Why resident work-hour restrictions should be individualized


There’s been a bit of kerfuffle over resident duty hours lately.

For those unfamiliar with the topic, physicians in training in the United States have traditionally lived in the hospital — hence why they were called residents — and available to patients 24/7.  Over time, concerns about patient safety led to limits on how many hours could be worked consecutively in the hospital.  In 2003, the maximum number of hours worked per week was restricted to 80 hours.  In 2011, the maximum work day for first-year residents was restricted to 16 hours — for second-year residents and above the limit was set at 24 hours of work plus 6 hours for handing off patient care.

These limitations have been bitterly controversial.  Some traditionalists point to steadily declining board pass rates and decreased confidence reported by graduating residents.  A large literature review found no evidence that the 2011 restrictions reduced fatigue or patient safety- but did find that restricting work hours negatively impacted education.  Many doctors, including residents now working 30 hour shifts, agree that longer continuous hours allow doctors to get to know patients better, and prevents errors that arise from doctors who do not know the patients well trying to make decisions about their care.

On the other side are many who believe the work hours are simply too long, and point to studies showing declines in attention/concentration and reaction time and decreases in cognitive performance in doctors working long hours.  Not to mention the well known increase in car accidents after long shifts.

The controversy recently took on a new turn: in order to figure out the answer to the question of whether or not the most recent duty hour restrictions (16 hour shifts vs. 30 hour shifts for first-year residents) had improved or worsened patient safety, the Institute of Medicine called for a study comparing doctors under both sets of restrictions.  The ACGME and other bodies duly got such a study going — only to be heavily criticized by several groups, including participating residents, for allowing such studies at all.  Negative comments focused on allowing such studies without patient consent and for accidentally failing to notify psychiatry interns who rotate on internal medicine that they would be subject to these rules.

I happen to think such criticism is absurd.  Patient consent is waived when study risks to patients are minimal — and considering the mountain of data which shows that no measurable increase in safety has occurred as a result of duty hour restrictions up to this point, it is entirely reasonable to proceed.  It would be impossible to consent every patient that enters a hospital and arrange scheduling such that non-consenting patients would be seen only by interns working the old system.  As for the residents, while the oversight with the psychiatry residents not being warned about this study was unfortunate, all the medicine interns who interviewed the year before were told of the potential of inclusion in this study as they were applying.  If strict informed consent requirements were applied, with interns able to opt-out after selecting the program to match to and patients able to choose not to be seen by a team with the new schedule, this study would never have been carried out.  And indeed in studies of hand-off procedures, which vastly affect patient care, patient and resident informed consent is routinely waived.

Opponents of the study cite a recent public survey which reveals that the U.S. is overwhelmingly against long shifts for medical residents — but such opposition is ill-informed.  To patients, every doctor is all-knowing and can figure out what’s wrong with their loved in the middle of the night as they are crashing in the SICU.  They do not understand just how difficult it is for a resident overnight who has never seen the patient before in their life to figure out what is going on — relative to a tired resident who nonetheless took their history and knows their hospital course like the back of their hand.

But all of this dust-up misses a larger point: Why in the era of personalized everything do we still treat all residents as if they are the same?

From personal experience, I can testify to the wide variability in ability of medical students, residents, and even attendings to handle a night without sleep.  I was awed by an elderly trauma surgeon who had the stamina to run me into the ground at 3 a.m. — and was less than impressed on occasion with how some residents handled the strain of long shifts.  One of the studies from earlier found that 75.8 percent of intensivists had lower performances after a night shift on various cognitive tasks.  This means that 24.2 percent did not.  Now, the clinical relevance of these cognitive tests is still up for debate — again, a tired intensivist who knows the patient may be much better than a fresh one who has no clue what’s going on.

But regardless, isn’t it amazing that 1/4 of doctors can apparently shrug off the lack of sleep with no deficiency in performance?  And on the other side, how badly are the worst affected?  From that same study, cognitive flexibility (a marker of executive function whereby higher scores indicate worse performance) increased from 41 to 44 (the scale is 32 to 64, with 32 being the best).  However, the worst 5 percent affected (assuming normal distribution) had their cognitive flexibility worsen to about 47.  How significant that is is again up for debate — but I believe it supports my point that how sleep deprivation affects doctors is highly variable.

And this is not to get into the reality of medical conditions like bipolar disorder, schizophrenia, lupus, and inflammatory bowel disease.  While examples abound of professionals who work safely and effectively with severe medical conditions under good control, sleep deprivation is a known trigger for such conditions to worsen.  Such doctors should not ever be required to work long hours.

All this suggests a different approach to duty hour restrictions: individualization.  Different programs have different needs.  It’s pretty unlikely that a family medicine (FM) resident working at Kaiser (a large private hospital chain) will need to be up most nights.  On the other hand at Ventura County, home to one of the strongest FM residencies in the country, FM residents are the only trainees in a level II trauma center and the only tertiary care provider in the county.  These FM residents have to be there day and night for car crashes, gunshot wounds, and septic babies who are an hour away from death without prompt intervention.  Granted, hours like that are rare for an FM residency — but they exist.

Medical students should be evaluated before they apply for residency for their ability to handle long, heavy calls.  They should work at least one rotation with 30-hour call shifts, and undergo some sort of psychometric testing at the end to see if they can function effectively.  The Kaiser FM residency will probably be more than happy to take a student who simply does not function as well after long hours but who is still bright, affable, enthusiastic, and professional.  Ventura County, probably not so much — if that resident can’t be trusted at 2 a.m. with a crashing neonate, they’re probably not the best fit for the program.

Some of this already happens, and is self-selected.  I purposefully sought out rotations with heavy call, and volunteered for extra call shifts on other rotations to prove to myself that I could work those hours and still function.  After all, I plan on applying into general surgery, a field with almost uniformly brutal hours for residency.  But even in general surgery, programs exist with lighter hours — though they are as rare as Ventura County’s more heavy schedule is among FM residencies.  And after a general surgery residency, few who specialize in endocrine, colorectal, or bariatric surgery will get called in the middle of the night while those who specialize in pediatric or acute care surgery can count on rarely getting sleep.

Ultimately the country has a need both for doctors who can work long hours and for doctors who have other strengths.  I can personally testify, and I believe the evidence will soon bear out that long call shifts are often times better for both the patient and the physician, and that the 16-hour restriction was probably a bad idea.  But there are many fine doctors who probably should not be forced to work 30 hours at a stretch.  Programs ought to clearly delineate their needs, and medical students ought to be evaluated uniformly to see what kind of hours they can handle — before they pick a program where they may be miserable and their patients unsafe.

Vamsi Aribindi is a medical student who blogs at the Medical Intellectual.

Image credit: Shutterstock.com


View 21 Comments >

Most Popular

✓ Join 150,000+ subscribers
✓ Get KevinMD's most popular stories