by J. Lance Lichtor, M.D., James Zacny, Ph.D., Bradford S. Lane, B.A., Michael L. Good, M.D.
Should medical personnel be permitted to go home in the morning after a night on call, work for 36 hours more or longer, or are 12 hour shifts more appropriate? Is performance impaired even on the day following normal sleep if it has been preceded by a night without sleep?
That lack of sleep may affect the performance of house officers has been a concern for many years. The problem was dramatically publicized in the case of Libby Zion, an 18-year-old who died in a New York hospital. Strong allegations were made that inadequate care by an overworked and sleep-deprived house officer contributed to her death. The case received wide attention in the media (the patient's father is an attorney and writer for the New York Times) and in the medical literature.(1,2) One of the recommendations of the grand jury investigating Libby Zion's death was that "The State Department of Health should promulgate regulations to limit consecutive working hours for interns and junior residents in teaching hospitals."(3) Subsequent rules issued in New York limiting the hours worked by residents are estimated to cost New York State more than $350 million dollars per year.(4) Those rules may have been based more on politics than on actual scientific findings.
Many authors have attempted to determine scientifically how lack of sleep affects physician performance. Performance in anesthesiologists after a night on call has not received as much study. Certainly, the pattern of practice by anesthesiologists is markedly different from that in most other specialties because the degree of concentration required during anethesia care is more intense.
We performed two studies to determine how loss of sleep affected performance in anesthesiologists who work 24 hour shifts compared to other medical personnel who work 12 or 36 hour shifts. In the first study, we determined how cognitive and psychomotor function (and moods as well) compare in anesthesiologists, surgeons, and nurses after a night on call and after a night not on call. We compared our results with similar data from another study in which healthy volunteers drank alcohol. Impairment on both a coordination task and a vision task (Maddox Wing) at the end of call was similar to impairment seen after low-dose alcohol (0.5 g/kg) ingestion. Further, residual performance decrements were seen the next day after a night of sleep at home that had been preceded by a night on call. Although performance deficits were seen the following morning, the majority were seen the following afternoon.
In a second study, we used a realistic testing environment created by a modified anesthesia simulator.(5) The anesthesia simulator replicates a clinical workstation in the ICU or operating room and comprises: 1. a life-like patient mannequin, 2. a physiological lung model that exhales carbon dioxide, 3. a standard anesthesia gas machine, mechanical ventilator, and circle anesthesia breathing system modified with electromechanical actuators to create malfunctions, 4. standard(6) intraoperative monitoring instruments, including electrocardiograph (ECG), noninvasive blood pressure monitor, pulse oximeter, capnograph, oxygen analyzer, and thermometer, 5. a computer screen that displays dosing and fluid problems to be solved, and 6. a controlling micro computer. Also, the task of mentally retaining story material was used to simulate the ability to recall a patient's history.
Anesthesiologists who worked 24 hour shifts were compared to surgeons and pediatricians who worked 36 hour shifts. Simulator performance by anesthesia, surgical, and pediatric personnel was not impaired after an on-call period, although anesthesiologists were better able to perform on this task compared to surgeons or pediatricians. Subjects did demonstrate impairment the day after call in story retention: the less sleep, the greater was the impairment in long-term ability to recall a story.
Some impairment was demonstrated in both studies, although the impairment was not as great as we expected. We were very careful to make certain that subjects were not fatigued before they actually started both studies: subjects had to have a normal night's sleep before a night when they were tested while on-call. This has been a problem in some studies, where performance of residents in the "control" group also may have been impaired from chronic sleep deprivation.(7) We also attempted to maintain the motivation of the volunteers, in that a portion received extra cash if they could perform better than others.
Our work suggests that length of call should be limited. However, motivation and chronic sleep deprivation should also be considered in making a decision how long individuals should remain in the hospital after a night of no sleep.
Supported in part by NIMH RO3MH453860, and a grant from APSF.
Dr. Lichtor and colleagues are in the Department of Anesthesia and Critical Care at the University of Chicago; Dr. Good is from the University of Florida.