Medical Students on Their Internal Medicine Clerkship Experience Short Sleep Duration

Abstract

Background

Third-year medical students during their internal medicine clerkship may be predisposed to short sleep duration (<7 h of sleep per night) because of rigorous clinical schedules and academic demands.

Objective

To evaluate the prevalence of short sleep duration, its impact on performance, and perceived causes among third-year medical students on their internal medicine clerkship.

Methods

During the 2023–2024 academic year, third-year medical students at Rutgers Robert Wood Johnson Medical School (RWJMS) and Rutgers New Jersey Medical School completed a survey at the end of their internal medicine clerkship regarding their sleep practices. The schools differed in clerkship structure and grading system. Data were analyzed and compared between schools using the Fisher exact test. Responses for an open-ended question on sleep strategies were categorized into themes by sleep physicians.

Results

Of the 314 third-year medical students invited, 222 (70.7%) completed the survey. Short sleep duration was reported by nearly two-thirds of our cohort (143, 64.4%), with a significantly higher prevalence at RWJMS. Most reported sleep-related impairment on clinical duties or educational activities for 1–3 days per week (125, 56.3%), with more reporting ⩾4 days per week at RWJMS. About one-third of our cohort reported drowsy driving either 1–3 days (82, 36.9%) or ⩾4 days (68, 30.6%) per week. Despite high rates of short sleep duration, the overwhelming majority (196, 88.3%) had never received education on sleep management.

Conclusion

Short sleep duration is prevalent among third-year medical students during their internal medicine clerkship, potentially impacting their clinical performance and safety.

The American Academy of Sleep Medicine (AASM) and Sleep Research Society (SRS) recommend adults regularly sleep ⩾7 hours per night (1). One-third of adults in the United States report short sleep duration (defined as <7 h per night), which impairs cognitive function and has a dose–response relationship with wide-ranging adverse effects, including cardiovascular events, metabolic syndrome, diabetes, and mortality (2–7). Lifestyle choices (e.g., consumption of alcohol and tobacco), occupation, and sleep disorders can predispose toward short sleep duration (8). Healthcare professionals have among the highest prevalence of short sleep duration because of long workdays, inconsistent schedules, and undiagnosed or untreated sleep disorders, which contribute to high rates of burnout (9–11).

Medical students also face similar challenges as healthcare professionals because of their educational and professional obligations (12). Insufficient sleep profoundly impacts learning and wellness, both of which are integral components of the Accreditation Council for Graduate Medical Education (ACGME) clinical education milestones (13–15). These milestones serve as a framework for assessing the educational skills of faculty members involved in undergraduate and graduate medical education. From an educational perspective, sufficient sleep is vital for fostering both effective learning and overall well-being, which are foundational pillars of successful clinical educator milestones. Problematically, surveys at U.S. medical schools found short sleep duration among medical students contributed to daytime sleepiness, drowsy driving, lower professional efficacy, burnout, and poor mental health (16–18). In addition, medical students reported worse sleep quality than a college-age cohort (19). Transitioning into clinical clerkships can introduce new educational and professional obligations that compete with getting sufficient sleep, and during the third year of medical school sleep duration falls to its lowest (19, 20).

The internal medicine (IM) clerkship is a core clinical rotation during which third-year medical students learn to diagnose, manage, and treat a wide range of adult illnesses through hands-on clinical experiences, didactics, and self-directed learning. Despite the IM clerkship’s known rigor and long hours, the prevalence and impact of short sleep duration in this population have not been formally studied. In this study, we surveyed third-year students on their IM clerkship across two medical schools to identify subjective sleep duration and assess its perceived causes and impact.

Methods

During the 2023–2024 academic year, third-year medical students at Rutgers Robert Wood Johnson Medical School (RWJMS) and Rutgers New Jersey Medical School (NJMS) were invited to complete a survey toward the end of their IM clerkship. At RWJMS, the IM clerkship is condensed into 6 weeks of inpatient general medicine, whereas NJMS divides its IM clerkship into three 3-week blocks, encompassing two blocks of inpatient general medicine and a block of medical subspecialty service. At both schools, the National Board of Medical Examiners shelf exam was administered at the end of the clerkship. Both RWJMS and NJMS limit medical student work hours to 80 hours per week. RWJMS has a five-tiered grading system (honors, high pass, pass, conditional pass, or fail) based 80% on clinical feedback and 20% on shelf exam score, whereas NJMS uses a four-tiered grading system (honors, high pass, pass, or fail) based 40% on clinical feedback, 30% on shelf exam score, 15% on an objective structured clinical exam, 10% on a graded history and physical note, and 5% on completion of online clinical cases.

