Impact of duty hours on competency‐related knowledge acquisition among community hospital residents

Abstract Background The effect of duty hour (DH) restrictions on postgraduate residents' acquisition of clinical competencies is unclear. We evaluated the relationship between DHs and competency‐related knowledge acquisition using the General Medicine In‐training Examination (GM‐ITE). Methods We conducted a multicenter, cross‐sectional study of community hospital residents among 2019 GM‐ITE examinees. Self‐reported average DHs per week were classified into five DH categories and the competency domains were classified into four areas: symptomatology and clinical reasoning (CR), physical examination and clinical procedure (PP), medical interview and professionalism (MP), and disease knowledge (DK). The association between these scores and DHs was examined using random‐intercept linear models with and without adjustment for confounding factors. Results We included 4753 participants in the analyses. Of these, 31% were women, and 49.1% were in the postgraduate year (PGY) 2. Mean CR and MP scores were lower among residents in Category 1 (<50 h) than in residents in Category 3 (≥60 and <70 h; reference group). Mean DK scores were lower among residents in Categories 1 and 2 (≥50 and <60 h) than in the reference group. PGY‐2 residents in Categories 1 and 2 had lower CR scores than those in Category 3; however, PGY‐1 residents in Category 5 showed higher scores. Conclusions The relationship between DHs and each competency area is not strictly linear. The acquisition of knowledge of physical examination and clinical procedures skills in particular may not be related to DHs.


| INTRODUC TI ON
Since 2018, workstyle reforms in Japan have affected all workers, including physicians and postgraduate residents. 1  There has been minimal research on the association between the number of DHs and resident competencies in Japan as well as globally. One report analyzed the association between resident DHs and in-training examination scores in Japan. 2 The study demonstrated that fewer than 60-65 DHs per week were independently associated with lower performance; however, exceeding 65 DHs per week did not improve performance. 2 Another research further reported that Japanese residents with fewer than 60 DHs per week tended to spend less time in self-study than did those with more than 65 DHs per week. 3 Surveys about the impact of DH restrictions on residents in the United States found that, while resident health and burnout improved, there were negative effects on resident education and patient care. [4][5][6] These studies assessed the overall impact on resident education and suggested that DH restrictions had negative impacts on education. A systematic review researching the effect of DH restrictions on postgraduate medical education found that the effect on education was assessed mostly by measuring access to educational opportunities, total in-training examination scores, and caseloads. 7 In this context, we found no previous studies that directly examined the impact on DHs and residents' competency in various areas. Investigating the relationship between DHs and each resident's competency could prompt program directors, training program developers, and policymakers to consider ways to mitigate the negative educational effects of DH restrictions.
In this study, we examined the relationship between DHs and scores in each competency area of the General Medicine In-training Examination (GM-ITE), an in-training examination in Japan among residents affiliated with community hospitals in Japan.

| Study populations
In Japan, there is a 2-year postgraduate clinical training system.
We conducted a multicenter, cross-sectional study of Japanese residents in postgraduate year 1 (PGY-1) and postgraduate year 2 (PGY-2) of their training. Before conducting the survey, we obtained written informed consent for study participation from all participants. The research consent document stated that the questionnaire results would be anonymized. An appropriate research ethics committee approved the study. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.
In Japan, teaching hospitals are divided into three categories: independent hospitals, administrative hospitals, and cooperative hospitals. 8,9 Independent hospitals provide training programs almost exclusively on their own, while administrative hospitals operate programs in collaboration with cooperative hospitals. Many university hospitals are administrative hospitals, and many of them have programs, so-called tasukigake programs, where training at one cooperative hospital lasts 1 year. Thus, the work styles of university hospital residents are more diverse than those of community hospital residents. Because this study used the average DHs of the entire training program for analysis, university hospital residents, whose DHs would be difficult to interpret, were excluded from the study, and only community hospital residents were included in our study.
As part of a training environment questionnaire completed immediately after the examination, participants were asked about average DHs during their residency program. The questionnaire also included the number of emergency department (ED) duties per month, the mean number of assigned inpatients, and self-study time per day. reference group). Mean DK scores were lower among residents in Categories 1 and 2 (≥50 and <60 h) than in the reference group. PGY-2 residents in Categories 1 and 2 had lower CR scores than those in Category 3; however, PGY-1 residents in Category 5 showed higher scores.

