David Schmit, MD

Background: Fever is a common symptom in patients presenting to the ED. In patients with hemodialysis-dependent ESRD, the literature on febrile response during infection is scarce. In this study, authors compared ED triage temperatures of S. aureus bacteremic patients with and without hemodialysis-dependent ESRD.

Study design: Paired, retrospective cohort study.

Setting: Tertiary care referral center.

Synopsis: A total of 74 patients with methicillin-resistant or methicillin-susceptible S. aureus bacteremia were included in this study (37 patients with and 37 patients without hemodialysis-dependent ESRD). Upon triage, 54% (95% confidence interval, 38%-70%) and 82% (95% CI, 65%-91%) of hemodialysis and nonhemodialysis patients did not have a detectable fever (less than 100.4° F), respectively. The estimated mean ED triage temperatures were 100.5° F in the hemodialysis-dependent patients and 99.0° F in the non–hemodialysis-dependent patients (P < .001). The authors note the significant lack of fevers may be the result of insensitive methods for measuring body temperature, such as peripheral thermometers.

Bottom line: In this small retrospective cohort study, these data suggest a high incidence of afebrile bacteremia in patients with ESRD, especially those patients not dialysis dependent. This may lead to delays in obtaining blood cultures and initiating antibiotics. However, given the study design, the authors were unable to conclude a causal relationship between ESRD and febrile response.

Citation: Weatherall SL et al. Do bacteremic patients with end-stage renal disease have a fever when presenting to the emergency department? A paired, retrospective cohort study. BMC Emerg Med. 2020;20:2.

Dr. Schmit is a hospitalist and associate professor of medicine at University of Texas Health, San Antonio.

Background: Fever is a common symptom in patients presenting to the ED. In patients with hemodialysis-dependent ESRD, the literature on febrile response during infection is scarce. In this study, authors compared ED triage temperatures of S. aureus bacteremic patients with and without hemodialysis-dependent ESRD.

Study design: Paired, retrospective cohort study.

Setting: Tertiary care referral center.

Synopsis: A total of 74 patients with methicillin-resistant or methicillin-susceptible S. aureus bacteremia were included in this study (37 patients with and 37 patients without hemodialysis-dependent ESRD). Upon triage, 54% (95% confidence interval, 38%-70%) and 82% (95% CI, 65%-91%) of hemodialysis and nonhemodialysis patients did not have a detectable fever (less than 100.4° F), respectively. The estimated mean ED triage temperatures were 100.5° F in the hemodialysis-dependent patients and 99.0° F in the non–hemodialysis-dependent patients (P < .001). The authors note the significant lack of fevers may be the result of insensitive methods for measuring body temperature, such as peripheral thermometers.

Bottom line: In this small retrospective cohort study, these data suggest a high incidence of afebrile bacteremia in patients with ESRD, especially those patients not dialysis dependent. This may lead to delays in obtaining blood cultures and initiating antibiotics. However, given the study design, the authors were unable to conclude a causal relationship between ESRD and febrile response.

Citation: Weatherall SL et al. Do bacteremic patients with end-stage renal disease have a fever when presenting to the emergency department? A paired, retrospective cohort study. BMC Emerg Med. 2020;20:2.

Dr. Schmit is a hospitalist and associate professor of medicine at University of Texas Health, San Antonio.

Background: Fever is a common symptom in patients presenting to the ED. In patients with hemodialysis-dependent ESRD, the literature on febrile response during infection is scarce. In this study, authors compared ED triage temperatures of S. aureus bacteremic patients with and without hemodialysis-dependent ESRD.

Study design: Paired, retrospective cohort study.

Setting: Tertiary care referral center.

Synopsis: A total of 74 patients with methicillin-resistant or methicillin-susceptible S. aureus bacteremia were included in this study (37 patients with and 37 patients without hemodialysis-dependent ESRD). Upon triage, 54% (95% confidence interval, 38%-70%) and 82% (95% CI, 65%-91%) of hemodialysis and nonhemodialysis patients did not have a detectable fever (less than 100.4° F), respectively. The estimated mean ED triage temperatures were 100.5° F in the hemodialysis-dependent patients and 99.0° F in the non–hemodialysis-dependent patients (P < .001). The authors note the significant lack of fevers may be the result of insensitive methods for measuring body temperature, such as peripheral thermometers.

Bottom line: In this small retrospective cohort study, these data suggest a high incidence of afebrile bacteremia in patients with ESRD, especially those patients not dialysis dependent. This may lead to delays in obtaining blood cultures and initiating antibiotics. However, given the study design, the authors were unable to conclude a causal relationship between ESRD and febrile response.

Citation: Weatherall SL et al. Do bacteremic patients with end-stage renal disease have a fever when presenting to the emergency department? A paired, retrospective cohort study. BMC Emerg Med. 2020;20:2.

Dr. Schmit is a hospitalist and associate professor of medicine at University of Texas Health, San Antonio.

Background: Management of primary spontaneous pneumothorax is usually with the insertion of a chest tube and typically requires hospitalization. This procedure can result in pain, organ injury, bleeding, and infection, and, if unresolved, may require surgery, introducing additional risks and complications. Few data exist from randomized trials comparing conservative versus interventional management.

In complete-case analysis, 129 out of 131 (98.5%) patients in the intervention group had resolution within 8 weeks, compared with 118 of 125 (94.4%) in the conservative group (risk difference, –4.1 percentage points; 95% confidence interval, –8.6 to 0.5, P = .02 for noninferiority).

In sensitivity analysis, in which missing data after the 8-week period were imputed as treatment failures, re-expansion occurred in 129 out of 138 (93.5%) patients in the intervention group and 118 out of 143 (82.5%) in the conservative group (risk difference, –11.0 percentage points; 95% CI, –18.4 to –3.5), which is outside the noninferiority margin of –9.0.

Overall, 41 patients in the intervention group and 13 in the conservative group had at least one adverse event.

Bottom line: Missing data limit the ability to make strong conclusions, but this trial suggests that conservative management of primary spontaneous pneumothorax was noninferior to interventional management with lower risk of serious adverse events.

Citation: Brown SG et al. Conservative versus interventional treatment for spontaneous pneumothorax. N Engl J Med. 2020; 382:405-15.

Dr. Schmit is a hospitalist and associate professor of medicine at University of Texas Health, San Antonio.

