of other staff. Third, our data were observational and were collected only on weekdays. Fourth, we did not differentiate interns from residents, and thus, potentially the level of training could have affected these results. Similarly, the decision for a “possible” and a “definite” discharge is likely dependent on the knowledge base of the participant, such that less experienced participants may have had differing perspectives than someone with more experience. Fifth, the sites did vary based on the infrastructure and support but also had several similarities. All sites had social work and case management involved in care, although at some sites, they were assigned according to team and at others according to geographic location. Similarly, rounding times varied. Most of the services surveyed did not utilize advanced practice providers (the exception was the nonteaching services at Denver Health, and their presence was variable). These differences in staffing models could also have affected these results.

Our study also has a number of strengths. First, we assessed the barriers at five different hospitals. Second, we collected real-time data related to specific barriers at multiple time points throughout the day, allowing us to assess the dynamic nature of identifying patients as being ready or nearly ready for discharge. Third, we assessed the perceptions of barriers to discharge from physicians working on teaching as well as nonteaching services and from physicians utilizing a variety of rounding styles. Fourth, we had a very high participation rate (100%), probably due to the fact that our study was strategically aligned with participants’ daily work activities.

In conclusion, we found two distinct categories of issues that physicians perceived as most commonly delaying writing discharge orders on their patients. The first pertained to patients thought to definitely be ready for discharge and was related to the physicians having to care for other patients on their team. The second pertained to patients identified as possibly ready for discharge and was related to the need for care to be completed by a variety of ancillary services. Addressing each of these barriers would require different interventions and a need to weigh the potential improvements that could be achieved against the increased costs and/or delays in care for other patients that may result.

Disclosures

The authors report no conflicts of interest relevant to this work.

Hospital discharges frequently occur in the afternoon or evening hours.^1-5 Late discharges can adversely affect patient flow throughout the hospital,^3,6-9 which, in turn, can result in delays in care,^10-16 more medication errors,¹⁷ increased mortality,^18-20 longer lengths of stay,^20-22 higher costs,²³ and lower patient satisfaction.²⁴

Various interventions have been employed in the attempts to find ways of moving discharge times to earlier in the day, including preparing the discharge paperwork and medications the previous night,²⁵ using checklists,^1,25 team huddles,² providing real-time feedback to unit staff,¹ and employing multidisciplinary teamwork.^1,2,6,25,26

The purpose of this study was to identify and determine the relative frequency of barriers to writing discharge orders in the hopes of identifying issues that might be addressed by targeted interventions. We also assessed the effects of daily team census, patients being on teaching versus nonteaching services, and how daily rounds were structured at the time that the discharge orders were written.

Study Design, Setting, and Participants

We conducted a prospective, cross-sectional survey of house-staff and attending physicians on general medicine teaching and nonteaching services from November 13, 2014, through May 31, 2016. The study was conducted at the following five hospitals: Denver Health Medical Center (DHMC) and Presbyterian/Saint Luke’s Medical Center (PSL) in Denver, Colorado; Ronald Reagan University (UCLA) and Los Angeles County/University of Southern California Medical Center (LAC+USC) in Los Angeles, California; and Harborview Medical Center (HMC) in Seattle, Washington. The study was approved by the Colorado Multi-Institutional Review Board as well as by the review boards of the other participating sites.

Data Collection

The results of the focus groups composed of attending physicians at DHMC were used to develop our initial data collection template. Additional sites joining the study provided feedback, leading to modifications (Appendix 1).

Physicians were surveyed at three different time points on study days that were selected according to the convenience of the investigators. The sampling occurred only on weekdays and was done based on the investigators’ availability. Investigators would attempt to survey as many teams as they were able to but, secondary to feasibility, not all teams could be surveyed on study days. The specific time points varied as a function of physician workflows but were standardized as much as possible to occur in the early morning, around noon, and midafternoon on weekdays. Physicians were contacted either in person or by telephone for verbal consent prior to administering the first survey. All general medicine teams were eligible. For teaching teams, the order of contact was resident, intern, and then attending based on which physician was available at the time of the survey and on which member of the team was thought to know the patients the best. For the nonteaching services, the attending physicians were contacted.

During the initial survey, the investigators assessed the provider role (ie, attending or housestaff), whether the service was a teaching or a nonteaching service, and the starting patient census on that service primarily based on interviewing the provider of record for the team and looking at team census lists. Physicians were asked about their rounding style (ie, sickest patients first, patients likely to be discharged first, room-by-room, most recently admitted patients first, patients on the team the longest, or other) and then to identify all patients they thought would be definite discharges sometime during the day of the survey. Definite discharges were defined as patients whom the provider thought were either currently ready for discharge or who had only minor barriers that, if unresolved, would not prevent same-day discharge. They were asked if the discharge order had been entered and, if not, what was preventing them from doing so, if the discharge could in their opinion have occurred the day prior and, if so, why this did not occur. We also obtained the date and time of the admission and discharge orders, the actual discharge time, as well as the length of stay either through chart review (majority of sites) or from data warehouses (Denver Health and Presbyterian St. Lukes had length of stay data retrieved from their data warehouse).

Physicians were also asked to identify all patients whom they thought might possibly be discharged that day. Possible discharges were defined as patients with barriers to discharge that, if unresolved, would prevent same-day discharge. For each of these, the physicians were asked to list whatever issues needed to be resolved prior to placing the discharge order (Appendix 1).

The second survey was administered late morning on the same day, typically between 11 am and 12 pm. In this survey, the physicians were asked to reassess the patients previously classified as definite and possible discharges for changes in status and/or barriers and to identify patients who had become definite or possible discharges since the earlier survey. Newly identified possible or definite discharges were evaluated in a similar manner as the initial survey.

The third survey was administered midafternoon, typically around 3 PM similar to the first two surveys, with the exception that the third survey did not attempt to identify new definite or possible discharges.

Sample Size

We stopped collecting data after obtaining a convenience sample of 5% of total discharges at each study site or on the study end date, which was May 31, 2016, whichever came first.

Data Analysis

Data were collected and managed using a secure, web-based application electronic data capture tool (REDCap), hosted at Denver Health. REDCap (Research Electronic Data Capture, Nashville, Tennessee) is designed to support data collection for research studies.²⁷ Data were then analyzed using SAS Enterprise Guide 5.1 (SAS Institute, Inc., Cary, North Carolina). All data entered into REDCap were reviewed by the principal investigator to ensure that data were not missing, and when there were missing data, a query was sent to verify if the data were retrievable. If retrievable, then the data would be entered. The volume of missing data that remained is described in our results.

Continuous variables were described using means and standard deviations (SD) or medians and interquartile ranges (IQR) based on tests of normality. Differences in the time that the discharge orders were placed in the electronic medical record according to morning patient census, teaching versus nonteaching service, and rounding style were compared using the Wilcoxon rank sum test. Linear regression was used to evaluate the effect of patient census on discharge order time. P < .05 was considered as significant.

We conducted 1,584 patient evaluations through surveys of 254 physicians over 156 days. Given surveys coincided with the existing work we had full participation (ie, 100% participation) and no dropout during the study days. Median (IQR) survey time points were 8:30 am (7:51 am, 9:12 am), 11:45 am (11:30 am, 12:17 pm), and 3:20 pm (3:00 pm, 4:06 pm).

The characteristics of the five hospitals participating in the study, the patients’ final discharge status, the types of physicians surveyed, the services on which they were working, the rounding styles employed, and the median starting daily census are summarized in Table 1. The majority of the physicians surveyed were housestaff working on teaching services, and only a small minority structured rounds such that patients ready for discharge were seen first.

The important findings of this study are that (1) the large majority of issues thought to delay discharging patients identified as definite discharges were related to physicians caring for other patients on their team, (2) although 91% of patients ever identified as being definite discharges were discharged on the day of the survey, only 48% of those identified as possible discharges became definite discharges by the afternoon time point, largely because the anticipated clinical improvement did not occur or care being provided by ancillary services had not been completed, and (3) discharge orders on patients identified as definite discharges were written on average 50 minutes earlier by physicians on teams with a smaller starting patient census, on nonteaching services, or when the rounding style was to see patients ready for discharges first.

Previous research has reported that physician-perceived barriers to discharge were extrinsic to providers and even extrinsic to the hospital setting (eg, awaiting subacute nursing placement and transportation).^28,29 However, many of the barriers that we identified were related directly to the providers’ workload and rounding styles and whether the patients were on teaching versus nonteaching services. We also found that delays in the ability of hospital services to complete care also contributed to delayed discharges.

Our observational data suggest that delays resulting from caring for other patients might be reduced by changing rounding styles such that patients ready for discharge are seen first and are discharged prior to seeing other patients on the team, as previously reported by Beck et al.³⁰ Intuitively, this would seem to be a straightforward way of freeing up beds earlier in the day, but such a change will, of necessity, lead to delaying care for other patients, which, in turn, could increase their length of stays. Durvasula et al. suggested that discharges could be moved to earlier in the day by completing orders and paperwork the day prior to discharge.²⁵ Such an approach might be effective on an Obstetrical or elective Orthopedic service on which patients predictably are hospitalized for a fixed number of days (or even hours) but may be less relevant to patients on internal medicine services where lengths of stay are less predictable. Interventions to improve discharge times have resulted in earlier discharge times in some studies,^2,4 but the overall length of stay either did not decrease²⁵ or increased³¹ in others. Werthheimer et al.¹ did find earlier discharge times, but other interventions also occurred during the study period (eg, extending social work services to include weekends).^1,32

We found that discharge times were approximately 50 minutes earlier on teams with a smaller starting census, on nonteaching compared with teaching services, or when the attending’s rounding style was to see patients ready for discharges first. Although 50 minutes may seem like a small change in discharge time, Khanna et al.³³ found that when discharges occur even 1 hour earlier, hospital overcrowding is reduced. To have a lower team census would require having more teams and more providers to staff these teams, raising cost-effectiveness concerns. Moving to more nonteaching services could represent a conflict with respect to one of the missions of teaching hospitals and raises a cost-benefit issue as several teaching hospitals receive substantial funding in support of their teaching activities and housestaff would have to be replaced with more expensive providers.

Delays attributable to ancillary services indicate imbalances between demand and availability of these services. Inappropriate demand and inefficiencies could be reduced by systems redesign, but in at least some instances, additional resources will be needed to add staff, increase space, or add additional equipment.

Our study has several limitations. First, we surveyed only physicians working in university-affiliated hospitals, and three of these were public safety-net hospitals. Accordingly, our results may not be generalizable to different patient populations. Second, we surveyed only physicians, and Minichiello et al.²⁹ found that barriers to discharge perceived by physicians were different from those of other staff. Third, our data were observational and were collected only on weekdays. Fourth, we did not differentiate interns from residents, and thus, potentially the level of training could have affected these results. Similarly, the decision for a “possible” and a “definite” discharge is likely dependent on the knowledge base of the participant, such that less experienced participants may have had differing perspectives than someone with more experience. Fifth, the sites did vary based on the infrastructure and support but also had several similarities. All sites had social work and case management involved in care, although at some sites, they were assigned according to team and at others according to geographic location. Similarly, rounding times varied. Most of the services surveyed did not utilize advanced practice providers (the exception was the nonteaching services at Denver Health, and their presence was variable). These differences in staffing models could also have affected these results.

Our study also has a number of strengths. First, we assessed the barriers at five different hospitals. Second, we collected real-time data related to specific barriers at multiple time points throughout the day, allowing us to assess the dynamic nature of identifying patients as being ready or nearly ready for discharge. Third, we assessed the perceptions of barriers to discharge from physicians working on teaching as well as nonteaching services and from physicians utilizing a variety of rounding styles. Fourth, we had a very high participation rate (100%), probably due to the fact that our study was strategically aligned with participants’ daily work activities.

In conclusion, we found two distinct categories of issues that physicians perceived as most commonly delaying writing discharge orders on their patients. The first pertained to patients thought to definitely be ready for discharge and was related to the physicians having to care for other patients on their team. The second pertained to patients identified as possibly ready for discharge and was related to the need for care to be completed by a variety of ancillary services. Addressing each of these barriers would require different interventions and a need to weigh the potential improvements that could be achieved against the increased costs and/or delays in care for other patients that may result.

Disclosures

The authors report no conflicts of interest relevant to this work.

Syncope, defined as a transient loss of consciousness and postural tone followed by complete, spontaneous return to neurological baseline, accounts for over 1 million (or approximately 1%) of all emergency department (ED) visits per year in the United States (US).¹² Given the breadth of etiologies for syncope, including certain life-threatening conditions, extensive diagnostic evaluation and hospitalization for this complaint is common.^3-7 The estimated costs of syncope-related hospitalizations are over $2.4 billion annually in the US.⁸

The 2011 American College of Cardiology Foundation appropriate use criteria for echocardiography state that syncope is an appropriate indication for transthoracic echocardiography (TTE) even when there are no other symptoms or signs of cardiovascular disease.⁹ This broad recommendation may be appropriate since a finding of severe valvular disease would generally merit consultation with a cardiothoracic surgeon to assess the potential for surgical intervention.¹⁰ However, routine use of echocardiogram in all syncope patients could result in increased healthcare costs, patient discomfort, and incidental findings of unclear significance, while rarely changing diagnosis or management.^11,12

In an attempt to reduce potentially unnecessary TTE testing, several studies have tried to identify patients at very low risk of structural heart disease.^13-17 These investigations suggest that TTE is not indicated in syncope patients with a normal ECG and a normal cardiac exam. However, this literature is limited by retrospective study design and/or small sample sizes. The 2017 American Heart Association/American College of Cardiology/Heart Rhythm Society syncope guidelines recommend TTE for a patient in whom structural heart disease is suspected, but they are not explicit about how to make this determination. ¹⁸ Thus, it is still unclear which syncope patients require TTE since a standardized approach to assessing risk of clinically significant findings on TTE has not yet been rigorously developed.