Conclusions:
The relationship between DHs and each competency area is not strictly linear. The acquisition of knowledge of physical examination and clinical procedures skills in particular may not be related to DHs.

K E Y W O R D S
clinical competency, community hospital, duty hour restriction, General Medicine In-training Examination, Japan, postgraduate resident

| Measurements
The primary independent variable in this study was the average self-reported DHs per week. We defined DHs as the total hours of weekday work duty, night ED duty (including restrained time when not providing medical care), and weekend work duty. DHs were answered in eight categories, which were reorganized into five cat- Although both CR and DK areas contain case-based questions on diagnosis and management, CR mainly focuses on the diagnosis, and DK focuses on management. In the PP area, we applied audio and video questions. We excluded one question in the PP area after the examination as an inappropriate question. Participants completed the GM-ITE from January 21 to 28, 2020.

| Statistical analyses
We examined the association between resident DHs and each competency area score (one point per question) using random-intercept linear models, accounting for hospital variability as normal random intercepts. Category 3 was set as the DH reference for the analyses, considering that 60 h per week was the basic upper limit of DHs for all doctors. 1 We excluded residents who did not provide information about DHs from the analyses. We repeated the analyses, stratified by PGY. We adjusted for the following variables that could potentially be associated with the total GM-ITE score: sex, PGY, ED duties, number of assigned inpatients, and self-study time. 2,12 All analyses were performed using SAS version 9.4 (SAS Inst.).

| RE SULTS
The GM-ITE survey included 6869 postgraduate residents from 539 residency programs. The survey response rate was 89.7% (6164/6869). We excluded 571 residents who did not provide DH information, and also excluded another 840 university and university-affiliated hospital residents. Finally, the analysis included 4753 participants. Among them, 31% were women and 50.9% were PGY-1. The mean (± standard deviation) of total GM-ITE scores of all the participants was 29.7 ± 5.3. The scores in the various competency areas were as follows: CR, 9.9 ± 2.3; PP, 9.2 ± 2.3; MP, 2.9 ± 1.2; and DK, 8.2 ± 2.2. Table 1 summarizes resident characteristics categorized by DHs, and Table S1 in Appendix S1 lists them stratified by PGY.   Table S3 in Appendix S2 present the adjusted results of the association between resident DHs and each competency area score stratified by PGY. PGY-2 residents in Categories 1 and 2 had lower CR scores than those in Category 3; however, PGY-1 residents in Category 5 had higher CR scores than those in Category 3. DK scores were lower for PGY-1 Category 1 residents than for the reference category residents, but there was no difference in DK scores between categories for PGY-2 residents. MP scores were lower for PGY-2, but not for PGY-1, Category 1 residents than for the reference category residents. There was no difference in mean scores between each DH Category for PGY-2 and PGY-1 residents in the PP competency areas.