Background: Management of primary spontaneous pneumothorax is usually with the insertion of a chest tube and typically requires hospitalization. This procedure can result in pain, organ injury, bleeding, and infection, and, if unresolved, may require surgery, introducing additional risks and complications. Few data exist from randomized trials comparing conservative versus interventional management.

In complete-case analysis, 129 out of 131 (98.5%) patients in the intervention group had resolution within 8 weeks, compared with 118 of 125 (94.4%) in the conservative group (risk difference, –4.1 percentage points; 95% confidence interval, –8.6 to 0.5, P = .02 for noninferiority).

In sensitivity analysis, in which missing data after the 8-week period were imputed as treatment failures, re-expansion occurred in 129 out of 138 (93.5%) patients in the intervention group and 118 out of 143 (82.5%) in the conservative group (risk difference, –11.0 percentage points; 95% CI, –18.4 to –3.5), which is outside the noninferiority margin of –9.0.

Overall, 41 patients in the intervention group and 13 in the conservative group had at least one adverse event.

Bottom line: Missing data limit the ability to make strong conclusions, but this trial suggests that conservative management of primary spontaneous pneumothorax was noninferior to interventional management with lower risk of serious adverse events.

Citation: Brown SG et al. Conservative versus interventional treatment for spontaneous pneumothorax. N Engl J Med. 2020; 382:405-15.

Dr. Schmit is a hospitalist and associate professor of medicine at University of Texas Health, San Antonio.

Background: Management of primary spontaneous pneumothorax is usually with the insertion of a chest tube and typically requires hospitalization. This procedure can result in pain, organ injury, bleeding, and infection, and, if unresolved, may require surgery, introducing additional risks and complications. Few data exist from randomized trials comparing conservative versus interventional management.

In complete-case analysis, 129 out of 131 (98.5%) patients in the intervention group had resolution within 8 weeks, compared with 118 of 125 (94.4%) in the conservative group (risk difference, –4.1 percentage points; 95% confidence interval, –8.6 to 0.5, P = .02 for noninferiority).

In sensitivity analysis, in which missing data after the 8-week period were imputed as treatment failures, re-expansion occurred in 129 out of 138 (93.5%) patients in the intervention group and 118 out of 143 (82.5%) in the conservative group (risk difference, –11.0 percentage points; 95% CI, –18.4 to –3.5), which is outside the noninferiority margin of –9.0.

Overall, 41 patients in the intervention group and 13 in the conservative group had at least one adverse event.

Bottom line: Missing data limit the ability to make strong conclusions, but this trial suggests that conservative management of primary spontaneous pneumothorax was noninferior to interventional management with lower risk of serious adverse events.

Citation: Brown SG et al. Conservative versus interventional treatment for spontaneous pneumothorax. N Engl J Med. 2020; 382:405-15.

Dr. Schmit is a hospitalist and associate professor of medicine at University of Texas Health, San Antonio.

Background: Abdominal paracentesis is a commonly performed procedure, and with appropriate training, hospitalists can deliver similar outcomes when compared to interventional radiologists.

To improve clinical outcomes, the authors recommended ultrasound guidance in performing paracentesis to reduce the risk of serious complications, to avoid attempting paracentesis with insufficient fluid, and to improve overall procedure success.

The authors advocated for several technique recommendations, including using the ultrasound to assess volume and location of intraperitoneal fluid, to identify the needle insertion site and confirm in multiple planes, to use color flow Doppler to identify abdominal wall vessels, to mark the insertion site immediately prior to the procedure, and to consider real-time ultrasound guidance.

When health care professionals are learning ultrasound-guided paracentesis, the authors recommended use of dedicated training sessions with simulation if available and that competency should be demonstrated before independently attempting the procedure.

Bottom line: These recommendations from SHM POCUS Task Force provides consensus guidelines on the use of ultrasound guidance when performing or learning abdominal paracentesis.

Citation: Cho J et al. Recommendations on the use of ultrasound guidance for adult abdominal paracentesis: A position statement of the Society of Hospital Medicine. 2019 Jan 2. doi: 10.12788/jhm.3095.

Dr. Schmit is an associate professor of medicine in the division of general and hospital medicine at UT Health San Antonio and a hospitalist at South Texas Veterans Health Care System, also in San Antonio.

Background: Abdominal paracentesis is a commonly performed procedure, and with appropriate training, hospitalists can deliver similar outcomes when compared to interventional radiologists.

To improve clinical outcomes, the authors recommended ultrasound guidance in performing paracentesis to reduce the risk of serious complications, to avoid attempting paracentesis with insufficient fluid, and to improve overall procedure success.

The authors advocated for several technique recommendations, including using the ultrasound to assess volume and location of intraperitoneal fluid, to identify the needle insertion site and confirm in multiple planes, to use color flow Doppler to identify abdominal wall vessels, to mark the insertion site immediately prior to the procedure, and to consider real-time ultrasound guidance.

When health care professionals are learning ultrasound-guided paracentesis, the authors recommended use of dedicated training sessions with simulation if available and that competency should be demonstrated before independently attempting the procedure.

Bottom line: These recommendations from SHM POCUS Task Force provides consensus guidelines on the use of ultrasound guidance when performing or learning abdominal paracentesis.

Citation: Cho J et al. Recommendations on the use of ultrasound guidance for adult abdominal paracentesis: A position statement of the Society of Hospital Medicine. 2019 Jan 2. doi: 10.12788/jhm.3095.

Dr. Schmit is an associate professor of medicine in the division of general and hospital medicine at UT Health San Antonio and a hospitalist at South Texas Veterans Health Care System, also in San Antonio.

Background: Abdominal paracentesis is a commonly performed procedure, and with appropriate training, hospitalists can deliver similar outcomes when compared to interventional radiologists.

To improve clinical outcomes, the authors recommended ultrasound guidance in performing paracentesis to reduce the risk of serious complications, to avoid attempting paracentesis with insufficient fluid, and to improve overall procedure success.

The authors advocated for several technique recommendations, including using the ultrasound to assess volume and location of intraperitoneal fluid, to identify the needle insertion site and confirm in multiple planes, to use color flow Doppler to identify abdominal wall vessels, to mark the insertion site immediately prior to the procedure, and to consider real-time ultrasound guidance.

When health care professionals are learning ultrasound-guided paracentesis, the authors recommended use of dedicated training sessions with simulation if available and that competency should be demonstrated before independently attempting the procedure.

Bottom line: These recommendations from SHM POCUS Task Force provides consensus guidelines on the use of ultrasound guidance when performing or learning abdominal paracentesis.