The objective of this study was to develop a risk-stratification tool to identify older adults at very low risk of having a major, clinically significant finding on rest TTE after presenting to the ED with syncope or near-syncope. Using clinical, ECG, and cardiac biomarker data, we created the ROMEO (Risk Of Major Echocardiography findings in Older adults with syncope) score to help optimize resource utilization for syncope.

Study Design and Setting

We conducted a large, multicenter, prospective, observational cohort study of older adults who presented to an ED with syncope or near-syncope (ClinicalTrials.gov identifier: NCT01802398). The study was approved by the institutional review boards at all sites and written informed consent was obtained from all participating subjects. The study was conducted at 11 academic EDs across the US (See Appendix Table 1).

Study Population

Patient inclusion criteria for eligibility were age ≥60 years with a complaint of syncope or near-syncope. Syncope was defined as transient loss of consciousness, associated with postural loss of tone, with immediate, spontaneous, and complete recovery. Near syncope was defined as the sensation of imminent loss of consciousness. Patients were excluded if their symptoms were thought to be due to intoxication, seizure, stroke, head trauma, or hypoglycemia. Additional exclusion criteria were the need for medical intervention to restore consciousness (eg, defibrillation), new or worsening confusion, and inability to obtain informed consent from the patient or a legally authorized representative.

This analysis included only patients who received a TTE during the index visit (either in the ED, observation unit, or while admitted to the hospital). This dataset was also used for other analyses addressing questions relevant to the ED management of syncope.

Measurements

All patients underwent a standardized history, physical examination, laboratory, and 12-lead ECG testing. Trained research assistants (RA) directly queried patients about symptoms associated with the syncopal episode. Data on the patient’s past medical history, medications, and physical examination findings were collected prospectively from treating providers.

Research staff obtained blood samples for testing at a core laboratory (University of Rochester, Rochester, NY). Two assays were performed using the Roche Elecsys platform: N-terminal pro B-type natriuretic peptide (NT-proBNP) and the 5th generation high-sensitivity cardiac troponin T (hs-TnT). NT-proBNP was classified as abnormal above a cutoff of 125 pg/mL. Hs-TnT was classified as abnormal above the 99th percentile for a reference population (14 pg/mL). Although hs-TnT was not approved by the U.S. Food and Drug Administration (FDA) at the time of the study, we anticipated that this assay would receive approval and be integrated into future standard of care (FDA approval was granted in January 2017). Rest TTEs were ordered at the discretion of the treating providers.

Outcome Measures

The primary outcome for this secondary analysis was a major, clinically significant finding on TTE.^13,14,16,19 These included severe aortic stenosis (<1 cm²), severe mitral stenosis, severe aortic/mitral regurgitation, reduced ejection fraction (defined either quantitatively as less than 45% or qualitatively as “severe left ventricular dysfunction”), hypertrophic cardiomyopathy with outflow tract obstruction, severe pulmonary hypertension, right ventricular dysfunction/strain, large pericardial effusion, atrial myxoma, or regional wall motion abnormalities.

All echocardiogram reports were independently reviewed by two research physicians. Discrepant reviews were resolved by the research physicians and two of the study investigators (BS, CB). Of note, all the TTEs obtained were formal echocardiographic studies, not bedside ultrasonography performed by the emergency physician.

Candidate Predictors

Potential candidate predictors were identified through a prior expert panel process.^20,21 Candidate predictors included age, sex, abnormal heart sounds, exertional syncope, shortness of breath, chest pain, near-syncope, family history of sudden cardiac death, high (>180 mm Hg) or low (<90 mm Hg) systolic blood pressure, abnormal ECG, elevated hs-TnT, elevated NT-proBNP, and history of the following: hypertension, cardiac dysrhythmia, renal failure, diabetes, congestive heart failure (CHF), and coronary artery disease (CAD).

The first obtained ECG was abstracted by one of five research study physicians blinded to all clinical data. Research study physicians demonstrated high interrater reliability (kappa > 0.80) in distinguishing normal from abnormal ECGs in a training set of 50 ECGs. Abnormal ECG interpretations included nonsinus rhythms (including paced rhythms), multiple premature ventricular complexes, sinus bradycardias (≤40 bpm), ventricular hypertrophies, short PR segment intervals (<100 milliseconds [ms]), axis deviations, first degree blocks (>200 ms), complete bundle branch blocks, Brugada patterns, Wolff-Parkinson-White patterns, abnormal QRS duration (>120 ms) or abnormal QTc prolongations (>450 ms), and Q/ST/T segment abnormalities suggestive of acute or chronic ischemia.

Statistical Analysis

We calculated descriptive statistics for each predictor variable, stratified by the presence or absence of TTE findings. Chi-square and t-tests were used to test associations between categorical or continuous variables and TTE findings using a significance level of 0.05 and 2-sided hypothesis testing. To identify a robust set of predictors of the primary outcome, we used multivariate logistic regression with the LASSO (Least Absolute Shrinkage and Selection Operator) to fit a parsimonious model.²² The LASSO selects variables and shrinks the associated coefficients to avoid overfitting.^23-25 We then used a bootstrap to generate confidence intervals for coefficient estimates. Cases with missing echocardiography reports were excluded from the analysis. Bootstrap results were summarized as the percentage of bootstrap iterations in which each variable’s coefficient was 1) chosen and negative, 2) shrunk to zero, or 3) chosen and positive.

We assessed different weighting schemes to generate a risk score from significant variables identified by regression modeling. These included weighting by regression coefficients rounded to the nearest integer and simple summation of the presence or absence of each variable.

Based on these results, a predictive score was developed to risk stratify patients on their probability of major, clinically significant findings on TTE. The sensitivity and specificity of a score of zero to predict findings on TTE was calculated. For confidence intervals, we used Wilson’s method for binomial confidence intervals.²⁶ The receiver operating characteristic (ROC) curve and its associated area under the curve (AUC) were calculated, and a confidence interval for the AUC was obtained through bootstrap resampling with 2,000 iterations. As part of our sensitivity analyses, we also calculated the ROC curve and AUC after excluding the patients with a known history of CHF and significant finding on TTE. Data analyses were performed in R.²⁷ Two sensitivity analyses were performed: 1) we used multiple imputation to impute 1,000 complete data sets and then used the same LASSO methodology as with the complete data to assess whether incorporating missing data changed the results; and 2) we simulated a conventional troponin assay by raising the positive threshold for hs-TnT to >30 pg/mL (corresponding to the limit of detection for conventional troponin).²⁸

Characteristics of Study Subjects

Patient screening occurred from April 2013 to September 2016. There were 6,930 patients who met eligibility criteria, of whom 3,686 (53%) consented and enrolled in the study (See Figure 1). Of these, 995 (27%) received TTE. The mean age of patients receiving TTE was 74 years; 55% were male. Characteristics of patients obtaining and not obtaining TTE are presented in Appendix Table 2. Patients who received TTE were more likely to be older, have abnormal heart sounds, abnormal EKGs, elevated hs-TnT, elevated NT-proBNP, and have a history of CHF. Of the 995 subjects receiving TTE, 215 (21.6%) had a major, clinically significant finding.

Main Results

Univariate analysis identified 14 variables significantly associated with major findings on TTE. These included male gender, shortness of breath, abnormal heart sounds, history of renal failure, diabetes, CHF, CAD, abnormal ECG, and elevated cardiac biomarkers, among others (See Table 1). The most common major finding on TTE was regional wall motion abnormality, followed by reduced left ventricular ejection fraction (See Table 2). Of the 995 patients who received TTE, 20 (2%) were discharged directly from the ED, 444 (45%) were observed, and 531 (53%) were admitted. On average, patients who received TTE had a longer length of stay than did those that did not (3.4 days vs 1.9 days).

LASSO multivariable logistic regression produced five predictors associated with major findings on TTE: 1) history of CHF, 2) history of CAD, 3) abnormal ECG, 4) hs-TnT above 14 pg/mL, and 5) NT-proBNP above 125 pg/mL (See Table 3).

Older adults with syncope often present to the ED and undergo a variety of diagnostic tests, including TTE, and a significant proportion are admitted to the hospital.² There is currently no standardized, evidence-based approach to guide TTE ordering for these patients. Using a large, prospective dataset of syncope patients, we sought to develop a risk-stratification tool to help clinicians identify which syncope patients would be at very low risk for clinically significant findings on TTE. We found that in the absence of these five high-risk clinical variables, the rate of significant findings on TTE in our sample was less than 1%. All five high-risk variables included in the tool remained predictive in our sensitivity analyses, speaking to the robustness of our model.

Other retrospective, and smaller prospective, studies have identified a combination of low-risk criteria including: a normal ECG alone,¹⁵ a normal physical exam and normal ECG,^14,17 a negative cardiac history and normal ECG.¹⁶ Han et al. performed a chart review of 241 patients presenting to the ED with syncope and identified three risk factors for abnormal TTE findings using multiple logistic regression: age, abnormal ECG, and BNP greater than 100 pg/mL.¹³ While these studies’ results are generally consistent with ours, the retrospective nature and small sample size of these studies limit the generalizability of these results. Thus, using a large, multicenter prospective dataset, we derived a clinical decision instrument (the ROMEO score) to determine which older adults with syncope are at very low risk for major, clinically significant findings on TTE.

Our results add to the recent American College of Cardiology/American Heart Association/Heart Rhythm Society guidelines on the management of syncope which recommend TTE in “selected patients presenting with syncope if structural heart disease is suspected.”¹⁸ Our risk-stratification tool offers a simple, standardized approach to determine specifically when to defer TTE testing.

Our findings can guide clinicians in deciding when to obtain TTE for ED syncope patients in the following way: Older adults presenting with syncope or near-syncope to the ED who have none of the ROMEO criteria are at extremely low risk for clinically significant findings on TTE and thus need not undergo such testing solely because of the syncopal event. Patients who have only one or more high-risk clinical variables are at higher risk (7.3%-56%) of significant TTE findings. In this subset, other factors, (eg, physician gestalt, recent previous echocardiography, patient preference, availability of echocardiography) can help guide TTE ordering. Patients with a greater number of high-risk variables may benefit from a more urgent echocardiographic evaluation.

Although on average, patients undergoing TTE had a longer length of stay than those that did not, this finding does not necessarily imply that ordering a TTE was the cause of the increased length of stay. It is possible that this positive association was due to greater underlying medical complexity or acuity of illness that resulted in a greater likelihood of admission/observation, and in turn, a greater length of stay.

Prior to implementation, our results should be externally validated in other clinical settings. In the interim, this risk-stratification tool may be used by clinicians, in conjunction with clinical judgement, to help guide the appropriate use of TTE in older adults presenting with syncope.

Our study has certain limitations. As we only enrolled patients 60 years and older, our findings may not necessarily be valid in younger populations of syncope patients. However, structural heart disease is less common in younger patients and is generally more of a concern for clinicians when evaluating syncope patients in the older age range.²⁹ In our study, 47% of eligible patients declined to participate and thus sampling bias may have occurred. TTEs were ordered at the discretion of treating providers, which was likely subject to physician, institutional, and regional variation; the prevalence of major TTE findings may be lower in the overall cohort than in patients who received TTE. Prior TTE reports were not available; therefore, we were not able to determine if these major findings were previously known. Importantly, we did not perform an internal or external validation of the ROMEO score due to time and resource constraints. Thus, this study represents a derivation of the score solely and would require external validation prior to clinical implementation. Also, to calculate the ROMEO score, both an hs-TnT and NT-proBNP level must be obtained. Thus, the cost savings of any potential reduction in TTE ordering may be partially offset by the costs of increased laboratory testing. Lastly, hs-TnT assays are not currently widely available in hospitals in the United States; earlier generation cardiac troponin assays may not be a perfect substitute for hs-TnT assays. Our sensitivity analysis using an elevated threshold for hs-TnT attempted to mitigate this limitation and resulted in similar findings.

In summary, this risk-stratification tool, using five simple criteria, could help clinicians determine which older adult syncope patients can safely forgo TTE. If validated, this tool could help optimize resource utilization, and increase the value of healthcare for patients presenting with syncope.

The authors would like to thank the research assistants at all 11 sites who enrolled patients and collected data for this study.

