Reductions in residency training hours do not significantly affect quality of patient care, including inpatient mortality, according to a study published online today in the BMJ.
With the data showing similar patient outcomes, one expert said the focus now needs to shift from discussing the total number of hours that residents spend in the hospital to how those hours are spent
In 2003, the Accreditation Council for Graduate Medical Education (ACGME) instituted a requirement that work weeks for residents not exceed 80 hours and that shifts not be longer than 30 hours. The changes came in response to a widely publicized death in a New York teaching hospital and to growing concerns about the safety of patients cared for by fatigued residents. Subsequent reforms in 2011 capped shift lengths at 16 hours for interns and 28 hours for trainees. However, in 2017, the ACGME allowed interns in some programs to work longer shifts, and made other changes as well.
"The reduction in resident work hours sparked debate as to whether working fewer hours during residency training would lead to physicians entering independent practice who were inadequately prepared," author Anupam B. Jena, MD, PhD, Harvard Medical School, Boston, Massachusetts, told Medscape Medical News via email.
To answer the question, Jena and colleagues compared patient outcomes for physicians trained in internal medicine before and after the residency work hour reforms occurred. They found that the reduction in work hours was not linked to hospital mortality, readmissions, and costs of care.
"These findings should certainly inform the debate, though may not end it," Jena said.
"It's important to recognize that hospital care is different than what it was 20 to 30 years ago, in a way that reduces the role of a single physician in driving patient outcomes. It's possible that the trainee of the future may adequately be prepared for independent practice with less than 80 hours a week during residency," he explained.
When asked for comment, Sanjay Desai, MD, vice chair of education and program director for internal medicine at Johns Hopkins University, Baltimore, Maryland, said that the study is important because it evaluated practicing physicians after they had completed training. Previous studies, by contrast, evaluated physicians while in training.
For example, results from the Individualized Comparative Effectiveness of Models Optimizing Patient Safety and Resident Education (iCOMPARE) showed that with restricted hours, there was no change in patient outcomes and no drop in educational outcomes for trainees.
However, few studies have evaluated the long-term effects of a reduction in training hours. Only one other study, which was conducted in Florida, has done so, and results may not be generalizable to other states, said Desai, who has served on the ACGME committee that set residency work hour limitations and whose research is focused in this area.
The study by Jena and colleagues "is a unique contribution because it actually assesses the performance of clinicians after they finish training. It affirms for me that the discussion around the number of hours spent in the hospital during training is not an important discussion anymore," Desai said.
Instead of talking about the absolute number of hours spent working, the focus should be on how those hours are spent, he explained.
One of the most important variables that needs to be assessed is physician burnout, he stressed. Other important variables include professionalization, physician attitudes, communication skills, development of specific clinical skills, and the impact of work hours on physician health.
"This study confirms, in my view, that we need to shift away from the number of hours to differences in how we spend those hours and how that associates with these other outcomes, including physician well-being. That type of research is desperately needed for the country," he said.
Nearly 500,000 Admissions Analyzed
For the study, Jena and colleagues analyzed 485,685 admissions for patients hospitalized under Medicare part B between January 2000 and December 2012.
They compared outcomes among patients cared for by physicians who were in their first year of independent practice and who finished residency before (2000–2006) and after (2007–2012) the ACGME reforms with outcomes of patients cared for by senior internists who were in their 10th year of independent practice during the same years. A differences-in-differences analysis showed no significant difference between training periods for any of the outcomes examined.
Among first-year internists, the 30-day mortality rate was 10.6% for those who completed training before reforms and 9.6% for those who completed training after reforms. Among senior physicians, the rates were 11.2% and 10.6% for the same periods.
Likewise, 30-day readmission rates among patients cared for by pre-reform and post-reform first-year internists were the same, at 20.4%. For senior internists, the rates were 20.1% and 20.5% during the same years.
Inpatient spending was also similar. Among first-year internists, spending was $1161 and $1267 per hospital admission for those in the pre-reform and post-reform groups, respectively. Inpatient spending among senior internists was $1331 and $1599 for the same periods.
The study has several potential limitations, including its observational design and the inclusion only of internists. The authors caution that the results may not be generalizable to other types of residents, particularly surgical residents, for whom exposure to a certain volume of procedures may make a difference.
Also, use of the 30-day mortality endpoint may have obscured differences between study groups. However, policy makers are likely most interested in this endpoint and would change work hour rules if any differences were found, Desai explained.
The study was sponsored by the National Institutes of Health. One or more authors have received consulting fees from one or more of the following: Pfizer, Hill Rom Services, Bristol-Myers Squibb, Novartis, Amgen, Eli Lilly, Vertex Pharmaceuticals, AstraZeneca, Celgene, Tesaro, Sanofi Aventis, Biogen, Precision Health Economics, Analysis Group, and Precision Health Economics. One author is an employee of Devoted Health. Desai has disclosed no relevant financial relationships.