Citation: Cho J et al. Recommendations on the use of ultrasound guidance for adult abdominal paracentesis: A position statement of the Society of Hospital Medicine. 2019 Jan 2. doi: 10.12788/jhm.3095.

Dr. Schmit is an associate professor of medicine in the division of general and hospital medicine at UT Health San Antonio and a hospitalist at South Texas Veterans Health Care System, also in San Antonio.

Background: Anemia is common in hospitalized patients and is associated with short- and long-term morbidity and mortality. Current evidence shows that reduced red blood cell (RBC) use and more restrictive transfusion practices do not increase short-term mortality; however, few data exist on the long-term outcomes of anemia.

Given the retrospective design of this study, data must be interpreted with caution in determining a causal relationship. The authors also acknowledge that there may be unmeasured confounding variables not accounted for in the study results.

Bottom line: Despite higher rates of moderate anemia at discharge, there was not an associated rise in subsequent RBC transfusions, readmissions, or mortality in the 6 months after hospital discharge.

Citation: Roubinian NH et al. Long-term outcomes among patients discharged from the hospital with moderate anemia: A retrospective cohort study. Ann Intern Med. 2019 Jan 14. doi: 10.7326/M17-3253.

Dr. Schmit is an associate professor of medicine in the division of general and hospital medicine at UT Health San Antonio and a hospitalist at South Texas Veterans Health Care System, also in San Antonio.

Background: Anemia is common in hospitalized patients and is associated with short- and long-term morbidity and mortality. Current evidence shows that reduced red blood cell (RBC) use and more restrictive transfusion practices do not increase short-term mortality; however, few data exist on the long-term outcomes of anemia.

Given the retrospective design of this study, data must be interpreted with caution in determining a causal relationship. The authors also acknowledge that there may be unmeasured confounding variables not accounted for in the study results.

Bottom line: Despite higher rates of moderate anemia at discharge, there was not an associated rise in subsequent RBC transfusions, readmissions, or mortality in the 6 months after hospital discharge.

Citation: Roubinian NH et al. Long-term outcomes among patients discharged from the hospital with moderate anemia: A retrospective cohort study. Ann Intern Med. 2019 Jan 14. doi: 10.7326/M17-3253.

Dr. Schmit is an associate professor of medicine in the division of general and hospital medicine at UT Health San Antonio and a hospitalist at South Texas Veterans Health Care System, also in San Antonio.

Background: Anemia is common in hospitalized patients and is associated with short- and long-term morbidity and mortality. Current evidence shows that reduced red blood cell (RBC) use and more restrictive transfusion practices do not increase short-term mortality; however, few data exist on the long-term outcomes of anemia.

Given the retrospective design of this study, data must be interpreted with caution in determining a causal relationship. The authors also acknowledge that there may be unmeasured confounding variables not accounted for in the study results.

Bottom line: Despite higher rates of moderate anemia at discharge, there was not an associated rise in subsequent RBC transfusions, readmissions, or mortality in the 6 months after hospital discharge.

Citation: Roubinian NH et al. Long-term outcomes among patients discharged from the hospital with moderate anemia: A retrospective cohort study. Ann Intern Med. 2019 Jan 14. doi: 10.7326/M17-3253.

Dr. Schmit is an associate professor of medicine in the division of general and hospital medicine at UT Health San Antonio and a hospitalist at South Texas Veterans Health Care System, also in San Antonio.

Clinical question: What is the relationship between crowding in the ED and the number of interventions adopted by the ED to address this?

Background: ED crowding results in long waits, prolonged lengths of stay, and delays in providing treatments, which can result in adverse events. Numerous interventions, including bedside registration, ED observation units, fast track, bed czar, surgical schedule smoothing, and pooled nursing, have been implemented to reduce crowding.

Study design: Retrospective, cross-sectional analysis.

Setting: U.S. hospitals in the National Hospital Ambulatory Medical Care Survey (NHAMCS).

Synopsis: From 2007 to 2010, an average of 341 hospitals per year were analyzed from the NHAMCS, representing 139,502 patient encounters. This study evaluated the adoption of nine crowding interventions at the emergency department level (bedside registration, electronic dashboard, RFID tracking, etc.) and eight crowding interventions at the hospital level (bed czar, pooled nursing, full-capacity protocol, board patients in inpatient hallways, etc.).

Bedside registration, electronic dashboard, RFID tracking, bed census, pooled nursing, full-capacity protocol, and boarding patients in the hallway had the highest statistically significant increases in adoption over the study period.

The average number of interventions adopted increased to 6.6 from 5.2, and more-crowded EDs adopted a greater number of interventions than less-crowded EDs. However, in the most-crowded quartile of EDs, 19% did not use bedside registration, and 94% did not use surgical schedule smoothing.

Given that this study is a retrospective, cross-sectional study, it is difficult to determine causality.

Bottom line: More interventions are being adopted by EDs and hospitals to decrease ED crowding, but several of the busiest EDs and hospitals have room for improvement.

Citation: Warner LS, Pines JM, Chambers JG, Schuur JD. The most crowded US hospital emergency departments did not adopt effective interventions to improve flow, 2007–10. Health Aff. 2015;34(12):2151-2159.

Clinical question: What is the relationship between crowding in the ED and the number of interventions adopted by the ED to address this?

Background: ED crowding results in long waits, prolonged lengths of stay, and delays in providing treatments, which can result in adverse events. Numerous interventions, including bedside registration, ED observation units, fast track, bed czar, surgical schedule smoothing, and pooled nursing, have been implemented to reduce crowding.

Study design: Retrospective, cross-sectional analysis.

Setting: U.S. hospitals in the National Hospital Ambulatory Medical Care Survey (NHAMCS).

Synopsis: From 2007 to 2010, an average of 341 hospitals per year were analyzed from the NHAMCS, representing 139,502 patient encounters. This study evaluated the adoption of nine crowding interventions at the emergency department level (bedside registration, electronic dashboard, RFID tracking, etc.) and eight crowding interventions at the hospital level (bed czar, pooled nursing, full-capacity protocol, board patients in inpatient hallways, etc.).

Bedside registration, electronic dashboard, RFID tracking, bed census, pooled nursing, full-capacity protocol, and boarding patients in the hallway had the highest statistically significant increases in adoption over the study period.

The average number of interventions adopted increased to 6.6 from 5.2, and more-crowded EDs adopted a greater number of interventions than less-crowded EDs. However, in the most-crowded quartile of EDs, 19% did not use bedside registration, and 94% did not use surgical schedule smoothing.