Disclosures

Dr. Adler has received research funding from Roche. Dr. Bastani has received research funding from Radiometer and Portola and has been a consultant for Portola. Dr. Baugh has received advisory board and speaker’s fees from Roche, research funding from Janssen and Boehringer Ingelheim, and consulting and advisory board fees from Janssen. Dr. Casterino has received funding from Astra Zeneca. Dr. Clark has received research funding from Radiometer, Ortho Clinical Trials, Janssen, Pfizer, NIH, Portola, Biocryst, Glaxo Smith Klein, Hospital Quality Foundation, and Abbott. She is a consultant for Portola, Janssen, and Hospital Quality Foundation. Dr. Diercks is a consultant for Siemens, Janssen, and Roche has received institutional research support from Novartis, Ortho Scientific, and Roche. Dr. Hollander has received research funding from Alere, Siemens, Roche, Portola, and Trinity. Dr. Hollander has also received royalties from UpToDate. Dr. Nishijima has received an honorarium from Pfizer. Dr. Storrow is a consultant for Siemens and Quidel, has received speaking fees from MCM Education, and is on the Data and Safety Monitoring Board for Trevena. Dr. Sun is a consultant for Medtronic. The other authors report no relevant conflicts of interest.

Funding

This project was supported by the National Heart, Lung, And Blood Institute of the National Institutes of Health under Award Number R01 HL111033. Dr. Probst is supported by the National Heart, Lung, And Blood Institute of the National Institutes of Health under Award Number K23HL132052-02. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Roche Diagnostics supplied the high-sensitivity troponin-T assays. The sponsoring organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, or review of the manuscript.

Syncope, defined as a transient loss of consciousness and postural tone followed by complete, spontaneous return to neurological baseline, accounts for over 1 million (or approximately 1%) of all emergency department (ED) visits per year in the United States (US).¹² Given the breadth of etiologies for syncope, including certain life-threatening conditions, extensive diagnostic evaluation and hospitalization for this complaint is common.^3-7 The estimated costs of syncope-related hospitalizations are over $2.4 billion annually in the US.⁸

The 2011 American College of Cardiology Foundation appropriate use criteria for echocardiography state that syncope is an appropriate indication for transthoracic echocardiography (TTE) even when there are no other symptoms or signs of cardiovascular disease.⁹ This broad recommendation may be appropriate since a finding of severe valvular disease would generally merit consultation with a cardiothoracic surgeon to assess the potential for surgical intervention.¹⁰ However, routine use of echocardiogram in all syncope patients could result in increased healthcare costs, patient discomfort, and incidental findings of unclear significance, while rarely changing diagnosis or management.^11,12

In an attempt to reduce potentially unnecessary TTE testing, several studies have tried to identify patients at very low risk of structural heart disease.^13-17 These investigations suggest that TTE is not indicated in syncope patients with a normal ECG and a normal cardiac exam. However, this literature is limited by retrospective study design and/or small sample sizes. The 2017 American Heart Association/American College of Cardiology/Heart Rhythm Society syncope guidelines recommend TTE for a patient in whom structural heart disease is suspected, but they are not explicit about how to make this determination. ¹⁸ Thus, it is still unclear which syncope patients require TTE since a standardized approach to assessing risk of clinically significant findings on TTE has not yet been rigorously developed.

The objective of this study was to develop a risk-stratification tool to identify older adults at very low risk of having a major, clinically significant finding on rest TTE after presenting to the ED with syncope or near-syncope. Using clinical, ECG, and cardiac biomarker data, we created the ROMEO (Risk Of Major Echocardiography findings in Older adults with syncope) score to help optimize resource utilization for syncope.

Study Design and Setting

We conducted a large, multicenter, prospective, observational cohort study of older adults who presented to an ED with syncope or near-syncope (ClinicalTrials.gov identifier: NCT01802398). The study was approved by the institutional review boards at all sites and written informed consent was obtained from all participating subjects. The study was conducted at 11 academic EDs across the US (See Appendix Table 1).

Study Population

Patient inclusion criteria for eligibility were age ≥60 years with a complaint of syncope or near-syncope. Syncope was defined as transient loss of consciousness, associated with postural loss of tone, with immediate, spontaneous, and complete recovery. Near syncope was defined as the sensation of imminent loss of consciousness. Patients were excluded if their symptoms were thought to be due to intoxication, seizure, stroke, head trauma, or hypoglycemia. Additional exclusion criteria were the need for medical intervention to restore consciousness (eg, defibrillation), new or worsening confusion, and inability to obtain informed consent from the patient or a legally authorized representative.

This analysis included only patients who received a TTE during the index visit (either in the ED, observation unit, or while admitted to the hospital). This dataset was also used for other analyses addressing questions relevant to the ED management of syncope.

Measurements

All patients underwent a standardized history, physical examination, laboratory, and 12-lead ECG testing. Trained research assistants (RA) directly queried patients about symptoms associated with the syncopal episode. Data on the patient’s past medical history, medications, and physical examination findings were collected prospectively from treating providers.

Research staff obtained blood samples for testing at a core laboratory (University of Rochester, Rochester, NY). Two assays were performed using the Roche Elecsys platform: N-terminal pro B-type natriuretic peptide (NT-proBNP) and the 5th generation high-sensitivity cardiac troponin T (hs-TnT). NT-proBNP was classified as abnormal above a cutoff of 125 pg/mL. Hs-TnT was classified as abnormal above the 99th percentile for a reference population (14 pg/mL). Although hs-TnT was not approved by the U.S. Food and Drug Administration (FDA) at the time of the study, we anticipated that this assay would receive approval and be integrated into future standard of care (FDA approval was granted in January 2017). Rest TTEs were ordered at the discretion of the treating providers.

Outcome Measures

The primary outcome for this secondary analysis was a major, clinically significant finding on TTE.^13,14,16,19 These included severe aortic stenosis (<1 cm²), severe mitral stenosis, severe aortic/mitral regurgitation, reduced ejection fraction (defined either quantitatively as less than 45% or qualitatively as “severe left ventricular dysfunction”), hypertrophic cardiomyopathy with outflow tract obstruction, severe pulmonary hypertension, right ventricular dysfunction/strain, large pericardial effusion, atrial myxoma, or regional wall motion abnormalities.

All echocardiogram reports were independently reviewed by two research physicians. Discrepant reviews were resolved by the research physicians and two of the study investigators (BS, CB). Of note, all the TTEs obtained were formal echocardiographic studies, not bedside ultrasonography performed by the emergency physician.

Candidate Predictors

Potential candidate predictors were identified through a prior expert panel process.^20,21 Candidate predictors included age, sex, abnormal heart sounds, exertional syncope, shortness of breath, chest pain, near-syncope, family history of sudden cardiac death, high (>180 mm Hg) or low (<90 mm Hg) systolic blood pressure, abnormal ECG, elevated hs-TnT, elevated NT-proBNP, and history of the following: hypertension, cardiac dysrhythmia, renal failure, diabetes, congestive heart failure (CHF), and coronary artery disease (CAD).

The first obtained ECG was abstracted by one of five research study physicians blinded to all clinical data. Research study physicians demonstrated high interrater reliability (kappa > 0.80) in distinguishing normal from abnormal ECGs in a training set of 50 ECGs. Abnormal ECG interpretations included nonsinus rhythms (including paced rhythms), multiple premature ventricular complexes, sinus bradycardias (≤40 bpm), ventricular hypertrophies, short PR segment intervals (<100 milliseconds [ms]), axis deviations, first degree blocks (>200 ms), complete bundle branch blocks, Brugada patterns, Wolff-Parkinson-White patterns, abnormal QRS duration (>120 ms) or abnormal QTc prolongations (>450 ms), and Q/ST/T segment abnormalities suggestive of acute or chronic ischemia.

Statistical Analysis

We calculated descriptive statistics for each predictor variable, stratified by the presence or absence of TTE findings. Chi-square and t-tests were used to test associations between categorical or continuous variables and TTE findings using a significance level of 0.05 and 2-sided hypothesis testing. To identify a robust set of predictors of the primary outcome, we used multivariate logistic regression with the LASSO (Least Absolute Shrinkage and Selection Operator) to fit a parsimonious model.²² The LASSO selects variables and shrinks the associated coefficients to avoid overfitting.^23-25 We then used a bootstrap to generate confidence intervals for coefficient estimates. Cases with missing echocardiography reports were excluded from the analysis. Bootstrap results were summarized as the percentage of bootstrap iterations in which each variable’s coefficient was 1) chosen and negative, 2) shrunk to zero, or 3) chosen and positive.

We assessed different weighting schemes to generate a risk score from significant variables identified by regression modeling. These included weighting by regression coefficients rounded to the nearest integer and simple summation of the presence or absence of each variable.

Based on these results, a predictive score was developed to risk stratify patients on their probability of major, clinically significant findings on TTE. The sensitivity and specificity of a score of zero to predict findings on TTE was calculated. For confidence intervals, we used Wilson’s method for binomial confidence intervals.²⁶ The receiver operating characteristic (ROC) curve and its associated area under the curve (AUC) were calculated, and a confidence interval for the AUC was obtained through bootstrap resampling with 2,000 iterations. As part of our sensitivity analyses, we also calculated the ROC curve and AUC after excluding the patients with a known history of CHF and significant finding on TTE. Data analyses were performed in R.²⁷ Two sensitivity analyses were performed: 1) we used multiple imputation to impute 1,000 complete data sets and then used the same LASSO methodology as with the complete data to assess whether incorporating missing data changed the results; and 2) we simulated a conventional troponin assay by raising the positive threshold for hs-TnT to >30 pg/mL (corresponding to the limit of detection for conventional troponin).²⁸

Characteristics of Study Subjects

Patient screening occurred from April 2013 to September 2016. There were 6,930 patients who met eligibility criteria, of whom 3,686 (53%) consented and enrolled in the study (See Figure 1). Of these, 995 (27%) received TTE. The mean age of patients receiving TTE was 74 years; 55% were male. Characteristics of patients obtaining and not obtaining TTE are presented in Appendix Table 2. Patients who received TTE were more likely to be older, have abnormal heart sounds, abnormal EKGs, elevated hs-TnT, elevated NT-proBNP, and have a history of CHF. Of the 995 subjects receiving TTE, 215 (21.6%) had a major, clinically significant finding.

Main Results

Univariate analysis identified 14 variables significantly associated with major findings on TTE. These included male gender, shortness of breath, abnormal heart sounds, history of renal failure, diabetes, CHF, CAD, abnormal ECG, and elevated cardiac biomarkers, among others (See Table 1). The most common major finding on TTE was regional wall motion abnormality, followed by reduced left ventricular ejection fraction (See Table 2). Of the 995 patients who received TTE, 20 (2%) were discharged directly from the ED, 444 (45%) were observed, and 531 (53%) were admitted. On average, patients who received TTE had a longer length of stay than did those that did not (3.4 days vs 1.9 days).

LASSO multivariable logistic regression produced five predictors associated with major findings on TTE: 1) history of CHF, 2) history of CAD, 3) abnormal ECG, 4) hs-TnT above 14 pg/mL, and 5) NT-proBNP above 125 pg/mL (See Table 3).

Older adults with syncope often present to the ED and undergo a variety of diagnostic tests, including TTE, and a significant proportion are admitted to the hospital.² There is currently no standardized, evidence-based approach to guide TTE ordering for these patients. Using a large, prospective dataset of syncope patients, we sought to develop a risk-stratification tool to help clinicians identify which syncope patients would be at very low risk for clinically significant findings on TTE. We found that in the absence of these five high-risk clinical variables, the rate of significant findings on TTE in our sample was less than 1%. All five high-risk variables included in the tool remained predictive in our sensitivity analyses, speaking to the robustness of our model.

Other retrospective, and smaller prospective, studies have identified a combination of low-risk criteria including: a normal ECG alone,¹⁵ a normal physical exam and normal ECG,^14,17 a negative cardiac history and normal ECG.¹⁶ Han et al. performed a chart review of 241 patients presenting to the ED with syncope and identified three risk factors for abnormal TTE findings using multiple logistic regression: age, abnormal ECG, and BNP greater than 100 pg/mL.¹³ While these studies’ results are generally consistent with ours, the retrospective nature and small sample size of these studies limit the generalizability of these results. Thus, using a large, multicenter prospective dataset, we derived a clinical decision instrument (the ROMEO score) to determine which older adults with syncope are at very low risk for major, clinically significant findings on TTE.

Our results add to the recent American College of Cardiology/American Heart Association/Heart Rhythm Society guidelines on the management of syncope which recommend TTE in “selected patients presenting with syncope if structural heart disease is suspected.”¹⁸ Our risk-stratification tool offers a simple, standardized approach to determine specifically when to defer TTE testing.

Our findings can guide clinicians in deciding when to obtain TTE for ED syncope patients in the following way: Older adults presenting with syncope or near-syncope to the ED who have none of the ROMEO criteria are at extremely low risk for clinically significant findings on TTE and thus need not undergo such testing solely because of the syncopal event. Patients who have only one or more high-risk clinical variables are at higher risk (7.3%-56%) of significant TTE findings. In this subset, other factors, (eg, physician gestalt, recent previous echocardiography, patient preference, availability of echocardiography) can help guide TTE ordering. Patients with a greater number of high-risk variables may benefit from a more urgent echocardiographic evaluation.

Although on average, patients undergoing TTE had a longer length of stay than those that did not, this finding does not necessarily imply that ordering a TTE was the cause of the increased length of stay. It is possible that this positive association was due to greater underlying medical complexity or acuity of illness that resulted in a greater likelihood of admission/observation, and in turn, a greater length of stay.