The sleep-related survey questions were designed to assess the following: average sleep duration, sleepiness during education and clinical duties, drowsy driving, and both receipt of and desire for education on sleep management (see Appendix E1 in the data supplement). We used the AASM definition of drowsy driving as operation of a motor vehicle while impaired by a lack of adequate sleep, whether due to chronic or acute sleep loss (21). In addition, we included an open-ended question inviting students to share their strategies for maintaining healthy sleep habits.

Data were summarized as frequencies, and differences between schools were determined using Fisher exact test (P < 0.05). All analyses were conducted using R Studio (4.0.3). The qualitative analysis was informed by the Health Belief Model, which proposes that the decision to take protective health actions—including applying strategies for maintaining health sleep habits—is motivated by perceived threats, benefits, self-efficacy, and barriers (22–24). An inductive and deductive approach was taken by two board-certified sleep physicians (25). After identifying main emerging themes through an initial reading of the responses, they then independently organized each response as belonging to neither, either, or multiple themes. Interrater reliability for each theme was reported as Cohen’s kappa and categorized as fair (κ = 0.21–0.40), moderate (κ = 0.41–0.61), substantial (κ = 0.61–0.80), or almost perfect (κ = 0.81–1.00), and disagreements were resolved by a third board-certified sleep physician. Within each theme, responses were coded during further reads into subthemes related to the Health Belief Model.

Results

Survey Responses

Of the 314 third-year medical students invited to participate across both sites, 222 (70.7%) completed the survey. Nearly two-thirds of our cohort (143, 64.4%) slept <7 hours per night during the IM clerkship (Table 1). More than half of our cohort (125, 56.3%) reported 1–3 days per week when they felt inadequate sleep impacted their ability to fulfill clinical duties and/or participate in educational activities. About two-thirds of our cohort spent 1–3 days (82, 36.9%) or >4 days (68, 30.6%) drowsy driving per week. An overwhelming majority (196, 88.3%) had never received any formal education on sleep management, and more than a third reported interest in receiving it (85, 38.3%) (Table 2).

Table 1.

Short sleep duration and impact on performance and safety

Characteristic	Total (N = 222)	MS-3 at NJMS (n = 139)	MS-3 at RWJMS (n = 83)	P Value
Average sleep duration per night
⩽6 h	143 (64.4)	80 (57.6)	63 (75.9)	0.006
⩾7 h	79 (35.6)	59 (42.4)	20 (24.1)
Days per week spent tired during clinical or educational activities
Never	38 (17.1)	30 (21.6)	8 (9.6)	0.011
1–3 times	125 (56.3)	80 (57.6)	45 (54.2)
4–7 times	59 (26.6)	29 (20.9)	30 (36.1)
Days per week spent drowsy driving
Never	72 (32.4)	43 (30.9)	29 (34.9)	0.26
1–3 times	82 (36.9)	57 (41.0)	25 (30.1)
4–7 times	68 (30.6)	39 (28.1)	29 (34.9)

Definition of abbreviations: MS-3 = third-year medical student; NJMS = New Jersey Medical School; RWJMS = Robert Wood Johnson Medical School.

Data are given as n (%). This table shows there is a significant difference between the two schools in the proportion of third-year medical students who had short sleep duration and its impact on clinical or education activities while on their internal medicine clerkship.

Table 2.

Lack of and need for sleep education

Characteristic	Total (N = 222)	MS-3 at NJMS (n = 139)	MS-3 at RWJMS (n = 83)	P Value
Had prior education on sleep management
No	196 (88.3)	120 (86.3)	76 (91.6)	0.29
Yes	26 (11.7)	19 (13.7)	7 (8.4)
Requested education on sleep management
No	137 (61.7)	90.0 (64.7)	47 (56.6)	0.25
Yes	85 (38.3)	49 (35.3)	36 (43.4)

Definition of abbreviations: MS-3 = third-year medical student; NJMS = New Jersey Medical School; RWJMS = Robert Wood Johnson Medical School.

Data are given as n (%). This table demonstrates that majority of third-year medical students at both schools had never received education on sleep management, and a sizable proportion of them would find it helpful.

Responses to the open-ended question were shared by most of our cohort (197, 88.7%) and coded into two themes by board-certified sleep physicians: sleep hygiene and clerkship responsibilities (Table 3). The kappa value was 0.7 for sleep hygiene and 0.8 for clerkship responsibilities, indicating substantial agreement, and the third board-certified sleep physician resolved disagreements for 39 (19.8%) responses. Example responses to the open-ended questions are shared in Appendix E2.