Given that this study is a retrospective, cross-sectional study, it is difficult to determine causality.

Bottom line: More interventions are being adopted by EDs and hospitals to decrease ED crowding, but several of the busiest EDs and hospitals have room for improvement.

Citation: Warner LS, Pines JM, Chambers JG, Schuur JD. The most crowded US hospital emergency departments did not adopt effective interventions to improve flow, 2007–10. Health Aff. 2015;34(12):2151-2159.

Clinical question: What is the relationship between crowding in the ED and the number of interventions adopted by the ED to address this?

Background: ED crowding results in long waits, prolonged lengths of stay, and delays in providing treatments, which can result in adverse events. Numerous interventions, including bedside registration, ED observation units, fast track, bed czar, surgical schedule smoothing, and pooled nursing, have been implemented to reduce crowding.

Study design: Retrospective, cross-sectional analysis.

Setting: U.S. hospitals in the National Hospital Ambulatory Medical Care Survey (NHAMCS).

Synopsis: From 2007 to 2010, an average of 341 hospitals per year were analyzed from the NHAMCS, representing 139,502 patient encounters. This study evaluated the adoption of nine crowding interventions at the emergency department level (bedside registration, electronic dashboard, RFID tracking, etc.) and eight crowding interventions at the hospital level (bed czar, pooled nursing, full-capacity protocol, board patients in inpatient hallways, etc.).

Bedside registration, electronic dashboard, RFID tracking, bed census, pooled nursing, full-capacity protocol, and boarding patients in the hallway had the highest statistically significant increases in adoption over the study period.

The average number of interventions adopted increased to 6.6 from 5.2, and more-crowded EDs adopted a greater number of interventions than less-crowded EDs. However, in the most-crowded quartile of EDs, 19% did not use bedside registration, and 94% did not use surgical schedule smoothing.

Given that this study is a retrospective, cross-sectional study, it is difficult to determine causality.

Bottom line: More interventions are being adopted by EDs and hospitals to decrease ED crowding, but several of the busiest EDs and hospitals have room for improvement.

Citation: Warner LS, Pines JM, Chambers JG, Schuur JD. The most crowded US hospital emergency departments did not adopt effective interventions to improve flow, 2007–10. Health Aff. 2015;34(12):2151-2159.

Clinical question: What is the appropriate frequency of INR monitoring in the hospital and its relationship to the risk of over-anticoagulation and warfarin-related adverse events?

Background: Warfarin use is a common cause of adverse drug events in hospitalized patients due to narrow therapeutic windows, drug interactions, and variability of metabolism. Current guidelines, including those by the American College of Chest Physicians, do not provide recommendations on how often to monitor INR or adjust warfarin dosing in the hospital.

Study design: Retrospective cohort.

Setting: Hospitalized patients included in the Medicare Patient Safety Monitoring System.

Synopsis: The study included 14,217 adult patients ≥18 years of age from the Medicare Patient Safety Monitoring System admitted from 2009 to 2013 with pneumonia, acute cardiac disease (myocardial infarction or congestive heart failure), or surgery and taking warfarin. Of those, 1,055 (7.4%) developed a warfarin-associated adverse event (bleeding, drop in hematocrit ≥3, hematoma, death, intracranial bleeding, or cardiac arrest). Patients admitted for acute cardiac disease (acute myocardial infarction or heart failure) or surgery on warfarin for ≥3 days but not monitored for ≥2 days had more warfarin-associated adverse events (OR 1.48; 95% CI, 1.02–2.17), but this association was not true in pneumonia patients. Cardiac and pneumonia patients with ≥1 day without INR being measured had higher rates of INR ≥6.0 (OR 1.61; 95% CI, 1.07–2.41, and OR 1.92, 95% CI, 1.36–2.71, respectively). A single-day rise in INR ≥0.9 had a likelihood ratio of 4.2 in predicting subsequent INR ≥6.0.

Bottom line: Frequent monitoring of INR may decrease warfarin-associated adverse events in hospitalized patients.

Citation: Metersky ML, Eldridge N, Wang Y, et al. Predictors of warfarin-associated adverse events in hospitalized patients: opportunities to prevent harm. J Hosp Med. 2016;11(4):276-282.

Short Take

CDC Guidelines on Prescribing Opioids

New CDC guidelines for chronic pain management stress the importance of non-pharmacologic (physical therapy, etc.) and non-opioid therapy (NSAIDs, etc.), using opioid therapy only if the expected benefits outweigh the risks.

Citation: CDC. CDC guideline for prescribing opioids for chronic pain. Available at: http://www.cdc.gov/drugoverdose/prescribing/guideline.html. Published March 16, 2016. Accessed April 8, 2016.

Clinical question: What is the appropriate frequency of INR monitoring in the hospital and its relationship to the risk of over-anticoagulation and warfarin-related adverse events?

Background: Warfarin use is a common cause of adverse drug events in hospitalized patients due to narrow therapeutic windows, drug interactions, and variability of metabolism. Current guidelines, including those by the American College of Chest Physicians, do not provide recommendations on how often to monitor INR or adjust warfarin dosing in the hospital.

Study design: Retrospective cohort.

Setting: Hospitalized patients included in the Medicare Patient Safety Monitoring System.

Synopsis: The study included 14,217 adult patients ≥18 years of age from the Medicare Patient Safety Monitoring System admitted from 2009 to 2013 with pneumonia, acute cardiac disease (myocardial infarction or congestive heart failure), or surgery and taking warfarin. Of those, 1,055 (7.4%) developed a warfarin-associated adverse event (bleeding, drop in hematocrit ≥3, hematoma, death, intracranial bleeding, or cardiac arrest). Patients admitted for acute cardiac disease (acute myocardial infarction or heart failure) or surgery on warfarin for ≥3 days but not monitored for ≥2 days had more warfarin-associated adverse events (OR 1.48; 95% CI, 1.02–2.17), but this association was not true in pneumonia patients. Cardiac and pneumonia patients with ≥1 day without INR being measured had higher rates of INR ≥6.0 (OR 1.61; 95% CI, 1.07–2.41, and OR 1.92, 95% CI, 1.36–2.71, respectively). A single-day rise in INR ≥0.9 had a likelihood ratio of 4.2 in predicting subsequent INR ≥6.0.

Bottom line: Frequent monitoring of INR may decrease warfarin-associated adverse events in hospitalized patients.