Prior to implementation, our results should be externally validated in other clinical settings. In the interim, this risk-stratification tool may be used by clinicians, in conjunction with clinical judgement, to help guide the appropriate use of TTE in older adults presenting with syncope.

Our study has certain limitations. As we only enrolled patients 60 years and older, our findings may not necessarily be valid in younger populations of syncope patients. However, structural heart disease is less common in younger patients and is generally more of a concern for clinicians when evaluating syncope patients in the older age range.²⁹ In our study, 47% of eligible patients declined to participate and thus sampling bias may have occurred. TTEs were ordered at the discretion of treating providers, which was likely subject to physician, institutional, and regional variation; the prevalence of major TTE findings may be lower in the overall cohort than in patients who received TTE. Prior TTE reports were not available; therefore, we were not able to determine if these major findings were previously known. Importantly, we did not perform an internal or external validation of the ROMEO score due to time and resource constraints. Thus, this study represents a derivation of the score solely and would require external validation prior to clinical implementation. Also, to calculate the ROMEO score, both an hs-TnT and NT-proBNP level must be obtained. Thus, the cost savings of any potential reduction in TTE ordering may be partially offset by the costs of increased laboratory testing. Lastly, hs-TnT assays are not currently widely available in hospitals in the United States; earlier generation cardiac troponin assays may not be a perfect substitute for hs-TnT assays. Our sensitivity analysis using an elevated threshold for hs-TnT attempted to mitigate this limitation and resulted in similar findings.

In summary, this risk-stratification tool, using five simple criteria, could help clinicians determine which older adult syncope patients can safely forgo TTE. If validated, this tool could help optimize resource utilization, and increase the value of healthcare for patients presenting with syncope.

The authors would like to thank the research assistants at all 11 sites who enrolled patients and collected data for this study.

Disclosures

Dr. Adler has received research funding from Roche. Dr. Bastani has received research funding from Radiometer and Portola and has been a consultant for Portola. Dr. Baugh has received advisory board and speaker’s fees from Roche, research funding from Janssen and Boehringer Ingelheim, and consulting and advisory board fees from Janssen. Dr. Casterino has received funding from Astra Zeneca. Dr. Clark has received research funding from Radiometer, Ortho Clinical Trials, Janssen, Pfizer, NIH, Portola, Biocryst, Glaxo Smith Klein, Hospital Quality Foundation, and Abbott. She is a consultant for Portola, Janssen, and Hospital Quality Foundation. Dr. Diercks is a consultant for Siemens, Janssen, and Roche has received institutional research support from Novartis, Ortho Scientific, and Roche. Dr. Hollander has received research funding from Alere, Siemens, Roche, Portola, and Trinity. Dr. Hollander has also received royalties from UpToDate. Dr. Nishijima has received an honorarium from Pfizer. Dr. Storrow is a consultant for Siemens and Quidel, has received speaking fees from MCM Education, and is on the Data and Safety Monitoring Board for Trevena. Dr. Sun is a consultant for Medtronic. The other authors report no relevant conflicts of interest.

Funding

This project was supported by the National Heart, Lung, And Blood Institute of the National Institutes of Health under Award Number R01 HL111033. Dr. Probst is supported by the National Heart, Lung, And Blood Institute of the National Institutes of Health under Award Number K23HL132052-02. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Roche Diagnostics supplied the high-sensitivity troponin-T assays. The sponsoring organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, or review of the manuscript.

Syncope, defined as a transient loss of consciousness and postural tone followed by complete, spontaneous return to neurological baseline, accounts for over 1 million (or approximately 1%) of all emergency department (ED) visits per year in the United States (US).¹² Given the breadth of etiologies for syncope, including certain life-threatening conditions, extensive diagnostic evaluation and hospitalization for this complaint is common.^3-7 The estimated costs of syncope-related hospitalizations are over $2.4 billion annually in the US.⁸

The 2011 American College of Cardiology Foundation appropriate use criteria for echocardiography state that syncope is an appropriate indication for transthoracic echocardiography (TTE) even when there are no other symptoms or signs of cardiovascular disease.⁹ This broad recommendation may be appropriate since a finding of severe valvular disease would generally merit consultation with a cardiothoracic surgeon to assess the potential for surgical intervention.¹⁰ However, routine use of echocardiogram in all syncope patients could result in increased healthcare costs, patient discomfort, and incidental findings of unclear significance, while rarely changing diagnosis or management.^11,12

In an attempt to reduce potentially unnecessary TTE testing, several studies have tried to identify patients at very low risk of structural heart disease.^13-17 These investigations suggest that TTE is not indicated in syncope patients with a normal ECG and a normal cardiac exam. However, this literature is limited by retrospective study design and/or small sample sizes. The 2017 American Heart Association/American College of Cardiology/Heart Rhythm Society syncope guidelines recommend TTE for a patient in whom structural heart disease is suspected, but they are not explicit about how to make this determination. ¹⁸ Thus, it is still unclear which syncope patients require TTE since a standardized approach to assessing risk of clinically significant findings on TTE has not yet been rigorously developed.

The objective of this study was to develop a risk-stratification tool to identify older adults at very low risk of having a major, clinically significant finding on rest TTE after presenting to the ED with syncope or near-syncope. Using clinical, ECG, and cardiac biomarker data, we created the ROMEO (Risk Of Major Echocardiography findings in Older adults with syncope) score to help optimize resource utilization for syncope.

Study Design and Setting

We conducted a large, multicenter, prospective, observational cohort study of older adults who presented to an ED with syncope or near-syncope (ClinicalTrials.gov identifier: NCT01802398). The study was approved by the institutional review boards at all sites and written informed consent was obtained from all participating subjects. The study was conducted at 11 academic EDs across the US (See Appendix Table 1).

Study Population

Patient inclusion criteria for eligibility were age ≥60 years with a complaint of syncope or near-syncope. Syncope was defined as transient loss of consciousness, associated with postural loss of tone, with immediate, spontaneous, and complete recovery. Near syncope was defined as the sensation of imminent loss of consciousness. Patients were excluded if their symptoms were thought to be due to intoxication, seizure, stroke, head trauma, or hypoglycemia. Additional exclusion criteria were the need for medical intervention to restore consciousness (eg, defibrillation), new or worsening confusion, and inability to obtain informed consent from the patient or a legally authorized representative.

This analysis included only patients who received a TTE during the index visit (either in the ED, observation unit, or while admitted to the hospital). This dataset was also used for other analyses addressing questions relevant to the ED management of syncope.

Measurements

All patients underwent a standardized history, physical examination, laboratory, and 12-lead ECG testing. Trained research assistants (RA) directly queried patients about symptoms associated with the syncopal episode. Data on the patient’s past medical history, medications, and physical examination findings were collected prospectively from treating providers.

Research staff obtained blood samples for testing at a core laboratory (University of Rochester, Rochester, NY). Two assays were performed using the Roche Elecsys platform: N-terminal pro B-type natriuretic peptide (NT-proBNP) and the 5th generation high-sensitivity cardiac troponin T (hs-TnT). NT-proBNP was classified as abnormal above a cutoff of 125 pg/mL. Hs-TnT was classified as abnormal above the 99th percentile for a reference population (14 pg/mL). Although hs-TnT was not approved by the U.S. Food and Drug Administration (FDA) at the time of the study, we anticipated that this assay would receive approval and be integrated into future standard of care (FDA approval was granted in January 2017). Rest TTEs were ordered at the discretion of the treating providers.

Outcome Measures

The primary outcome for this secondary analysis was a major, clinically significant finding on TTE.^13,14,16,19 These included severe aortic stenosis (<1 cm²), severe mitral stenosis, severe aortic/mitral regurgitation, reduced ejection fraction (defined either quantitatively as less than 45% or qualitatively as “severe left ventricular dysfunction”), hypertrophic cardiomyopathy with outflow tract obstruction, severe pulmonary hypertension, right ventricular dysfunction/strain, large pericardial effusion, atrial myxoma, or regional wall motion abnormalities.

All echocardiogram reports were independently reviewed by two research physicians. Discrepant reviews were resolved by the research physicians and two of the study investigators (BS, CB). Of note, all the TTEs obtained were formal echocardiographic studies, not bedside ultrasonography performed by the emergency physician.

Candidate Predictors

Potential candidate predictors were identified through a prior expert panel process.^20,21 Candidate predictors included age, sex, abnormal heart sounds, exertional syncope, shortness of breath, chest pain, near-syncope, family history of sudden cardiac death, high (>180 mm Hg) or low (<90 mm Hg) systolic blood pressure, abnormal ECG, elevated hs-TnT, elevated NT-proBNP, and history of the following: hypertension, cardiac dysrhythmia, renal failure, diabetes, congestive heart failure (CHF), and coronary artery disease (CAD).

The first obtained ECG was abstracted by one of five research study physicians blinded to all clinical data. Research study physicians demonstrated high interrater reliability (kappa > 0.80) in distinguishing normal from abnormal ECGs in a training set of 50 ECGs. Abnormal ECG interpretations included nonsinus rhythms (including paced rhythms), multiple premature ventricular complexes, sinus bradycardias (≤40 bpm), ventricular hypertrophies, short PR segment intervals (<100 milliseconds [ms]), axis deviations, first degree blocks (>200 ms), complete bundle branch blocks, Brugada patterns, Wolff-Parkinson-White patterns, abnormal QRS duration (>120 ms) or abnormal QTc prolongations (>450 ms), and Q/ST/T segment abnormalities suggestive of acute or chronic ischemia.

Statistical Analysis

We calculated descriptive statistics for each predictor variable, stratified by the presence or absence of TTE findings. Chi-square and t-tests were used to test associations between categorical or continuous variables and TTE findings using a significance level of 0.05 and 2-sided hypothesis testing. To identify a robust set of predictors of the primary outcome, we used multivariate logistic regression with the LASSO (Least Absolute Shrinkage and Selection Operator) to fit a parsimonious model.²² The LASSO selects variables and shrinks the associated coefficients to avoid overfitting.^23-25 We then used a bootstrap to generate confidence intervals for coefficient estimates. Cases with missing echocardiography reports were excluded from the analysis. Bootstrap results were summarized as the percentage of bootstrap iterations in which each variable’s coefficient was 1) chosen and negative, 2) shrunk to zero, or 3) chosen and positive.

We assessed different weighting schemes to generate a risk score from significant variables identified by regression modeling. These included weighting by regression coefficients rounded to the nearest integer and simple summation of the presence or absence of each variable.

Based on these results, a predictive score was developed to risk stratify patients on their probability of major, clinically significant findings on TTE. The sensitivity and specificity of a score of zero to predict findings on TTE was calculated. For confidence intervals, we used Wilson’s method for binomial confidence intervals.²⁶ The receiver operating characteristic (ROC) curve and its associated area under the curve (AUC) were calculated, and a confidence interval for the AUC was obtained through bootstrap resampling with 2,000 iterations. As part of our sensitivity analyses, we also calculated the ROC curve and AUC after excluding the patients with a known history of CHF and significant finding on TTE. Data analyses were performed in R.²⁷ Two sensitivity analyses were performed: 1) we used multiple imputation to impute 1,000 complete data sets and then used the same LASSO methodology as with the complete data to assess whether incorporating missing data changed the results; and 2) we simulated a conventional troponin assay by raising the positive threshold for hs-TnT to >30 pg/mL (corresponding to the limit of detection for conventional troponin).²⁸

Characteristics of Study Subjects

Patient screening occurred from April 2013 to September 2016. There were 6,930 patients who met eligibility criteria, of whom 3,686 (53%) consented and enrolled in the study (See Figure 1). Of these, 995 (27%) received TTE. The mean age of patients receiving TTE was 74 years; 55% were male. Characteristics of patients obtaining and not obtaining TTE are presented in Appendix Table 2. Patients who received TTE were more likely to be older, have abnormal heart sounds, abnormal EKGs, elevated hs-TnT, elevated NT-proBNP, and have a history of CHF. Of the 995 subjects receiving TTE, 215 (21.6%) had a major, clinically significant finding.

Main Results

Univariate analysis identified 14 variables significantly associated with major findings on TTE. These included male gender, shortness of breath, abnormal heart sounds, history of renal failure, diabetes, CHF, CAD, abnormal ECG, and elevated cardiac biomarkers, among others (See Table 1). The most common major finding on TTE was regional wall motion abnormality, followed by reduced left ventricular ejection fraction (See Table 2). Of the 995 patients who received TTE, 20 (2%) were discharged directly from the ED, 444 (45%) were observed, and 531 (53%) were admitted. On average, patients who received TTE had a longer length of stay than did those that did not (3.4 days vs 1.9 days).

LASSO multivariable logistic regression produced five predictors associated with major findings on TTE: 1) history of CHF, 2) history of CAD, 3) abnormal ECG, 4) hs-TnT above 14 pg/mL, and 5) NT-proBNP above 125 pg/mL (See Table 3).