Table 3.

Prevalence of themes in student responses judged by sleep physicians

Characteristic	Total (N = 197)	MS-3 at NJMS (n = 119)	MS-3 at RWJMS (n = 78)	P Value
Sleep hygiene
No	26 (13.2)	9 (7.6)	17 (21.8)	0.005
Yes	171 (86.8)	110 (92.4)	61 (78.2)
Clerkship responsibilities
No	163 (82.7)	107 (89.9)	56 (71.8)	0.002
Yes	34 (17.3)	12 (10.1)	22 (28.2)

Definition of abbreviations: MS-3 = third-year medical student; NJMS = New Jersey Medical School; RWJMS = Robert Wood Johnson Medical School.

Data are given as n (%). In our qualitative analysis, we categorized each response and compared the frequency of each theme between the schools.

Theme 1: Sleep Hygiene

Most responses detailed sleep hygiene behaviors (171, 86.8%). Many adhered to a routine sleep schedule and tried to achieve a certain number of hours of sleep each night. Others mentioned reducing coffee intake and screen usage, avoiding late meals, winding down before heading to bed, and catching up on sleep through naps or during the weekend. Less common sleep hygiene behaviors included taking melatonin, controlling their sleep environment (e.g., ear plugs, curtains, room temperature), limiting activities in bed aside from sleep and sex, and engaging in regular exercise and meditation.

Responses were mixed in demonstrating self-efficacy. Although some were “adamant about getting my 7–8 hours” and maintained a “strict sleep schedule,” others conveyed less confidence, stating they “try to sleep” by a certain time. In addition, a few responses reflected a perceived threat of short sleep duration and/or and benefit of healthy sleep, as participants chose to “prioritize sleep,” even at the “expense of preparing for rounds, reading up on patients, and doing practice questions.”

Theme 2: Clerkship Responsibilities

A smaller proportion of responses shared concerns and frustrations about clerkship responsibilities (34, 17.3%). The primary complaint was that between long work hours, charting, finishing notes, and studying for the shelf exam, the IM clerkship afforded insufficient time for sleep.

Responses within this theme mainly highlighted barriers to obtaining healthy sleep. Some suggested reduced work hours, fewer weekend calls, and dedicated study time. Interestingly, a few at RWJMS asked for a longer rotation, as there are “so many UWorld questions to do in 6 weeks.” One medical student articulated a perceived threat of short sleep duration by mentioning they needed at least 6 hours of sleep to “function at a satisfactory level during the day.” Another wrote that “if you want to do very well you will be tired,” indicating that, for some, the perceived benefit of healthy sleep may be outweighed by the desire to excel on the rotation.

Discussion

Our study confirms that third-year medical students transitioning to clinical years surveyed at the end of their IM clerkship experience short sleep duration, potentially impacting their performance and safety. When prompted to share strategies to maintain healthy sleep habits, most mentioned practicing sleep hygiene behaviors, whereas a few cited clerkship responsibilities that limited time available for sleep. The majority of the medical students had not received education on maintaining healthy sleep.

Although not mandated, work hour limitations have been overwhelmingly adapted for medical students based on policy changes made over concerns for patient safety due to sleep deprivation among residents (26, 27). In 2003, the ACGME capped resident work hours to 80 hours per week and no more than 24 hours consecutively, plus an additional 6 hours for transfer of care and/or didactics (28). In 2011, the ACGME further reduced maximum shift lengths for first-year residents to 16 hours; although this decision was rescinded in 2017, programs could choose to retain the limitation (29, 30). These reforms did not yield a significant difference in resident performance on board exams and patient outcomes like mortality, nor did they definitively improve resident sleep and well-being (31–37). Notably, medical students surveyed after these changes reported residents provided more feedback and teaching, but this depended on how programs accounted for gaps left by reduced resident work through additional staffing or coverage (38–40). In studies that followed residents in the intensive care unit (ICU), restricting work hours improved sleep duration and lowered the rate of attentional failures by half (41). Although Landrigan and colleagues determined that resident schedules in the ICU without extended shifts led to more serious errors, their secondary analysis suggests this association was confounded by an increased number of patients followed by each resident (42). Only a few medical schools require a fourth-year rotation in the ICU, during which sleep duration may differ from that of an IM clerkship because of distinct responsibilities and requirements (43). There are no studies that specifically examine the prevalence or impact of sleep among fourth-year medical students rotating in the ICU, where the hours and responsibilities may be even more demanding, particularly when students are serving as “sub-interns.”