BMJ. Published online July 10, 2019.
With the data showing similar patient outcomes, one expert said the focus now needs to shift from discussing the total number of hours that residents spend in the hospital to how those hours are spent
In 2003, the Accreditation Council for Graduate Medical Education (ACGME) instituted a requirement that work weeks for residents not exceed 80 hours and that shifts not be longer than 30 hours. The changes came in response to a widely publicized death in a New York teaching hospital and to growing concerns about the safety of patients cared for by fatigued residents. Subsequent reforms in 2011 capped shift lengths at 16 hours for interns and 28 hours for trainees. However, in 2017, the ACGME allowed interns in some programs to work longer shifts, and made other changes as well.
"The reduction in resident work hours sparked debate as to whether working fewer hours during residency training would lead to physicians entering independent practice who were inadequately prepared," author Anupam B. Jena, MD, PhD, Harvard Medical School, Boston, Massachusetts, told Medscape Medical News via email.
To answer the question, Jena and colleagues compared patient outcomes for physicians trained in internal medicine before and after the residency work hour reforms occurred. They found that the reduction in work hours was not linked to hospital mortality, readmissions, and costs of care.
"These findings should certainly inform the debate, though may not end it," Jena said.
"It's important to recognize that hospital care is different than what it was 20 to 30 years ago, in a way that reduces the role of a single physician in driving patient outcomes. It's possible that the trainee of the future may adequately be prepared for independent practice with less than 80 hours a week during residency," he explained.
When asked for comment, Sanjay Desai, MD, vice chair of education and program director for internal medicine at Johns Hopkins University, Baltimore, Maryland, said that the study is important because it evaluated practicing physicians after they had completed training. Previous studies, by contrast, evaluated physicians while in training.
For example, results from the Individualized Comparative Effectiveness of Models Optimizing Patient Safety and Resident Education (iCOMPARE) showed that with restricted hours, there was no change in patient outcomes and no drop in educational outcomes for trainees.
However, few studies have evaluated the long-term effects of a reduction in training hours. Only one other study, which was conducted in Florida, has done so, and results may not be generalizable to other states, said Desai, who has served on the ACGME committee that set residency work hour limitations and whose research is focused in this area.
The study by Jena and colleagues "is a unique contribution because it actually assesses the performance of clinicians after they finish training. It affirms for me that the discussion around the number of hours spent in the hospital during training is not an important discussion anymore," Desai said.
Instead of talking about the absolute number of hours spent working, the focus should be on how those hours are spent, he explained.
One of the most important variables that needs to be assessed is physician burnout, he stressed. Other important variables include professionalization, physician attitudes, communication skills, development of specific clinical skills, and the impact of work hours on physician health.
"This study confirms, in my view, that we need to shift away from the number of hours to differences in how we spend those hours and how that associates with these other outcomes, including physician well-being. That type of research is desperately needed for the country," he said.
Nearly 500,000 Admissions Analyzed
For the study, Jena and colleagues analyzed 485,685 admissions for patients hospitalized under Medicare part B between January 2000 and December 2012.
They compared outcomes among patients cared for by physicians who were in their first year of independent practice and who finished residency before (2000–2006) and after (2007–2012) the ACGME reforms with outcomes of patients cared for by senior internists who were in their 10th year of independent practice during the same years. A differences-in-differences analysis showed no significant difference between training periods for any of the outcomes examined.
Among first-year internists, the 30-day mortality rate was 10.6% for those who completed training before reforms and 9.6% for those who completed training after reforms. Among senior physicians, the rates were 11.2% and 10.6% for the same periods.
Likewise, 30-day readmission rates among patients cared for by pre-reform and post-reform first-year internists were the same, at 20.4%. For senior internists, the rates were 20.1% and 20.5% during the same years.
Inpatient spending was also similar. Among first-year internists, spending was $1161 and $1267 per hospital admission for those in the pre-reform and post-reform groups, respectively. Inpatient spending among senior internists was $1331 and $1599 for the same periods.
The study has several potential limitations, including its observational design and the inclusion only of internists. The authors caution that the results may not be generalizable to other types of residents, particularly surgical residents, for whom exposure to a certain volume of procedures may make a difference.
Also, use of the 30-day mortality endpoint may have obscured differences between study groups. However, policy makers are likely most interested in this endpoint and would change work hour rules if any differences were found, Desai explained.
The study was sponsored by the National Institutes of Health. One or more authors have received consulting fees from one or more of the following: Pfizer, Hill Rom Services, Bristol-Myers Squibb, Novartis, Amgen, Eli Lilly, Vertex Pharmaceuticals, AstraZeneca, Celgene, Tesaro, Sanofi Aventis, Biogen, Precision Health Economics, Analysis Group, and Precision Health Economics. One author is an employee of Devoted Health. Desai has disclosed no relevant financial relationships.
BMJ. Published online July 10, 2019.