Older adults with syncope often present to the ED and undergo a variety of diagnostic tests, including TTE, and a significant proportion are admitted to the hospital.² There is currently no standardized, evidence-based approach to guide TTE ordering for these patients. Using a large, prospective dataset of syncope patients, we sought to develop a risk-stratification tool to help clinicians identify which syncope patients would be at very low risk for clinically significant findings on TTE. We found that in the absence of these five high-risk clinical variables, the rate of significant findings on TTE in our sample was less than 1%. All five high-risk variables included in the tool remained predictive in our sensitivity analyses, speaking to the robustness of our model.

Other retrospective, and smaller prospective, studies have identified a combination of low-risk criteria including: a normal ECG alone,¹⁵ a normal physical exam and normal ECG,^14,17 a negative cardiac history and normal ECG.¹⁶ Han et al. performed a chart review of 241 patients presenting to the ED with syncope and identified three risk factors for abnormal TTE findings using multiple logistic regression: age, abnormal ECG, and BNP greater than 100 pg/mL.¹³ While these studies’ results are generally consistent with ours, the retrospective nature and small sample size of these studies limit the generalizability of these results. Thus, using a large, multicenter prospective dataset, we derived a clinical decision instrument (the ROMEO score) to determine which older adults with syncope are at very low risk for major, clinically significant findings on TTE.

Our results add to the recent American College of Cardiology/American Heart Association/Heart Rhythm Society guidelines on the management of syncope which recommend TTE in “selected patients presenting with syncope if structural heart disease is suspected.”¹⁸ Our risk-stratification tool offers a simple, standardized approach to determine specifically when to defer TTE testing.

Our findings can guide clinicians in deciding when to obtain TTE for ED syncope patients in the following way: Older adults presenting with syncope or near-syncope to the ED who have none of the ROMEO criteria are at extremely low risk for clinically significant findings on TTE and thus need not undergo such testing solely because of the syncopal event. Patients who have only one or more high-risk clinical variables are at higher risk (7.3%-56%) of significant TTE findings. In this subset, other factors, (eg, physician gestalt, recent previous echocardiography, patient preference, availability of echocardiography) can help guide TTE ordering. Patients with a greater number of high-risk variables may benefit from a more urgent echocardiographic evaluation.

Although on average, patients undergoing TTE had a longer length of stay than those that did not, this finding does not necessarily imply that ordering a TTE was the cause of the increased length of stay. It is possible that this positive association was due to greater underlying medical complexity or acuity of illness that resulted in a greater likelihood of admission/observation, and in turn, a greater length of stay.

Prior to implementation, our results should be externally validated in other clinical settings. In the interim, this risk-stratification tool may be used by clinicians, in conjunction with clinical judgement, to help guide the appropriate use of TTE in older adults presenting with syncope.

Our study has certain limitations. As we only enrolled patients 60 years and older, our findings may not necessarily be valid in younger populations of syncope patients. However, structural heart disease is less common in younger patients and is generally more of a concern for clinicians when evaluating syncope patients in the older age range.²⁹ In our study, 47% of eligible patients declined to participate and thus sampling bias may have occurred. TTEs were ordered at the discretion of treating providers, which was likely subject to physician, institutional, and regional variation; the prevalence of major TTE findings may be lower in the overall cohort than in patients who received TTE. Prior TTE reports were not available; therefore, we were not able to determine if these major findings were previously known. Importantly, we did not perform an internal or external validation of the ROMEO score due to time and resource constraints. Thus, this study represents a derivation of the score solely and would require external validation prior to clinical implementation. Also, to calculate the ROMEO score, both an hs-TnT and NT-proBNP level must be obtained. Thus, the cost savings of any potential reduction in TTE ordering may be partially offset by the costs of increased laboratory testing. Lastly, hs-TnT assays are not currently widely available in hospitals in the United States; earlier generation cardiac troponin assays may not be a perfect substitute for hs-TnT assays. Our sensitivity analysis using an elevated threshold for hs-TnT attempted to mitigate this limitation and resulted in similar findings.

In summary, this risk-stratification tool, using five simple criteria, could help clinicians determine which older adult syncope patients can safely forgo TTE. If validated, this tool could help optimize resource utilization, and increase the value of healthcare for patients presenting with syncope.

The authors would like to thank the research assistants at all 11 sites who enrolled patients and collected data for this study.

Disclosures

Dr. Adler has received research funding from Roche. Dr. Bastani has received research funding from Radiometer and Portola and has been a consultant for Portola. Dr. Baugh has received advisory board and speaker’s fees from Roche, research funding from Janssen and Boehringer Ingelheim, and consulting and advisory board fees from Janssen. Dr. Casterino has received funding from Astra Zeneca. Dr. Clark has received research funding from Radiometer, Ortho Clinical Trials, Janssen, Pfizer, NIH, Portola, Biocryst, Glaxo Smith Klein, Hospital Quality Foundation, and Abbott. She is a consultant for Portola, Janssen, and Hospital Quality Foundation. Dr. Diercks is a consultant for Siemens, Janssen, and Roche has received institutional research support from Novartis, Ortho Scientific, and Roche. Dr. Hollander has received research funding from Alere, Siemens, Roche, Portola, and Trinity. Dr. Hollander has also received royalties from UpToDate. Dr. Nishijima has received an honorarium from Pfizer. Dr. Storrow is a consultant for Siemens and Quidel, has received speaking fees from MCM Education, and is on the Data and Safety Monitoring Board for Trevena. Dr. Sun is a consultant for Medtronic. The other authors report no relevant conflicts of interest.

Funding

This project was supported by the National Heart, Lung, And Blood Institute of the National Institutes of Health under Award Number R01 HL111033. Dr. Probst is supported by the National Heart, Lung, And Blood Institute of the National Institutes of Health under Award Number K23HL132052-02. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Roche Diagnostics supplied the high-sensitivity troponin-T assays. The sponsoring organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, or review of the manuscript.

Resident physician workload has traditionally been measured by patient census.^1,2 However, census and other volume-based metrics such as daily admissions may not accurately reflect workload due to variation in patient complexity. Relative value units (RVUs) are another commonly used marker of workload, but the validity of this metric relies on accurate coding, usually done by the attending physician, and is less directly related to resident physician workload. Because much of hospital-based medicine is mediated through the electronic health record (EHR), which can capture differences in patient complexity,³ electronic records could be harnessed to more comprehensively describe residents’ work. Current government estimates indicate that several hundred companies offer certified EHRs, thanks in large part to the Health Information Technology for Economic and Clinical Health (HITECH) Act of 2009, which aimed to promote adoption and meaningful use of health information technology.^{4, 5} These systems can collect important data about the usage and operating patterns of physicians, which may provide an insight into workload.^6-8

Accurately measuring workload is important because of the direct link that has been drawn between physician workload and quality metrics. In a study of attending hospitalists, higher workload, as measured by patient census and RVUs, was associated with longer lengths of stay and higher costs of hospitalization.⁹ Another study among medical residents found that as daily admissions increased, length of stay, cost, and inpatient mortality appeared to rise.¹⁰ Although these studies used only volume-based workload metrics, the implication that high workload may negatively impact patient care hints at a possible trade-off between the two that should inform discussions of physician productivity.

In the current study, we examine whether data obtained from the EHR, particularly electronic order volume, could provide valuable information, in addition to patient volume, about resident physician workload. We first tested the feasibility and validity of using electronic order volume as an important component of clinical workload by examining the relationship between electronic order volume and well-established factors that are likely to increase the workload of residents, including patient level of care and severity of illness. Then, using order volume as a marker for workload, we sought to describe whether higher order volumes were associated with two discharge-related quality metrics, completion of a high-quality after-visit summary and timely discharge summary, postulating that quality metrics may suffer when residents are busier.

Study Design and Setting

We performed a single-center retrospective cohort study of patients admitted to the internal medicine service at the University of California, San Francisco (UCSF) Medical Center between May 1, 2015 and July 31, 2016. UCSF is a 600-bed academic medical center, and the inpatient internal medicine teaching service manages an average daily census of 80-90 patients. Medicine teams care for patients on the general acute-care wards, the step-down units (for patients with higher acuity of care), and also patients in the intensive care unit (ICU). ICU patients are comanaged by general medicine teams and intensive care teams; internal medicine teams enter all electronic orders for ICU patients, except for orders for respiratory care or sedating medications. The inpatient internal medicine teaching service comprises eight teams each supervised by an attending physician, a senior resident (in the second or third year of residency training), two interns, and a third- and/or fourth-year medical student. Residents place all clinical orders and complete all clinical documentation through the EHR (Epic Systems, Verona, Wisconsin).¹¹ Typically, the bulk of the orders and documentation, including discharge documentation, is performed by interns; however, the degree of senior resident involvement in these tasks is variable and team-dependent. In addition to the eight resident teams, there are also four attending hospitalist-only internal medicine teams, who manage a service of ~30-40 patients.

Study Population

Our study population comprised all hospitalized adults admitted to the eight resident-run teams on the internal medicine teaching service. Patients cared for by hospitalist-only teams were not included in this analysis. Because the focus of our study was on hospitalizations, individual patients may have been included multiple times over the course of the study. Hospitalizations were excluded if they did not have complete Medicare Severity-Diagnosis Related Group (MS-DRG) data,¹² since this was used as our severity of illness marker. This occurred either because patients were not discharged by the end of the study period or because they had a length of stay of less than one day, because this metric was not assigned to these short-stay (observation) patients.

Data Collection

All electronic orders placed during the study period were obtained by extracting data from Epic’s Clarity database. Our EHR allows for the use of order sets; each order in these sets was counted individually, so that an order set with several orders would not be identified as one order. We identified the time and date that the order was placed, the ordering physician, the identity of the patient for which the order was placed, and the location of the patient when the order was placed, to determine the level of care (ICU, step-down, or general medicine unit). To track the composite volume of orders placed by resident teams, we matched each ordering physician to his or her corresponding resident team using our physician scheduling database, Amion (Spiral Software). We obtained team census by tabulating the total number of patients that a single resident team placed orders on over the course of a given calendar day. From billing data, we identified the MS-DRG weight that was assigned at the end of each hospitalization. Finally, we collected data on adherence to two discharge-related quality metrics to determine whether increased order volume was associated with decreased rates of adherence to these metrics. Using departmental patient-level quality improvement data, we determined whether each metric was met on discharge at the patient level. We also extracted patient-level demographic data, including age, sex, and insurance status, from this departmental quality improvement database.

Discharge Quality Outcome Metrics

We hypothesized that as the total daily electronic orders of a resident team increased, the rate of completion of two discharge-related quality metrics would decline due to the greater time constraints placed on the teams. The first metric we used was the completion of a high-quality after-visit summary (AVS), which has been described by the Centers for Medicare and Medicaid Services as part of its Meaningful Use Initiative.¹³ It was selected by the residents in our program as a particularly high-priority quality metric. Our institution specifically defines a “high-quality” AVS as including the following three components: a principal hospital problem, patient instructions, and follow-up information. The second discharge-related quality metric was the completion of a timely discharge summary, another measure recognized as a critical component in high-quality care.¹⁴ To be considered timely, the discharge summary had to be filed no later than 24 hours after the discharge order was entered into the EHR. This metric was more recently tracked by the internal medicine department and was not selected by the residents as a high-priority metric.

Statistical Analysis

To examine how the order volume per day changed throughout each sequential day of hospital admission, mean orders per hospital day with 95% CIs were plotted. We performed an aggregate analysis of all orders placed for each patient per day across three different levels of care (ICU, step-down, and general medicine). For each day of the study period, we summed all orders for all patients according to their location and divided by the number of total patients in each location to identify the average number of orders written for an ICU, step-down, and general medicine patient that day. We then calculated the mean daily orders for an ICU, step-down, and general medicine patient over the entire study period. We used ANOVA to test for statistically significant differences between the mean daily orders between these locations.

To examine the relationship between severity of illness and order volume, we performed an unadjusted patient-level analysis of orders per patient in the first three days of each hospitalization and stratified the data by the MS-DRG payment weight, which we divided into four quartiles. For each quartile, we calculated the mean number of orders placed in the first three days of admission and used ANOVA to test for statistically significant differences. We restricted the orders to the first three days of hospitalization instead of calculating mean orders per day of hospitalization because we postulated that the majority of orders were entered in these first few days and that with increasing length of stay (which we expected to occur with higher MS-DRG weight), the order volume becomes highly variable, which would tend to skew the mean orders per day.

We used multivariable logistic regression to determine whether the volume of electronic orders on the day of a given patient’s discharge, and also on the day before a given patient’s discharge, was a significant predictor of receiving a high-quality AVS. We adjusted for team census on the day of discharge, MS-DRG weight, age, sex, and insurance status. We then conducted a separate analysis of the association between electronic order volume and likelihood of completing a timely discharge summary among patients where discharge summary data were available. Logistic regression for each case was performed independently, so that team orders on the day prior to a patient’s discharge were not included in the model for the relationship between team orders on the day of a patient’s discharge and the discharge-related quality metric of interest, and vice versa, since including both in the model would be potentially disruptive given that orders on the day before and day of a patient’s discharge are likely correlated.

We also performed a subanalysis in which we restricted orders to only those placed during the daytime hours (7 am-7 pm), since these reflect the work performed by the primary team, and excluded those placed by covering night-shift residents.

IRB Approval

The study was approved by the UCSF Institutional Review Board and was granted a waiver of informed consent.