Citation: Metersky ML, Eldridge N, Wang Y, et al. Predictors of warfarin-associated adverse events in hospitalized patients: opportunities to prevent harm. J Hosp Med. 2016;11(4):276-282.

Short Take

CDC Guidelines on Prescribing Opioids

New CDC guidelines for chronic pain management stress the importance of non-pharmacologic (physical therapy, etc.) and non-opioid therapy (NSAIDs, etc.), using opioid therapy only if the expected benefits outweigh the risks.

Citation: CDC. CDC guideline for prescribing opioids for chronic pain. Available at: http://www.cdc.gov/drugoverdose/prescribing/guideline.html. Published March 16, 2016. Accessed April 8, 2016.

Clinical question: What is the appropriate frequency of INR monitoring in the hospital and its relationship to the risk of over-anticoagulation and warfarin-related adverse events?

Background: Warfarin use is a common cause of adverse drug events in hospitalized patients due to narrow therapeutic windows, drug interactions, and variability of metabolism. Current guidelines, including those by the American College of Chest Physicians, do not provide recommendations on how often to monitor INR or adjust warfarin dosing in the hospital.

Study design: Retrospective cohort.

Setting: Hospitalized patients included in the Medicare Patient Safety Monitoring System.

Synopsis: The study included 14,217 adult patients ≥18 years of age from the Medicare Patient Safety Monitoring System admitted from 2009 to 2013 with pneumonia, acute cardiac disease (myocardial infarction or congestive heart failure), or surgery and taking warfarin. Of those, 1,055 (7.4%) developed a warfarin-associated adverse event (bleeding, drop in hematocrit ≥3, hematoma, death, intracranial bleeding, or cardiac arrest). Patients admitted for acute cardiac disease (acute myocardial infarction or heart failure) or surgery on warfarin for ≥3 days but not monitored for ≥2 days had more warfarin-associated adverse events (OR 1.48; 95% CI, 1.02–2.17), but this association was not true in pneumonia patients. Cardiac and pneumonia patients with ≥1 day without INR being measured had higher rates of INR ≥6.0 (OR 1.61; 95% CI, 1.07–2.41, and OR 1.92, 95% CI, 1.36–2.71, respectively). A single-day rise in INR ≥0.9 had a likelihood ratio of 4.2 in predicting subsequent INR ≥6.0.

Bottom line: Frequent monitoring of INR may decrease warfarin-associated adverse events in hospitalized patients.

Citation: Metersky ML, Eldridge N, Wang Y, et al. Predictors of warfarin-associated adverse events in hospitalized patients: opportunities to prevent harm. J Hosp Med. 2016;11(4):276-282.

Short Take

CDC Guidelines on Prescribing Opioids

New CDC guidelines for chronic pain management stress the importance of non-pharmacologic (physical therapy, etc.) and non-opioid therapy (NSAIDs, etc.), using opioid therapy only if the expected benefits outweigh the risks.

Citation: CDC. CDC guideline for prescribing opioids for chronic pain. Available at: http://www.cdc.gov/drugoverdose/prescribing/guideline.html. Published March 16, 2016. Accessed April 8, 2016.

Clinical question: What are the factors that promote education in delivering high-value, cost-conscious care?

Background: Healthcare costs are increasing, with most recent numbers showing U.S. expenditures on healthcare of more than $3 trillion, almost 18% of the gross domestic product. High-value care focuses on understanding the benefits, risks, and costs of care and promoting interventions that add value.

Study design: Systematic review.

Setting: Physicians, resident physicians, and medical students in North America, Asia, and Oceania.

Synopsis: Seventy-nine articles were included in the analysis, with 14 being RCTs. Most of the studies were conducted in North America (78.5%) and used a pre-post interventional design (58.2%). Practicing physicians (36.7%) made up the majority of participants in the study, with residents (15.2%) and medical students (6.3%) in smaller numbers. Analysis of the studies identified three factors for successful learning:

• effective transmission of knowledge about prices of services and general health economics, scientific evidence, and patient preferences;
• facilitation of reflective practice through feedback and/or stimulating reflection; and
• creation of a supportive environment.

Bottom line: The most-effective interventions in educating physicians, resident physicians, and medical students on high-value, cost-conscious care are effective transmission of knowledge, reflective practice, and supportive environment.

Citation: Stammen LA, Stalmeijer RE, Paternotte E, et al. Training physicians to provide high-value, cost-conscious care: a systematic review. JAMA. 2015;314(22):2384-2400.

Clinical question: What are the factors that promote education in delivering high-value, cost-conscious care?

Background: Healthcare costs are increasing, with most recent numbers showing U.S. expenditures on healthcare of more than $3 trillion, almost 18% of the gross domestic product. High-value care focuses on understanding the benefits, risks, and costs of care and promoting interventions that add value.

Study design: Systematic review.

Setting: Physicians, resident physicians, and medical students in North America, Asia, and Oceania.

Synopsis: Seventy-nine articles were included in the analysis, with 14 being RCTs. Most of the studies were conducted in North America (78.5%) and used a pre-post interventional design (58.2%). Practicing physicians (36.7%) made up the majority of participants in the study, with residents (15.2%) and medical students (6.3%) in smaller numbers. Analysis of the studies identified three factors for successful learning:

• effective transmission of knowledge about prices of services and general health economics, scientific evidence, and patient preferences;
• facilitation of reflective practice through feedback and/or stimulating reflection; and
• creation of a supportive environment.

Bottom line: The most-effective interventions in educating physicians, resident physicians, and medical students on high-value, cost-conscious care are effective transmission of knowledge, reflective practice, and supportive environment.

Citation: Stammen LA, Stalmeijer RE, Paternotte E, et al. Training physicians to provide high-value, cost-conscious care: a systematic review. JAMA. 2015;314(22):2384-2400.

Clinical question: What are the factors that promote education in delivering high-value, cost-conscious care?

Background: Healthcare costs are increasing, with most recent numbers showing U.S. expenditures on healthcare of more than $3 trillion, almost 18% of the gross domestic product. High-value care focuses on understanding the benefits, risks, and costs of care and promoting interventions that add value.

Study design: Systematic review.

Setting: Physicians, resident physicians, and medical students in North America, Asia, and Oceania.

Synopsis: Seventy-nine articles were included in the analysis, with 14 being RCTs. Most of the studies were conducted in North America (78.5%) and used a pre-post interventional design (58.2%). Practicing physicians (36.7%) made up the majority of participants in the study, with residents (15.2%) and medical students (6.3%) in smaller numbers. Analysis of the studies identified three factors for successful learning:

• effective transmission of knowledge about prices of services and general health economics, scientific evidence, and patient preferences;
• facilitation of reflective practice through feedback and/or stimulating reflection; and
• creation of a supportive environment.