In our study, a high proportion of medical students (36.9%) reported drowsy driving at least once per week, although this likely reflected chronic sleep restriction rather than acute sleep deprivation. Each site ensured one day off every 7 days and that weekly hours did not exceed 80 hours, but there was no oversight regarding the number of hours worked on each individual day. After about 17 hours without sleep, cognitive performance declines to a degree comparable to a blood alcohol concentration of 0.05% (44, 45). After 24 hours, the blood alcohol concentration equivalent becomes 0.10% which exceeds the legal driving limit in the United States (44, 45). We found a much higher prevalence of short sleep duration and drowsy driving than a survey done with second-year students, which implicates stressors brought by clinical responsibilities during the IM clerkship may play a role (17). Among residents, short sleep duration is associated with riskier driving, which is concerning, considering motor vehicle accidents are the third-leading cause of deaths among residents after neoplastic disease and suicide (46–48). Importantly, extended work hours are associated with drowsy driving, and their elimination reduces the risk of crashes, especially during the commute from work (47, 49). There are no specific data on car accidents involving third-year medical students. However, our findings indicate that a significant portion of these students drive while fatigued or drowsy multiple times per week, which may increase their risk of car accidents, a trend similarly observed among residents. A few schools have also begun implementing ride-sharing programs for their residents, although their availability for medical students is not widely known.

The overwhelming majority of third-year medical students (88.3%) we surveyed had never received formal sleep education, and one-third believed they would benefit from it. Because of an already packed curriculum, most medical schools devote only 2–4 hours to sleep education (50). The ACGME mandates that residency programs educate both residents and faculty to recognize the signs of sleep deprivation and implement fatigue mitigation strategies (51). However, there is currently no standardized approach or formal process for delivering this education (52). In addition, nonpharmacologic therapies, such as education for residents to mitigate sleep deprivation and fatigue, have variable effectiveness. Meanwhile, the Liaison Committee for Medical Education does not have any mandates regarding sleep and fatigue management for the medical students. Currently, there are no data on the impact of sleep education for medical students as they transition from nonclinical to clinical years. Providing sleep education can potentially help medical students recognize often underdiagnosed sleep disorders, apply sleep knowledge to improve their own sleep, and possibly provide evidence-based strategies to mitigate sleep and fatigue during highly demanding rotations (53–57). Health literacy frameworks would suggest that medial students and healthcare professionals would be best poised to translate sleep knowledge into healthy sleep behavior because of their ability to understand health information (58). However, sleep education as an intervention alone may not address the underlying causes of short sleep duration, as sleep knowledge gained throughout medical school has not been associated with improved sleep quality or duration (59, 60). Arora and colleagues found that delivering a 60- to 90-minute lecture to residents on the neurobiology of sleep, the effects of sleep loss, and effective countermeasures did not impact sleep (61). Although single education sessions may not have a significant impact on changing sleep habits, longitudinal educational programs may improve sleep while mitigating fatigue. In addition, better data on sleep patterns could allow medical schools and residency programs to identify stressors that contribute to short sleep duration and plan schedules that reduce the risk of fatigue (62).

Limitations

As a cross-sectional study, our results are limited by recall bias, as the survey was taken toward the end of clerkship. However, the high completion rate helps mitigate the risk of selection bias. In addition, our study did not incorporate validated sleep questionnaires, such as the Epworth Sleepiness Scale, Pittsburgh Sleep Quality Index, Sleep Hygiene Index, or Functional Outcomes of Sleep Questionnaire, because of concerns that including them would make the end-of-clerkship survey too lengthy for the medical students. Moreover, the lack of objective sleep data limited our ability to correlate reported sleepiness and fatigue with actual sleep patterns. Nevertheless, future studies that incorporate validated sleep questionnaires alongside wearable devices to measure sleep duration could provide a clearer understanding of sleep patterns and their effects on medical students (15, 41, 42, 63). The results of our study may have also been influenced by the variability between institutions and individual rotation weeks, including differences in rounding schedules, the balance of outpatient versus inpatient rotations, expectations for work outside the hospital, and individual commute times. Despite these variations, the findings highlight that third-year medical students frequently experienced insufficient sleep, which had a negative impact on their performance and safety. Future studies, in addition to collecting objective sleep data, could explore additional factors that may account for the differences observed in responses between medical schools, which were beyond the scope of this study. Overall, this study adds to the discussion of short sleep duration among medical students by being the first to survey those on their IM clerkship regarding sleep duration and the impact of fatigue on performance and safety.

Conclusions

Our multicenter study shows that third-year medical students on their IM clerkship reported high rates of short sleep duration, interfering with their performance and safety.