Population

We identified 7,296 eligible hospitalizations during the study period. After removing hospitalizations according to our exclusion criteria (Figure 1), there were 5,032 hospitalizations that were used in the analysis for which a total of 929,153 orders were written. The vast majority of patients received at least one order per day; fewer than 1% of encounter-days had zero associated orders. The top 10 discharge diagnoses identified in the cohort are listed in Appendix Table 1. A breakdown of orders by order type, across the entire cohort, is displayed in Appendix Table 2. The mean number of orders per patient per day of hospitalization is plotted in the Appendix Figure, which indicates that the number of orders is highest on the day of admission, decreases significantly after the first few days, and becomes increasingly variable with longer lengths of stay.

Patient Level of Care and Severity of Illness Metrics

Patients at a higher level of care had, on average, more orders entered per day. The mean order frequency was 40 orders per day for an ICU patient (standard deviation [SD] 13, range 13-134), 24 for a step-down patient (SD 6, range 11-48), and 19 for a general medicine unit patient (SD 3, range 10-31). The difference in mean daily orders was statistically significant (P < .001, Figure 2a).

Orders also correlated with increasing severity of illness. Patients in the lowest quartile of MS-DRG weight received, on average, 98 orders in the first three days of hospitalization (SD 35, range 2-349), those in the second quartile received 105 orders (SD 38, range 10-380), those in the third quartile received 132 orders (SD 51, range 17-436), and those in the fourth and highest quartile received 149 orders (SD 59, range 32-482). Comparisons between each of these severity of illness categories were significant (P < .001, Figure 2b).

Discharge-Related Quality Metrics

The median number of orders per internal medicine team per day was 343 (IQR 261- 446). Of the 5,032 total discharged patients, 3,657 (73%) received a high-quality AVS on discharge. After controlling for team census, severity of illness, and demographic factors, there was no statistically significant association between total orders on the day of discharge and odds of receiving a high-quality AVS (OR 1.01; 95% CI 0.96-1.06), or between team orders placed the day prior to discharge and odds of receiving a high-quality AVS (OR 0.99; 95% CI 0.95-1.04; Table 1). When we restricted our analysis to orders placed during daytime hours (7 am-7 pm), these findings were largely unchanged (OR 1.05; 95% CI 0.97-1.14 for orders on the day of discharge; OR 1.02; 95% CI 0.95-1.10 for orders on the day before discharge).

There were 3,835 patients for whom data on timing of discharge summary were available. Of these, 3,455 (91.2%) had a discharge summary completed within 24 hours. After controlling for team census, severity of illness, and demographic factors, there was no statistically significant association between total orders placed by the team on a patient’s day of discharge and odds of receiving a timely discharge summary (OR 0.96; 95% CI 0.88-1.05). However, patients were 12% less likely to receive a timely discharge summary for every 100 extra orders the team placed on the day prior to discharge (OR 0.88, 95% CI 0.82-0.95). Patients who received a timely discharge summary were cared for by teams who placed a median of 345 orders the day prior to their discharge, whereas those that did not receive a timely discharge summary were cared for by teams who placed a significantly higher number of orders (375) on the day prior to discharge (Table 2). When we restricted our analysis to only daytime orders, there were no significant changes in the findings (OR 1.00; 95% CI 0.88-1.14 for orders on the day of discharge; OR 0.84; 95% CI 0.75-0.95 for orders on the day prior to discharge).

We found that electronic order volume may be a marker for patient complexity, which encompasses both level of care and severity of illness, and could be a marker of resident physician workload that harnesses readily available data from an EHR. Recent time-motion studies of internal medicine residents indicate that the majority of trainees’ time is spent on computers, engaged in indirect patient care activities such as reading electronic charts, entering electronic orders, and writing computerized notes.^15-18 Capturing these tasks through metrics such as electronic order volume, as we did in this study, can provide valuable insights into resident physician workflow.

We found that ICU patients received more than twice as many orders per day than did general acute care-level patients. Furthermore, we found that patients whose hospitalizations fell into the highest MS-DRG weight quartile received approximately 50% more orders during the first three days of admission compared to that of patients whose hospitalizations fell into the lowest quartile. This strong association indicates that electronic order volume could provide meaningful additional information, in concert with other factors such as census, to describe resident physician workload.

We did not find that our workload measure was significantly associated with high-quality AVS completion. There are several possible explanations for this finding. First, adherence to this quality metric may be independent of workload, possibly because it is highly prioritized by residents at our institution. Second, adherence may only be impacted at levels of workload greater than what was experienced by the residents in our study. Finally, electronic order volume may not encompass enough of total workload to be reliably representative of resident work. However, the tight correlation between electronic order volume with severity of illness and level of care, in conjunction with the finding that patients were less likely to receive a timely discharge summary when workload was high on the day prior to a patient’s discharge, suggests that electronic order volume does indeed encompass a meaningful component of workload, and that with higher workload, adherence to some quality metrics may decline. We found that patients who received a timely discharge summary were discharged by teams who entered 30 fewer orders on the day before discharge compared with patients who did not receive a timely discharge summary. In addition to being statistically significant, it is also likely that this difference is clinically significant, although a determination of clinical significance is outside the scope of this study. Further exploration into the relationship between order volume and other quality metrics that are perhaps more sensitive to workload would be interesting.

The primary strength of our study is in how it demonstrates that EHRs can be harnessed to provide additional insights into clinical workload in a quantifiable and automated manner. Although there are a wide range of EHRs currently in use across the country, the capability to track electronic orders is common and could therefore be used broadly across institutions, with tailoring and standardization specific to each site. This technique is similar to that used by prior investigators who characterized the workload of pediatric residents by orders entered and notes written in the electronic medical record.¹⁹ However, our study is unique, in that we explored the relationship between electronic order volume and patient-level severity metrics as well as discharge-related quality metrics.

Our study is limited by several factors. When conceptualizing resident workload, several other elements that contribute to a sense of “busyness” may be independent of electronic orders and were not measured in our study.²⁰ These include communication factors (such as language discordance, discussion with consulting services, and difficult end-of-life discussions), environmental factors (such as geographic localization), resident physician team factors (such as competing clinical or educational responsibilities), timing (in terms of day of week as well as time of year, since residents in July likely feel “busier” than residents in May), and ultimate discharge destination for patients (those going to a skilled nursing facility may require discharge documentation more urgently). Additionally, we chose to focus on the workload of resident teams, as represented by team orders, as opposed to individual work, which may be more directly correlated to our outcomes of interest, completion of a high-quality AVS, and timely discharge summary, which are usually performed by individuals.

Furthermore, we did not measure the relationship between our objective measure of workload and clinical endpoints. Instead, we chose to focus on process measures because they are less likely to be confounded by clinical factors independent of physician workload.²¹ Future studies should also consider obtaining direct resident-level measures of “busyness” or burnout, or other resident-centered endpoints, such as whether residents left the hospital at times consistent with duty hour regulations or whether they were able to attend educational conferences.

These limitations pose opportunities for further efforts to more comprehensively characterize clinical workload. Additional research is needed to understand and quantify the impact of patient, physician, and environmental factors that are not reflected by electronic order volume. Furthermore, an exploration of other electronic surrogates for clinical workload, such as paging volume and other EHR-derived data points, could also prove valuable in further describing the clinical workload. Future studies should also examine whether there is a relationship between these novel markers of workload and further outcomes, including both process measures and clinical endpoints.

Electronic order volume may provide valuable additional information for estimating the workload of resident physicians caring for hospitalized patients. Further investigation to determine whether the statistically significant differences identified in this study are clinically significant, how the technique used in this work may be applied to different EHRs, an examination of other EHR-derived metrics that may represent workload, and an exploration of additional patient-centered outcomes may be warranted.

Disclosures

Rajkomar reports personal fees from Google LLC, outside the submitted work. Dr. Khanna reports that during the conduct of the study, his salary, and the development of CareWeb (a communication platform that includes a smartphone-based paging application in use in several inpatient clinical units at University of California, San Francisco [UCSF] Medical Center) were supported by funding from the Center for Digital Health Innovation at UCSF. The CareWeb software has been licensed by Voalte.

Disclaimer

The views expressed in the submitted article are of the authors and not an official position of the institution.

Resident physician workload has traditionally been measured by patient census.^1,2 However, census and other volume-based metrics such as daily admissions may not accurately reflect workload due to variation in patient complexity. Relative value units (RVUs) are another commonly used marker of workload, but the validity of this metric relies on accurate coding, usually done by the attending physician, and is less directly related to resident physician workload. Because much of hospital-based medicine is mediated through the electronic health record (EHR), which can capture differences in patient complexity,³ electronic records could be harnessed to more comprehensively describe residents’ work. Current government estimates indicate that several hundred companies offer certified EHRs, thanks in large part to the Health Information Technology for Economic and Clinical Health (HITECH) Act of 2009, which aimed to promote adoption and meaningful use of health information technology.^{4, 5} These systems can collect important data about the usage and operating patterns of physicians, which may provide an insight into workload.^6-8

Accurately measuring workload is important because of the direct link that has been drawn between physician workload and quality metrics. In a study of attending hospitalists, higher workload, as measured by patient census and RVUs, was associated with longer lengths of stay and higher costs of hospitalization.⁹ Another study among medical residents found that as daily admissions increased, length of stay, cost, and inpatient mortality appeared to rise.¹⁰ Although these studies used only volume-based workload metrics, the implication that high workload may negatively impact patient care hints at a possible trade-off between the two that should inform discussions of physician productivity.

In the current study, we examine whether data obtained from the EHR, particularly electronic order volume, could provide valuable information, in addition to patient volume, about resident physician workload. We first tested the feasibility and validity of using electronic order volume as an important component of clinical workload by examining the relationship between electronic order volume and well-established factors that are likely to increase the workload of residents, including patient level of care and severity of illness. Then, using order volume as a marker for workload, we sought to describe whether higher order volumes were associated with two discharge-related quality metrics, completion of a high-quality after-visit summary and timely discharge summary, postulating that quality metrics may suffer when residents are busier.

Study Design and Setting

We performed a single-center retrospective cohort study of patients admitted to the internal medicine service at the University of California, San Francisco (UCSF) Medical Center between May 1, 2015 and July 31, 2016. UCSF is a 600-bed academic medical center, and the inpatient internal medicine teaching service manages an average daily census of 80-90 patients. Medicine teams care for patients on the general acute-care wards, the step-down units (for patients with higher acuity of care), and also patients in the intensive care unit (ICU). ICU patients are comanaged by general medicine teams and intensive care teams; internal medicine teams enter all electronic orders for ICU patients, except for orders for respiratory care or sedating medications. The inpatient internal medicine teaching service comprises eight teams each supervised by an attending physician, a senior resident (in the second or third year of residency training), two interns, and a third- and/or fourth-year medical student. Residents place all clinical orders and complete all clinical documentation through the EHR (Epic Systems, Verona, Wisconsin).¹¹ Typically, the bulk of the orders and documentation, including discharge documentation, is performed by interns; however, the degree of senior resident involvement in these tasks is variable and team-dependent. In addition to the eight resident teams, there are also four attending hospitalist-only internal medicine teams, who manage a service of ~30-40 patients.

Study Population

Our study population comprised all hospitalized adults admitted to the eight resident-run teams on the internal medicine teaching service. Patients cared for by hospitalist-only teams were not included in this analysis. Because the focus of our study was on hospitalizations, individual patients may have been included multiple times over the course of the study. Hospitalizations were excluded if they did not have complete Medicare Severity-Diagnosis Related Group (MS-DRG) data,¹² since this was used as our severity of illness marker. This occurred either because patients were not discharged by the end of the study period or because they had a length of stay of less than one day, because this metric was not assigned to these short-stay (observation) patients.

Data Collection

All electronic orders placed during the study period were obtained by extracting data from Epic’s Clarity database. Our EHR allows for the use of order sets; each order in these sets was counted individually, so that an order set with several orders would not be identified as one order. We identified the time and date that the order was placed, the ordering physician, the identity of the patient for which the order was placed, and the location of the patient when the order was placed, to determine the level of care (ICU, step-down, or general medicine unit). To track the composite volume of orders placed by resident teams, we matched each ordering physician to his or her corresponding resident team using our physician scheduling database, Amion (Spiral Software). We obtained team census by tabulating the total number of patients that a single resident team placed orders on over the course of a given calendar day. From billing data, we identified the MS-DRG weight that was assigned at the end of each hospitalization. Finally, we collected data on adherence to two discharge-related quality metrics to determine whether increased order volume was associated with decreased rates of adherence to these metrics. Using departmental patient-level quality improvement data, we determined whether each metric was met on discharge at the patient level. We also extracted patient-level demographic data, including age, sex, and insurance status, from this departmental quality improvement database.

Discharge Quality Outcome Metrics

We hypothesized that as the total daily electronic orders of a resident team increased, the rate of completion of two discharge-related quality metrics would decline due to the greater time constraints placed on the teams. The first metric we used was the completion of a high-quality after-visit summary (AVS), which has been described by the Centers for Medicare and Medicaid Services as part of its Meaningful Use Initiative.¹³ It was selected by the residents in our program as a particularly high-priority quality metric. Our institution specifically defines a “high-quality” AVS as including the following three components: a principal hospital problem, patient instructions, and follow-up information. The second discharge-related quality metric was the completion of a timely discharge summary, another measure recognized as a critical component in high-quality care.¹⁴ To be considered timely, the discharge summary had to be filed no later than 24 hours after the discharge order was entered into the EHR. This metric was more recently tracked by the internal medicine department and was not selected by the residents as a high-priority metric.