Bottom line: The most-effective interventions in educating physicians, resident physicians, and medical students on high-value, cost-conscious care are effective transmission of knowledge, reflective practice, and supportive environment.

Citation: Stammen LA, Stalmeijer RE, Paternotte E, et al. Training physicians to provide high-value, cost-conscious care: a systematic review. JAMA. 2015;314(22):2384-2400.

Clinical question: Is there a clinical difference in rates of return of spontaneous circulation (ROSC) and survival to discharge in patients with in-hospital cardiac arrest (IHCA) depending on time of day and day of the week?

Background: Current U.S. data from the American Hospital Association’s “Get with the Guidelines-Resuscitation” (AHA GWTG-R) show hospital survival is lower at night and on the weekends. However, little data exist in the U.K. describing patients already hospitalized and the outcomes of in-hospital cardiac arrest with respect to time of day and day of the week.

Study design: Observational cohort study.

Setting: One hundred forty-six hospitals in the United Kingdom.

Synopsis: Study investigators included 27,700 patients ≥16 years of age receiving chest compressions and/or defibrillation from the U.K. National Cardiac Arrest Audit (NCAA) from April 2011 to September 2013. When compared to weekday daytime, the risk-adjusted rates of ROSC were worse for weekend daytime (odds ratio [OR] ROSC >20 min. 0.88; 95% CI, 0.81–0.95) and nighttime (OR ROSC >20 min. 0.72; 95% CI, 0.68–0.76). Hospital survival had similar trends, with OR for the weekend daytime of 0.72 (95% CI, 0.64–0.80) and OR for nighttime 0.58 (95% CI, 0.54–0.63; P value for all was <0.001).

IHCAs were equally likely to occur during the day and night, and the patients were broadly similar, thus suggesting differences in outcomes were secondary to care differences. However, unmeasured patient characteristics may have affected the outcomes. Given that the study was observational, it is difficult to attribute causality, but results are similar to the large, multicenter study published by the AHA GWTG-R registry.

Bottom line: IHCAs that occur during the night or on weekends have increased odds of worse outcomes.

Citation: Robinson EJ, Smith GB, Power GS, et al. Risk-adjusted survival for adults following in-hospital cardiac arrest by day of week and time of day: observational cohort study [published online ahead of print December 11, 2015]. BMJ Qual Saf. doi:10.1136/bmjqs-2015-004223.

Short Take

USPSTF Recommends Statins for More Americans

The U.S. Preventive Services Task Force recommends a low- to moderate-dose statin for adults ages 40–75 with no history of cardiovascular disease and a calculated 10-year cardiovascular disease event risk of ≥10%.

Citation: U.S. Preventive Services Task Force. Draft recommendation statement: statin use for the primary prevention of cardiovascular disease in adults: preventive medication. Available at:

http://www.uspreventiveservicestaskforce.org/Page/Document/draft-recommendation-statement175/statin-use-in-adults-preventive-medication1. Published December 2015. Accessed April 1, 2016.

Clinical question: Is there a clinical difference in rates of return of spontaneous circulation (ROSC) and survival to discharge in patients with in-hospital cardiac arrest (IHCA) depending on time of day and day of the week?

Background: Current U.S. data from the American Hospital Association’s “Get with the Guidelines-Resuscitation” (AHA GWTG-R) show hospital survival is lower at night and on the weekends. However, little data exist in the U.K. describing patients already hospitalized and the outcomes of in-hospital cardiac arrest with respect to time of day and day of the week.

Study design: Observational cohort study.

Setting: One hundred forty-six hospitals in the United Kingdom.

Synopsis: Study investigators included 27,700 patients ≥16 years of age receiving chest compressions and/or defibrillation from the U.K. National Cardiac Arrest Audit (NCAA) from April 2011 to September 2013. When compared to weekday daytime, the risk-adjusted rates of ROSC were worse for weekend daytime (odds ratio [OR] ROSC >20 min. 0.88; 95% CI, 0.81–0.95) and nighttime (OR ROSC >20 min. 0.72; 95% CI, 0.68–0.76). Hospital survival had similar trends, with OR for the weekend daytime of 0.72 (95% CI, 0.64–0.80) and OR for nighttime 0.58 (95% CI, 0.54–0.63; P value for all was <0.001).

IHCAs were equally likely to occur during the day and night, and the patients were broadly similar, thus suggesting differences in outcomes were secondary to care differences. However, unmeasured patient characteristics may have affected the outcomes. Given that the study was observational, it is difficult to attribute causality, but results are similar to the large, multicenter study published by the AHA GWTG-R registry.

Bottom line: IHCAs that occur during the night or on weekends have increased odds of worse outcomes.

Citation: Robinson EJ, Smith GB, Power GS, et al. Risk-adjusted survival for adults following in-hospital cardiac arrest by day of week and time of day: observational cohort study [published online ahead of print December 11, 2015]. BMJ Qual Saf. doi:10.1136/bmjqs-2015-004223.

Short Take

USPSTF Recommends Statins for More Americans

The U.S. Preventive Services Task Force recommends a low- to moderate-dose statin for adults ages 40–75 with no history of cardiovascular disease and a calculated 10-year cardiovascular disease event risk of ≥10%.

Citation: U.S. Preventive Services Task Force. Draft recommendation statement: statin use for the primary prevention of cardiovascular disease in adults: preventive medication. Available at:

http://www.uspreventiveservicestaskforce.org/Page/Document/draft-recommendation-statement175/statin-use-in-adults-preventive-medication1. Published December 2015. Accessed April 1, 2016.

Clinical question: Is there a clinical difference in rates of return of spontaneous circulation (ROSC) and survival to discharge in patients with in-hospital cardiac arrest (IHCA) depending on time of day and day of the week?

Background: Current U.S. data from the American Hospital Association’s “Get with the Guidelines-Resuscitation” (AHA GWTG-R) show hospital survival is lower at night and on the weekends. However, little data exist in the U.K. describing patients already hospitalized and the outcomes of in-hospital cardiac arrest with respect to time of day and day of the week.

Study design: Observational cohort study.

Setting: One hundred forty-six hospitals in the United Kingdom.