| DISCUSS ION
In this study of the effect of DHs on residents' competencies in different areas, we found that, overall, community hospital residents with <60 DHs per week had lower DK scores on the GM-ITE than those with 60-70 DHs per week, whereas those with >70 DHs per week did not have higher scores. Furthermore, <50 DHs were associated with lower CR and MP scores. When we stratified the analysis by PGY, CR scores decreased for residents in the ≤60 DHs group for PGY-2 and increased for those in the ≥80 DHs for PGY-1.
In residents with <50 DHs, the DK scores decreased only for PGY-1 residents, and the MP scores decreased only for PGY-2 residents.
Thus, there were discrepancies in the association between DHs and knowledge acquisition and its application in each competency area.
Knowledge acquisition for the competency areas of DK and CR was found to be related to resident DHs, with a positive trend up to 60-70 DHs per week, but no increase in the score when DHs exceeded 70 h. For CR knowledge acquisition, working more during PGY-1 temporarily raised the score, but for those in PGY-2, the scores of residents who worked more than 60 h per week eventually seemed to reach a plateau. Nevertheless, reducing DHs of PGY-1 could be a barrier to CR skill acquisition. Although these adjusted score differences are about 2%-4% in each category, the "cumula-  Note: Calculation of the average duty hours was based on the sum of weekday work duty, night emergency department (ED) duties, and weekend work duty. In this study, we found that working more than 60-70 h per week did not increase scores in any competency area for PGY-2 residents nearing the end of their training, although the relationship between scores in each of the competency areas and DHs was different.
McCoy et al. proposed four models for the relationship between patient encounter volume and knowledge acquisition: a linear relationship, threshold, Yerkes-Dodson curve (positive relationship up to a certain point, negative relationship thereafter), and null. 19 In their study, the patient encounter volume and in-training examination scores were positively associated, but in our study, we found that the relationship of DHs with CR and DK scores had a threshold. The in-training examination score of residents may improve as the patient encounter volume increases but increasing DHs further will not improve scores. Therefore, it may be necessary to set an appropriate upper limit for DHs. Our recent study also found that self-study time was shorter in residents who worked less than 60- [SD 5.6]). The residents who performed worse in the examination were more likely to refuse to participate in the study. Fourth, on F I G U R E 1 Estimates of mean score differences for each subcategory score between residents' duty hour categories, with multivariable adjustments. Note: Non-responders for Emergency Department duties and inpatients were included in "unknown." Non-responders for selfstudy time were excluded from the multivariable analysis. Error bars indicate 95% confidence interval. C1-C5, Category 1 to Category 5.
average, only one-third of all residents take the GM-ITE, which may introduce selection bias. Participation in GM-ITE is voluntary in training hospitals, which may be more education-oriented, and residents in-training hospitals may be more motivated to improve their skills. In addition, we excluded university hospital residents from our analysis as they tend to have greatly variable working patterns. To allow for a comparison of residents at university and community hospitals, the baseline characteristics of the university hospital residents are shown in Table S4 in Appendix S3. A summary of the examination scores is presented in Table S5 in Appendix S4. University hospital residents appear to have less ED duty, fewer assigned inpatients, less time for self-study, and lower scores than community hospital residents. Fifth, the study did not assess residents' baseline clinical knowledge. Therefore, to measure the impact of working hours on resident improvement directly, adjusting for baseline GM-ITE scores, national examination scores, and performance during medical school is necessary. Sixth, some community hospitals also have "tasukigake" programs. Thus, there are residents in community hospitals who work in a variety of work locations similar to those in university hospitals. However, this study did not examine the number of community hospital residents who partook in "tasukigake" programs.
In conclusion, our results suggest that the knowledge acquisition of CR, medical interviews, professionalism, and diseaserelated topics requires a certain level of DHs, while the application of knowledge of physical examination and procedures may not be related to DHs. In addition, the limit of 80 DHs per week starting in 2024 will have only a limited impact on the residents' acquisition of competency-related knowledge. In the future, these competencies should be evaluated using more suitable assessment methods, and their relationships with DHs should be investigated in more detail.

ACK N OWLED G M ENTS
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. The authors thank members of JAMEP and the question development committee and the peer-review committee of the GM-ITE for their assistance.
The authors thank Enago (www.enago.jp) for the English language review.

CO N FLI C T O F I NTE R E S T
The JAMEP was involved in collecting and managing data as the GM-ITE administrative organization. It did not participate in designing and conducting the study; data analysis and interpretation; preparation, review, or approval of the manuscript; and the decision to

F I G U R E 2
Estimates of mean score differences for each subcategory score between categories of residents' duty hours, with multivariable adjustments stratified by postgraduate year. Note: Non-responders for Emergency Department duties and inpatients were included in "unknown." Non-responders for self-study time were excluded from the multivariable analysis. Error bars indicate 95% confidence interval. C1-C5, Category 1 to Category 5.

DATA AVA I L A B I L I T Y S TAT E M E N T
Due to the nature of this research, participants of this study did not agree for their data to be shared publicly, so supporting data are not available.

E TH I C S A PPROVA L S TATEM ENT
The Ethics Review Board of Juntendo University School of Medicine approved the study.