Statistical Analysis

To examine how the order volume per day changed throughout each sequential day of hospital admission, mean orders per hospital day with 95% CIs were plotted. We performed an aggregate analysis of all orders placed for each patient per day across three different levels of care (ICU, step-down, and general medicine). For each day of the study period, we summed all orders for all patients according to their location and divided by the number of total patients in each location to identify the average number of orders written for an ICU, step-down, and general medicine patient that day. We then calculated the mean daily orders for an ICU, step-down, and general medicine patient over the entire study period. We used ANOVA to test for statistically significant differences between the mean daily orders between these locations.

To examine the relationship between severity of illness and order volume, we performed an unadjusted patient-level analysis of orders per patient in the first three days of each hospitalization and stratified the data by the MS-DRG payment weight, which we divided into four quartiles. For each quartile, we calculated the mean number of orders placed in the first three days of admission and used ANOVA to test for statistically significant differences. We restricted the orders to the first three days of hospitalization instead of calculating mean orders per day of hospitalization because we postulated that the majority of orders were entered in these first few days and that with increasing length of stay (which we expected to occur with higher MS-DRG weight), the order volume becomes highly variable, which would tend to skew the mean orders per day.

We used multivariable logistic regression to determine whether the volume of electronic orders on the day of a given patient’s discharge, and also on the day before a given patient’s discharge, was a significant predictor of receiving a high-quality AVS. We adjusted for team census on the day of discharge, MS-DRG weight, age, sex, and insurance status. We then conducted a separate analysis of the association between electronic order volume and likelihood of completing a timely discharge summary among patients where discharge summary data were available. Logistic regression for each case was performed independently, so that team orders on the day prior to a patient’s discharge were not included in the model for the relationship between team orders on the day of a patient’s discharge and the discharge-related quality metric of interest, and vice versa, since including both in the model would be potentially disruptive given that orders on the day before and day of a patient’s discharge are likely correlated.

We also performed a subanalysis in which we restricted orders to only those placed during the daytime hours (7 am-7 pm), since these reflect the work performed by the primary team, and excluded those placed by covering night-shift residents.

IRB Approval

The study was approved by the UCSF Institutional Review Board and was granted a waiver of informed consent.

Population

We identified 7,296 eligible hospitalizations during the study period. After removing hospitalizations according to our exclusion criteria (Figure 1), there were 5,032 hospitalizations that were used in the analysis for which a total of 929,153 orders were written. The vast majority of patients received at least one order per day; fewer than 1% of encounter-days had zero associated orders. The top 10 discharge diagnoses identified in the cohort are listed in Appendix Table 1. A breakdown of orders by order type, across the entire cohort, is displayed in Appendix Table 2. The mean number of orders per patient per day of hospitalization is plotted in the Appendix Figure, which indicates that the number of orders is highest on the day of admission, decreases significantly after the first few days, and becomes increasingly variable with longer lengths of stay.

Patient Level of Care and Severity of Illness Metrics

Patients at a higher level of care had, on average, more orders entered per day. The mean order frequency was 40 orders per day for an ICU patient (standard deviation [SD] 13, range 13-134), 24 for a step-down patient (SD 6, range 11-48), and 19 for a general medicine unit patient (SD 3, range 10-31). The difference in mean daily orders was statistically significant (P < .001, Figure 2a).

Orders also correlated with increasing severity of illness. Patients in the lowest quartile of MS-DRG weight received, on average, 98 orders in the first three days of hospitalization (SD 35, range 2-349), those in the second quartile received 105 orders (SD 38, range 10-380), those in the third quartile received 132 orders (SD 51, range 17-436), and those in the fourth and highest quartile received 149 orders (SD 59, range 32-482). Comparisons between each of these severity of illness categories were significant (P < .001, Figure 2b).

Discharge-Related Quality Metrics

The median number of orders per internal medicine team per day was 343 (IQR 261- 446). Of the 5,032 total discharged patients, 3,657 (73%) received a high-quality AVS on discharge. After controlling for team census, severity of illness, and demographic factors, there was no statistically significant association between total orders on the day of discharge and odds of receiving a high-quality AVS (OR 1.01; 95% CI 0.96-1.06), or between team orders placed the day prior to discharge and odds of receiving a high-quality AVS (OR 0.99; 95% CI 0.95-1.04; Table 1). When we restricted our analysis to orders placed during daytime hours (7 am-7 pm), these findings were largely unchanged (OR 1.05; 95% CI 0.97-1.14 for orders on the day of discharge; OR 1.02; 95% CI 0.95-1.10 for orders on the day before discharge).

There were 3,835 patients for whom data on timing of discharge summary were available. Of these, 3,455 (91.2%) had a discharge summary completed within 24 hours. After controlling for team census, severity of illness, and demographic factors, there was no statistically significant association between total orders placed by the team on a patient’s day of discharge and odds of receiving a timely discharge summary (OR 0.96; 95% CI 0.88-1.05). However, patients were 12% less likely to receive a timely discharge summary for every 100 extra orders the team placed on the day prior to discharge (OR 0.88, 95% CI 0.82-0.95). Patients who received a timely discharge summary were cared for by teams who placed a median of 345 orders the day prior to their discharge, whereas those that did not receive a timely discharge summary were cared for by teams who placed a significantly higher number of orders (375) on the day prior to discharge (Table 2). When we restricted our analysis to only daytime orders, there were no significant changes in the findings (OR 1.00; 95% CI 0.88-1.14 for orders on the day of discharge; OR 0.84; 95% CI 0.75-0.95 for orders on the day prior to discharge).

We found that electronic order volume may be a marker for patient complexity, which encompasses both level of care and severity of illness, and could be a marker of resident physician workload that harnesses readily available data from an EHR. Recent time-motion studies of internal medicine residents indicate that the majority of trainees’ time is spent on computers, engaged in indirect patient care activities such as reading electronic charts, entering electronic orders, and writing computerized notes.^15-18 Capturing these tasks through metrics such as electronic order volume, as we did in this study, can provide valuable insights into resident physician workflow.

We found that ICU patients received more than twice as many orders per day than did general acute care-level patients. Furthermore, we found that patients whose hospitalizations fell into the highest MS-DRG weight quartile received approximately 50% more orders during the first three days of admission compared to that of patients whose hospitalizations fell into the lowest quartile. This strong association indicates that electronic order volume could provide meaningful additional information, in concert with other factors such as census, to describe resident physician workload.

We did not find that our workload measure was significantly associated with high-quality AVS completion. There are several possible explanations for this finding. First, adherence to this quality metric may be independent of workload, possibly because it is highly prioritized by residents at our institution. Second, adherence may only be impacted at levels of workload greater than what was experienced by the residents in our study. Finally, electronic order volume may not encompass enough of total workload to be reliably representative of resident work. However, the tight correlation between electronic order volume with severity of illness and level of care, in conjunction with the finding that patients were less likely to receive a timely discharge summary when workload was high on the day prior to a patient’s discharge, suggests that electronic order volume does indeed encompass a meaningful component of workload, and that with higher workload, adherence to some quality metrics may decline. We found that patients who received a timely discharge summary were discharged by teams who entered 30 fewer orders on the day before discharge compared with patients who did not receive a timely discharge summary. In addition to being statistically significant, it is also likely that this difference is clinically significant, although a determination of clinical significance is outside the scope of this study. Further exploration into the relationship between order volume and other quality metrics that are perhaps more sensitive to workload would be interesting.

The primary strength of our study is in how it demonstrates that EHRs can be harnessed to provide additional insights into clinical workload in a quantifiable and automated manner. Although there are a wide range of EHRs currently in use across the country, the capability to track electronic orders is common and could therefore be used broadly across institutions, with tailoring and standardization specific to each site. This technique is similar to that used by prior investigators who characterized the workload of pediatric residents by orders entered and notes written in the electronic medical record.¹⁹ However, our study is unique, in that we explored the relationship between electronic order volume and patient-level severity metrics as well as discharge-related quality metrics.

Our study is limited by several factors. When conceptualizing resident workload, several other elements that contribute to a sense of “busyness” may be independent of electronic orders and were not measured in our study.²⁰ These include communication factors (such as language discordance, discussion with consulting services, and difficult end-of-life discussions), environmental factors (such as geographic localization), resident physician team factors (such as competing clinical or educational responsibilities), timing (in terms of day of week as well as time of year, since residents in July likely feel “busier” than residents in May), and ultimate discharge destination for patients (those going to a skilled nursing facility may require discharge documentation more urgently). Additionally, we chose to focus on the workload of resident teams, as represented by team orders, as opposed to individual work, which may be more directly correlated to our outcomes of interest, completion of a high-quality AVS, and timely discharge summary, which are usually performed by individuals.

Furthermore, we did not measure the relationship between our objective measure of workload and clinical endpoints. Instead, we chose to focus on process measures because they are less likely to be confounded by clinical factors independent of physician workload.²¹ Future studies should also consider obtaining direct resident-level measures of “busyness” or burnout, or other resident-centered endpoints, such as whether residents left the hospital at times consistent with duty hour regulations or whether they were able to attend educational conferences.

These limitations pose opportunities for further efforts to more comprehensively characterize clinical workload. Additional research is needed to understand and quantify the impact of patient, physician, and environmental factors that are not reflected by electronic order volume. Furthermore, an exploration of other electronic surrogates for clinical workload, such as paging volume and other EHR-derived data points, could also prove valuable in further describing the clinical workload. Future studies should also examine whether there is a relationship between these novel markers of workload and further outcomes, including both process measures and clinical endpoints.

Electronic order volume may provide valuable additional information for estimating the workload of resident physicians caring for hospitalized patients. Further investigation to determine whether the statistically significant differences identified in this study are clinically significant, how the technique used in this work may be applied to different EHRs, an examination of other EHR-derived metrics that may represent workload, and an exploration of additional patient-centered outcomes may be warranted.

Disclosures

Rajkomar reports personal fees from Google LLC, outside the submitted work. Dr. Khanna reports that during the conduct of the study, his salary, and the development of CareWeb (a communication platform that includes a smartphone-based paging application in use in several inpatient clinical units at University of California, San Francisco [UCSF] Medical Center) were supported by funding from the Center for Digital Health Innovation at UCSF. The CareWeb software has been licensed by Voalte.

Disclaimer

The views expressed in the submitted article are of the authors and not an official position of the institution.

Resident physician workload has traditionally been measured by patient census.^1,2 However, census and other volume-based metrics such as daily admissions may not accurately reflect workload due to variation in patient complexity. Relative value units (RVUs) are another commonly used marker of workload, but the validity of this metric relies on accurate coding, usually done by the attending physician, and is less directly related to resident physician workload. Because much of hospital-based medicine is mediated through the electronic health record (EHR), which can capture differences in patient complexity,³ electronic records could be harnessed to more comprehensively describe residents’ work. Current government estimates indicate that several hundred companies offer certified EHRs, thanks in large part to the Health Information Technology for Economic and Clinical Health (HITECH) Act of 2009, which aimed to promote adoption and meaningful use of health information technology.^{4, 5} These systems can collect important data about the usage and operating patterns of physicians, which may provide an insight into workload.^6-8

Accurately measuring workload is important because of the direct link that has been drawn between physician workload and quality metrics. In a study of attending hospitalists, higher workload, as measured by patient census and RVUs, was associated with longer lengths of stay and higher costs of hospitalization.⁹ Another study among medical residents found that as daily admissions increased, length of stay, cost, and inpatient mortality appeared to rise.¹⁰ Although these studies used only volume-based workload metrics, the implication that high workload may negatively impact patient care hints at a possible trade-off between the two that should inform discussions of physician productivity.

In the current study, we examine whether data obtained from the EHR, particularly electronic order volume, could provide valuable information, in addition to patient volume, about resident physician workload. We first tested the feasibility and validity of using electronic order volume as an important component of clinical workload by examining the relationship between electronic order volume and well-established factors that are likely to increase the workload of residents, including patient level of care and severity of illness. Then, using order volume as a marker for workload, we sought to describe whether higher order volumes were associated with two discharge-related quality metrics, completion of a high-quality after-visit summary and timely discharge summary, postulating that quality metrics may suffer when residents are busier.

Study Design and Setting

We performed a single-center retrospective cohort study of patients admitted to the internal medicine service at the University of California, San Francisco (UCSF) Medical Center between May 1, 2015 and July 31, 2016. UCSF is a 600-bed academic medical center, and the inpatient internal medicine teaching service manages an average daily census of 80-90 patients. Medicine teams care for patients on the general acute-care wards, the step-down units (for patients with higher acuity of care), and also patients in the intensive care unit (ICU). ICU patients are comanaged by general medicine teams and intensive care teams; internal medicine teams enter all electronic orders for ICU patients, except for orders for respiratory care or sedating medications. The inpatient internal medicine teaching service comprises eight teams each supervised by an attending physician, a senior resident (in the second or third year of residency training), two interns, and a third- and/or fourth-year medical student. Residents place all clinical orders and complete all clinical documentation through the EHR (Epic Systems, Verona, Wisconsin).¹¹ Typically, the bulk of the orders and documentation, including discharge documentation, is performed by interns; however, the degree of senior resident involvement in these tasks is variable and team-dependent. In addition to the eight resident teams, there are also four attending hospitalist-only internal medicine teams, who manage a service of ~30-40 patients.