Synopsis: Study investigators included 27,700 patients ≥16 years of age receiving chest compressions and/or defibrillation from the U.K. National Cardiac Arrest Audit (NCAA) from April 2011 to September 2013. When compared to weekday daytime, the risk-adjusted rates of ROSC were worse for weekend daytime (odds ratio [OR] ROSC >20 min. 0.88; 95% CI, 0.81–0.95) and nighttime (OR ROSC >20 min. 0.72; 95% CI, 0.68–0.76). Hospital survival had similar trends, with OR for the weekend daytime of 0.72 (95% CI, 0.64–0.80) and OR for nighttime 0.58 (95% CI, 0.54–0.63; P value for all was <0.001).

IHCAs were equally likely to occur during the day and night, and the patients were broadly similar, thus suggesting differences in outcomes were secondary to care differences. However, unmeasured patient characteristics may have affected the outcomes. Given that the study was observational, it is difficult to attribute causality, but results are similar to the large, multicenter study published by the AHA GWTG-R registry.

Bottom line: IHCAs that occur during the night or on weekends have increased odds of worse outcomes.

Citation: Robinson EJ, Smith GB, Power GS, et al. Risk-adjusted survival for adults following in-hospital cardiac arrest by day of week and time of day: observational cohort study [published online ahead of print December 11, 2015]. BMJ Qual Saf. doi:10.1136/bmjqs-2015-004223.

Short Take

USPSTF Recommends Statins for More Americans

The U.S. Preventive Services Task Force recommends a low- to moderate-dose statin for adults ages 40–75 with no history of cardiovascular disease and a calculated 10-year cardiovascular disease event risk of ≥10%.

Citation: U.S. Preventive Services Task Force. Draft recommendation statement: statin use for the primary prevention of cardiovascular disease in adults: preventive medication. Available at:

http://www.uspreventiveservicestaskforce.org/Page/Document/draft-recommendation-statement175/statin-use-in-adults-preventive-medication1. Published December 2015. Accessed April 1, 2016.

Clinical question: What is the harm associated with long-term beta-blocker therapy in patients with uncomplicated hypertension undergoing non-cardiac surgery?

Background: Given the recent concerns over the validity of prior studies, there is uncertainty about which patients benefit most from perioperative beta-blockade. Current guidelines suggest continuing beta-blockers in the perioperative period. More data are needed to delineate which patients maximally benefit from perioperative beta-blockade.

Study design: Association study.

Setting: Danish nationwide cohort of patients.

Synopsis: Study investigators included 55,320 uncomplicated hypertension (no cardiovascular, renal, or liver disease) patients >19 years of age on ≥2 antihypertensive drugs undergoing non-cardiac surgery. In the 14,664 patients who received beta-blockers, the rates of 30-day major adverse cardiovascular events (MACE; cardiovascular death, nonfatal ischemic stroke, and nonfatal myocardial infarction) and 30-day all-cause mortality were 1.32% and 1.93%, respectively. However, in the 40,676 patients who did not receive beta-blockers, 30-day MACEs and 30-day all-cause mortality rates were 0.84% and 1.32%, respectively (P<0.001). When looking at the individual MACEs, cardiovascular death was the only statistically significant event with higher incidence (0.9% versus 0.45%, P<0.001).

Combination therapy with beta-blocker and RAS inhibitor, calcium channel blockers, or thiazide was associated with statistically significant higher risks of MACEs and all-cause mortality when compared to the combination of RAS inhibitor plus thiazide. Men >70 years of age or undergoing urgent surgery had the highest risk of harm. This study was not a randomized control trial, so caution must be used when attributing causality to beta-blockers, MACEs, and all-cause mortality.

Bottom line: Antihypertensive regimens containing beta-blockers may increase risk of perioperative MACEs and all-cause mortality in patients with uncomplicated hypertension.

Citation: Jorgensen ME, Hlatky MA, Kober L, et al. β-blocker-associated risks in patients with uncomplicated hypertension undergoing noncardiac surgery. JAMA Intern Med. 2015;175(12):1923-1931.

Clinical question: What is the harm associated with long-term beta-blocker therapy in patients with uncomplicated hypertension undergoing non-cardiac surgery?

Background: Given the recent concerns over the validity of prior studies, there is uncertainty about which patients benefit most from perioperative beta-blockade. Current guidelines suggest continuing beta-blockers in the perioperative period. More data are needed to delineate which patients maximally benefit from perioperative beta-blockade.

Study design: Association study.

Setting: Danish nationwide cohort of patients.

Synopsis: Study investigators included 55,320 uncomplicated hypertension (no cardiovascular, renal, or liver disease) patients >19 years of age on ≥2 antihypertensive drugs undergoing non-cardiac surgery. In the 14,664 patients who received beta-blockers, the rates of 30-day major adverse cardiovascular events (MACE; cardiovascular death, nonfatal ischemic stroke, and nonfatal myocardial infarction) and 30-day all-cause mortality were 1.32% and 1.93%, respectively. However, in the 40,676 patients who did not receive beta-blockers, 30-day MACEs and 30-day all-cause mortality rates were 0.84% and 1.32%, respectively (P<0.001). When looking at the individual MACEs, cardiovascular death was the only statistically significant event with higher incidence (0.9% versus 0.45%, P<0.001).

Combination therapy with beta-blocker and RAS inhibitor, calcium channel blockers, or thiazide was associated with statistically significant higher risks of MACEs and all-cause mortality when compared to the combination of RAS inhibitor plus thiazide. Men >70 years of age or undergoing urgent surgery had the highest risk of harm. This study was not a randomized control trial, so caution must be used when attributing causality to beta-blockers, MACEs, and all-cause mortality.

Bottom line: Antihypertensive regimens containing beta-blockers may increase risk of perioperative MACEs and all-cause mortality in patients with uncomplicated hypertension.

Citation: Jorgensen ME, Hlatky MA, Kober L, et al. β-blocker-associated risks in patients with uncomplicated hypertension undergoing noncardiac surgery. JAMA Intern Med. 2015;175(12):1923-1931.

Clinical question: What is the harm associated with long-term beta-blocker therapy in patients with uncomplicated hypertension undergoing non-cardiac surgery?

Background: Given the recent concerns over the validity of prior studies, there is uncertainty about which patients benefit most from perioperative beta-blockade. Current guidelines suggest continuing beta-blockers in the perioperative period. More data are needed to delineate which patients maximally benefit from perioperative beta-blockade.

Study design: Association study.

Setting: Danish nationwide cohort of patients.

Synopsis: Study investigators included 55,320 uncomplicated hypertension (no cardiovascular, renal, or liver disease) patients >19 years of age on ≥2 antihypertensive drugs undergoing non-cardiac surgery. In the 14,664 patients who received beta-blockers, the rates of 30-day major adverse cardiovascular events (MACE; cardiovascular death, nonfatal ischemic stroke, and nonfatal myocardial infarction) and 30-day all-cause mortality were 1.32% and 1.93%, respectively. However, in the 40,676 patients who did not receive beta-blockers, 30-day MACEs and 30-day all-cause mortality rates were 0.84% and 1.32%, respectively (P<0.001). When looking at the individual MACEs, cardiovascular death was the only statistically significant event with higher incidence (0.9% versus 0.45%, P<0.001).

Combination therapy with beta-blocker and RAS inhibitor, calcium channel blockers, or thiazide was associated with statistically significant higher risks of MACEs and all-cause mortality when compared to the combination of RAS inhibitor plus thiazide. Men >70 years of age or undergoing urgent surgery had the highest risk of harm. This study was not a randomized control trial, so caution must be used when attributing causality to beta-blockers, MACEs, and all-cause mortality.

Bottom line: Antihypertensive regimens containing beta-blockers may increase risk of perioperative MACEs and all-cause mortality in patients with uncomplicated hypertension.

Citation: Jorgensen ME, Hlatky MA, Kober L, et al. β-blocker-associated risks in patients with uncomplicated hypertension undergoing noncardiac surgery. JAMA Intern Med. 2015;175(12):1923-1931.

Clinical question: What is the prevalence of depression or depressive symptoms in resident physicians?

Background: Depression in resident physicians can lead to poor-quality medical care, increased errors, and long-term morbidity. Prevalence of depression or depressive symptoms has varied in prior studies, and more data are needed to better understand the true prevalence.

Study design: Systematic review and meta-analysis.

Setting: Surgical and nonsurgical residency programs in North America, Asia, Europe, South America, and Africa

Synopsis: Thirty-one cross-sectional studies (9,447 individuals) and 23 longitudinal studies (8,113 individuals) from January 1963 to September 2015 were included in this analysis, with the majority using self-reporting to identify residents with depression or depressive symptoms. Overall prevalence of depression or depressive symptoms was 28.8%, with a range of 20.9% to 43.2%, depending on the screening tool (95% CI, 25.3%–32.5%; P<0.001). There was an increased prevalence in depression or depressive symptoms as the calendar year progressed (slope=0.5% per calendar year increase; 95% CI, 0.03%–0.09%), with no difference in prevalence rates between surgical versus nonsurgical residents, U.S. versus elsewhere, cross-sectional versus longitudinal, or interns versus upper-level residents.

Because studies were heterogeneous with respect to the screening tools and resident population, the prevalence of depression or depressive symptoms cannot be precisely determined.

Bottom line: Prevalence of depression or depressive symptoms ranged from 20.9% to 43.2%, with pooled prevalence of 28.8%, and increased with time.

Citation: Mata DA, Ramos MA, Bansal N, et al. Prevalence of depression and depressive symptoms among resident physicians: a systematic review and met-analysis. JAMA. 2015;314(22):2373-2383.

Clinical question: What is the prevalence of depression or depressive symptoms in resident physicians?

Background: Depression in resident physicians can lead to poor-quality medical care, increased errors, and long-term morbidity. Prevalence of depression or depressive symptoms has varied in prior studies, and more data are needed to better understand the true prevalence.

Study design: Systematic review and meta-analysis.

Setting: Surgical and nonsurgical residency programs in North America, Asia, Europe, South America, and Africa

Synopsis: Thirty-one cross-sectional studies (9,447 individuals) and 23 longitudinal studies (8,113 individuals) from January 1963 to September 2015 were included in this analysis, with the majority using self-reporting to identify residents with depression or depressive symptoms. Overall prevalence of depression or depressive symptoms was 28.8%, with a range of 20.9% to 43.2%, depending on the screening tool (95% CI, 25.3%–32.5%; P<0.001). There was an increased prevalence in depression or depressive symptoms as the calendar year progressed (slope=0.5% per calendar year increase; 95% CI, 0.03%–0.09%), with no difference in prevalence rates between surgical versus nonsurgical residents, U.S. versus elsewhere, cross-sectional versus longitudinal, or interns versus upper-level residents.

Because studies were heterogeneous with respect to the screening tools and resident population, the prevalence of depression or depressive symptoms cannot be precisely determined.

Bottom line: Prevalence of depression or depressive symptoms ranged from 20.9% to 43.2%, with pooled prevalence of 28.8%, and increased with time.

Citation: Mata DA, Ramos MA, Bansal N, et al. Prevalence of depression and depressive symptoms among resident physicians: a systematic review and met-analysis. JAMA. 2015;314(22):2373-2383.

Clinical question: What is the prevalence of depression or depressive symptoms in resident physicians?

Background: Depression in resident physicians can lead to poor-quality medical care, increased errors, and long-term morbidity. Prevalence of depression or depressive symptoms has varied in prior studies, and more data are needed to better understand the true prevalence.

Study design: Systematic review and meta-analysis.

Setting: Surgical and nonsurgical residency programs in North America, Asia, Europe, South America, and Africa

Synopsis: Thirty-one cross-sectional studies (9,447 individuals) and 23 longitudinal studies (8,113 individuals) from January 1963 to September 2015 were included in this analysis, with the majority using self-reporting to identify residents with depression or depressive symptoms. Overall prevalence of depression or depressive symptoms was 28.8%, with a range of 20.9% to 43.2%, depending on the screening tool (95% CI, 25.3%–32.5%; P<0.001). There was an increased prevalence in depression or depressive symptoms as the calendar year progressed (slope=0.5% per calendar year increase; 95% CI, 0.03%–0.09%), with no difference in prevalence rates between surgical versus nonsurgical residents, U.S. versus elsewhere, cross-sectional versus longitudinal, or interns versus upper-level residents.

Because studies were heterogeneous with respect to the screening tools and resident population, the prevalence of depression or depressive symptoms cannot be precisely determined.

Bottom line: Prevalence of depression or depressive symptoms ranged from 20.9% to 43.2%, with pooled prevalence of 28.8%, and increased with time.

Citation: Mata DA, Ramos MA, Bansal N, et al. Prevalence of depression and depressive symptoms among resident physicians: a systematic review and met-analysis. JAMA. 2015;314(22):2373-2383.