Study Population

Our study population comprised all hospitalized adults admitted to the eight resident-run teams on the internal medicine teaching service. Patients cared for by hospitalist-only teams were not included in this analysis. Because the focus of our study was on hospitalizations, individual patients may have been included multiple times over the course of the study. Hospitalizations were excluded if they did not have complete Medicare Severity-Diagnosis Related Group (MS-DRG) data,¹² since this was used as our severity of illness marker. This occurred either because patients were not discharged by the end of the study period or because they had a length of stay of less than one day, because this metric was not assigned to these short-stay (observation) patients.

Data Collection

All electronic orders placed during the study period were obtained by extracting data from Epic’s Clarity database. Our EHR allows for the use of order sets; each order in these sets was counted individually, so that an order set with several orders would not be identified as one order. We identified the time and date that the order was placed, the ordering physician, the identity of the patient for which the order was placed, and the location of the patient when the order was placed, to determine the level of care (ICU, step-down, or general medicine unit). To track the composite volume of orders placed by resident teams, we matched each ordering physician to his or her corresponding resident team using our physician scheduling database, Amion (Spiral Software). We obtained team census by tabulating the total number of patients that a single resident team placed orders on over the course of a given calendar day. From billing data, we identified the MS-DRG weight that was assigned at the end of each hospitalization. Finally, we collected data on adherence to two discharge-related quality metrics to determine whether increased order volume was associated with decreased rates of adherence to these metrics. Using departmental patient-level quality improvement data, we determined whether each metric was met on discharge at the patient level. We also extracted patient-level demographic data, including age, sex, and insurance status, from this departmental quality improvement database.

Discharge Quality Outcome Metrics

We hypothesized that as the total daily electronic orders of a resident team increased, the rate of completion of two discharge-related quality metrics would decline due to the greater time constraints placed on the teams. The first metric we used was the completion of a high-quality after-visit summary (AVS), which has been described by the Centers for Medicare and Medicaid Services as part of its Meaningful Use Initiative.¹³ It was selected by the residents in our program as a particularly high-priority quality metric. Our institution specifically defines a “high-quality” AVS as including the following three components: a principal hospital problem, patient instructions, and follow-up information. The second discharge-related quality metric was the completion of a timely discharge summary, another measure recognized as a critical component in high-quality care.¹⁴ To be considered timely, the discharge summary had to be filed no later than 24 hours after the discharge order was entered into the EHR. This metric was more recently tracked by the internal medicine department and was not selected by the residents as a high-priority metric.

Statistical Analysis

To examine how the order volume per day changed throughout each sequential day of hospital admission, mean orders per hospital day with 95% CIs were plotted. We performed an aggregate analysis of all orders placed for each patient per day across three different levels of care (ICU, step-down, and general medicine). For each day of the study period, we summed all orders for all patients according to their location and divided by the number of total patients in each location to identify the average number of orders written for an ICU, step-down, and general medicine patient that day. We then calculated the mean daily orders for an ICU, step-down, and general medicine patient over the entire study period. We used ANOVA to test for statistically significant differences between the mean daily orders between these locations.

To examine the relationship between severity of illness and order volume, we performed an unadjusted patient-level analysis of orders per patient in the first three days of each hospitalization and stratified the data by the MS-DRG payment weight, which we divided into four quartiles. For each quartile, we calculated the mean number of orders placed in the first three days of admission and used ANOVA to test for statistically significant differences. We restricted the orders to the first three days of hospitalization instead of calculating mean orders per day of hospitalization because we postulated that the majority of orders were entered in these first few days and that with increasing length of stay (which we expected to occur with higher MS-DRG weight), the order volume becomes highly variable, which would tend to skew the mean orders per day.

We used multivariable logistic regression to determine whether the volume of electronic orders on the day of a given patient’s discharge, and also on the day before a given patient’s discharge, was a significant predictor of receiving a high-quality AVS. We adjusted for team census on the day of discharge, MS-DRG weight, age, sex, and insurance status. We then conducted a separate analysis of the association between electronic order volume and likelihood of completing a timely discharge summary among patients where discharge summary data were available. Logistic regression for each case was performed independently, so that team orders on the day prior to a patient’s discharge were not included in the model for the relationship between team orders on the day of a patient’s discharge and the discharge-related quality metric of interest, and vice versa, since including both in the model would be potentially disruptive given that orders on the day before and day of a patient’s discharge are likely correlated.

We also performed a subanalysis in which we restricted orders to only those placed during the daytime hours (7 am-7 pm), since these reflect the work performed by the primary team, and excluded those placed by covering night-shift residents.

IRB Approval

The study was approved by the UCSF Institutional Review Board and was granted a waiver of informed consent.

Population

We identified 7,296 eligible hospitalizations during the study period. After removing hospitalizations according to our exclusion criteria (Figure 1), there were 5,032 hospitalizations that were used in the analysis for which a total of 929,153 orders were written. The vast majority of patients received at least one order per day; fewer than 1% of encounter-days had zero associated orders. The top 10 discharge diagnoses identified in the cohort are listed in Appendix Table 1. A breakdown of orders by order type, across the entire cohort, is displayed in Appendix Table 2. The mean number of orders per patient per day of hospitalization is plotted in the Appendix Figure, which indicates that the number of orders is highest on the day of admission, decreases significantly after the first few days, and becomes increasingly variable with longer lengths of stay.

Patient Level of Care and Severity of Illness Metrics

Patients at a higher level of care had, on average, more orders entered per day. The mean order frequency was 40 orders per day for an ICU patient (standard deviation [SD] 13, range 13-134), 24 for a step-down patient (SD 6, range 11-48), and 19 for a general medicine unit patient (SD 3, range 10-31). The difference in mean daily orders was statistically significant (P < .001, Figure 2a).

Orders also correlated with increasing severity of illness. Patients in the lowest quartile of MS-DRG weight received, on average, 98 orders in the first three days of hospitalization (SD 35, range 2-349), those in the second quartile received 105 orders (SD 38, range 10-380), those in the third quartile received 132 orders (SD 51, range 17-436), and those in the fourth and highest quartile received 149 orders (SD 59, range 32-482). Comparisons between each of these severity of illness categories were significant (P < .001, Figure 2b).

Discharge-Related Quality Metrics

The median number of orders per internal medicine team per day was 343 (IQR 261- 446). Of the 5,032 total discharged patients, 3,657 (73%) received a high-quality AVS on discharge. After controlling for team census, severity of illness, and demographic factors, there was no statistically significant association between total orders on the day of discharge and odds of receiving a high-quality AVS (OR 1.01; 95% CI 0.96-1.06), or between team orders placed the day prior to discharge and odds of receiving a high-quality AVS (OR 0.99; 95% CI 0.95-1.04; Table 1). When we restricted our analysis to orders placed during daytime hours (7 am-7 pm), these findings were largely unchanged (OR 1.05; 95% CI 0.97-1.14 for orders on the day of discharge; OR 1.02; 95% CI 0.95-1.10 for orders on the day before discharge).

There were 3,835 patients for whom data on timing of discharge summary were available. Of these, 3,455 (91.2%) had a discharge summary completed within 24 hours. After controlling for team census, severity of illness, and demographic factors, there was no statistically significant association between total orders placed by the team on a patient’s day of discharge and odds of receiving a timely discharge summary (OR 0.96; 95% CI 0.88-1.05). However, patients were 12% less likely to receive a timely discharge summary for every 100 extra orders the team placed on the day prior to discharge (OR 0.88, 95% CI 0.82-0.95). Patients who received a timely discharge summary were cared for by teams who placed a median of 345 orders the day prior to their discharge, whereas those that did not receive a timely discharge summary were cared for by teams who placed a significantly higher number of orders (375) on the day prior to discharge (Table 2). When we restricted our analysis to only daytime orders, there were no significant changes in the findings (OR 1.00; 95% CI 0.88-1.14 for orders on the day of discharge; OR 0.84; 95% CI 0.75-0.95 for orders on the day prior to discharge).

We found that electronic order volume may be a marker for patient complexity, which encompasses both level of care and severity of illness, and could be a marker of resident physician workload that harnesses readily available data from an EHR. Recent time-motion studies of internal medicine residents indicate that the majority of trainees’ time is spent on computers, engaged in indirect patient care activities such as reading electronic charts, entering electronic orders, and writing computerized notes.^15-18 Capturing these tasks through metrics such as electronic order volume, as we did in this study, can provide valuable insights into resident physician workflow.

We found that ICU patients received more than twice as many orders per day than did general acute care-level patients. Furthermore, we found that patients whose hospitalizations fell into the highest MS-DRG weight quartile received approximately 50% more orders during the first three days of admission compared to that of patients whose hospitalizations fell into the lowest quartile. This strong association indicates that electronic order volume could provide meaningful additional information, in concert with other factors such as census, to describe resident physician workload.

We did not find that our workload measure was significantly associated with high-quality AVS completion. There are several possible explanations for this finding. First, adherence to this quality metric may be independent of workload, possibly because it is highly prioritized by residents at our institution. Second, adherence may only be impacted at levels of workload greater than what was experienced by the residents in our study. Finally, electronic order volume may not encompass enough of total workload to be reliably representative of resident work. However, the tight correlation between electronic order volume with severity of illness and level of care, in conjunction with the finding that patients were less likely to receive a timely discharge summary when workload was high on the day prior to a patient’s discharge, suggests that electronic order volume does indeed encompass a meaningful component of workload, and that with higher workload, adherence to some quality metrics may decline. We found that patients who received a timely discharge summary were discharged by teams who entered 30 fewer orders on the day before discharge compared with patients who did not receive a timely discharge summary. In addition to being statistically significant, it is also likely that this difference is clinically significant, although a determination of clinical significance is outside the scope of this study. Further exploration into the relationship between order volume and other quality metrics that are perhaps more sensitive to workload would be interesting.

The primary strength of our study is in how it demonstrates that EHRs can be harnessed to provide additional insights into clinical workload in a quantifiable and automated manner. Although there are a wide range of EHRs currently in use across the country, the capability to track electronic orders is common and could therefore be used broadly across institutions, with tailoring and standardization specific to each site. This technique is similar to that used by prior investigators who characterized the workload of pediatric residents by orders entered and notes written in the electronic medical record.¹⁹ However, our study is unique, in that we explored the relationship between electronic order volume and patient-level severity metrics as well as discharge-related quality metrics.

Our study is limited by several factors. When conceptualizing resident workload, several other elements that contribute to a sense of “busyness” may be independent of electronic orders and were not measured in our study.²⁰ These include communication factors (such as language discordance, discussion with consulting services, and difficult end-of-life discussions), environmental factors (such as geographic localization), resident physician team factors (such as competing clinical or educational responsibilities), timing (in terms of day of week as well as time of year, since residents in July likely feel “busier” than residents in May), and ultimate discharge destination for patients (those going to a skilled nursing facility may require discharge documentation more urgently). Additionally, we chose to focus on the workload of resident teams, as represented by team orders, as opposed to individual work, which may be more directly correlated to our outcomes of interest, completion of a high-quality AVS, and timely discharge summary, which are usually performed by individuals.

Furthermore, we did not measure the relationship between our objective measure of workload and clinical endpoints. Instead, we chose to focus on process measures because they are less likely to be confounded by clinical factors independent of physician workload.²¹ Future studies should also consider obtaining direct resident-level measures of “busyness” or burnout, or other resident-centered endpoints, such as whether residents left the hospital at times consistent with duty hour regulations or whether they were able to attend educational conferences.

These limitations pose opportunities for further efforts to more comprehensively characterize clinical workload. Additional research is needed to understand and quantify the impact of patient, physician, and environmental factors that are not reflected by electronic order volume. Furthermore, an exploration of other electronic surrogates for clinical workload, such as paging volume and other EHR-derived data points, could also prove valuable in further describing the clinical workload. Future studies should also examine whether there is a relationship between these novel markers of workload and further outcomes, including both process measures and clinical endpoints.

Electronic order volume may provide valuable additional information for estimating the workload of resident physicians caring for hospitalized patients. Further investigation to determine whether the statistically significant differences identified in this study are clinically significant, how the technique used in this work may be applied to different EHRs, an examination of other EHR-derived metrics that may represent workload, and an exploration of additional patient-centered outcomes may be warranted.

Disclosures

Rajkomar reports personal fees from Google LLC, outside the submitted work. Dr. Khanna reports that during the conduct of the study, his salary, and the development of CareWeb (a communication platform that includes a smartphone-based paging application in use in several inpatient clinical units at University of California, San Francisco [UCSF] Medical Center) were supported by funding from the Center for Digital Health Innovation at UCSF. The CareWeb software has been licensed by Voalte.

Disclaimer

The views expressed in the submitted article are of the authors and not an official position of the institution.