Residency training is a wilderness that few non-physicians penetrate. Patients lying sick in a hospital bed rarely inquire of the person with a stethoscope whether he or she is an attending physician, a resident, an intern, or, for that matter, a medical student. (In some institutions, the length of the white coat is the giveaway, but few people know that short jackets identify medical students and knee-length coats “attending physicians,” the senior doctors formally in charge of a patient’s case.) More important, at least until the publicity around the Zion case, almost no one outside medicine appreciated the contradictions inherent in the residency system and their effects on the lives of residents and patients.
These contradictions start with the definition of a resident’s status and tasks. Residents are both advanced students and practitioners: the official term for the three-year resident program is “graduate medical education.” In fact, many teaching hospitals have dropped the labels “intern” and “resident” altogether, using instead the terms Post Graduate Year I, II, and III. Interns are in their first post-graduate year after taking their M.D. degree; residents are in their second and third. As graduate students, residents are expected to read widely in the medical literature and to consult regularly with their putative teachers—that is, the attending physicians. But residents are also practitioners, responsible on a day-to-day basis for diagnosing, treating, and monitoring anywhere from ten to thirty very sick patients. In many hospitals, they are also expected to do “scut work”: drawing blood, inserting intravenous lines, taking specimens to the lab, and so on.
The two roles do not fit together easily either in theory or in practice. Residents are supposed to be supervised as they care for their patients but not supervised too closely. According to the prevailing view, the only way they can learn to practice medicine is by actively making diagnoses, administering drugs, and performing various other medical procedures. If the process is monopolized by an attending physician, the resident’s training will suffer. Accordingly, the guidelines issued by the American Board of Internal Medicine, the certifying body for residency programs, state: “Residents must have continuing responsibility for most of the patients they admit…. Residents must be given the responsibility for decision making and for direct patient care.”2 Supervision is important, but direct experience with patients counts most.
While the official requirements are designed to give residents opportunities to make independent decisions, they have very wide discretion. The result is often what one reform-minded physician has called “resident-run residencies.” Residents usually have primary responsibility over ward patients (that is, poor patients paid for mainly by Medicaid), which gives them an opportunity to learn from treating patients who do not have personal doctors or specialists of their own. Senior physicians provide supervision, but only up to a point—they usually appear on the wards each morning for two hours and make the rounds of patients accompanied by residents and medical students; they may do the same for an hour or two in the afternoon if they are particularly diligent; they are very rarely present at night. For their part, residents on the wards are seldom eager to consult with the senior physicians; they want to work through the case themselves. In most teaching hospitals, to wake a supervisor with a question or to ask him to come to the hospital in the middle of the night would amount to a confession of ineptitude.
This system prompted Michel Foucault to observe that, with the rise of scientific medicine, the poor and the rich struck a Faustian bargain. The poor received treatment on the wards, and in return gave their bodies to the service of medical education. The rich benefited from this knowledge, and in return endowed and supported the hospitals.
But Foucault was only partly right. Residents learn not only by treating the poor but by caring for private patients as well. Even people who endow hospitals or have their own personal doctors or specialists seldom escape the system. All but a few private patients in teaching hospitals go “on service.” That is, they are seen regularly by the resident staff, which writes out their “orders,” including the tests and procedures they are to undergo and the medicine they are to be prescribed. Each patient has an “order book” in which his treatment is entered. It is not, as Robins sometimes suggests, officially closed to attending physicians in the sense that they are prohibited from writing in it; but in practice the orders are entered by the residents, both for “on service” patients as well as for ward patients. Indeed, the American Board of Internal Medicine flatly states: “Residents cannot be expected to show appropriate growth in understanding patient care responsibilities if others write orders for patients under the residents’ care.”
Robins describes both what she calls the “closed” order book—and “on service” care in general. Her discussion is useful in bringing practice in teaching hospitals into the open, but she ignores several critical points. First, the alternative to “on service” care is “off service,” which means that no one apart from nurses and the patient’s private doctor, whose office and home are both likely to be at some distance from the hospital, would be responsible for looking after the patient. (Hospitals usually have a so-called “house doctor,” who is often a moonlighting PGY III; he or she is usually available for emergencies, but little else.) Almost all private doctors put their seriously ill hospitalized patients “on service” so the patient will have around-the-clock scrutiny by the hospital staff, even if it is carried out by young residents.
Second, having residents in control of the order books is not as arbitrary a practice as it might appear. Since they are almost always assigned to a specific hospital floor, they are more likely to keep their eye on the order books of a particular group of patients; they observe the patient from hour to hour if necessary and this reduces the chance of errors arising from orders that are contradictory or should have been revised.
Finally, Robins fails to recognize that the ways doctors work with house staffs on behalf of private patients vary enormously. Many busy practitioners count on the residents to provide care for their patients and would never consider writing in the order books that the residents maintain. Some doctors insist on being called in if their patients begin sneezing; they refuse to give residents any leeway in making decisions and will take over the order book if the resident is not immediately on hand. Residents quickly learn each attending physician’s style; naturally they prefer the one who gives them the most latitude. The “closed” order book is not “the dirtiest little secret in medical education,” as Robins puts it, but, whether for reasons of ego or embarrassment, physicians generally fail to explain to patients how much of their treatment will be in the hands of residents.
Before Libby Zion’s death her family had little idea how a hospital worked. They assumed, for example, that since their doctor had sent Libby to the emergency room at New York Hospital, he would be in charge and on the scene. But the case was handled in the usual way—that is, Libby was put “on service.” In fact, the residents called and spoke to the attending physician several times. No record exists of these conversations, but it seems most unlikely from the evidence that the residents told him, or implied, that he should come to the hospital. Neither the admitting diagnosis (“viral syndrome with hysterical symptoms”) nor the admitting orders (to merely monitor the patient’s vital signs) suggested that Libby’s condition was critical. When a nurse asked an intern to see the increasingly agitated patient, the intern treated her request casually, as if there was no cause for intense concern. In those circumstances few private physicians would have gone to the hospital; most of them would have done so only if they had an urgent call from the house staff or if they had a special relationship with the patient.
Would a more experienced doctor have sent Libby to the intensive care unit? Should not the intern have come to see Libby as her fever rose? In hind-sight most of the physicians I know would say yes to both questions, but there is no way to be sure what should have been done at the time. All that can be said is that no senior physician on the hospital staff ever examined Libby. Neither New York Hospital nor most of the other hospitals I know of kept a senior doctor on hand to review decisions made by emergency room residents. That they should now do so was one of the two major proposals for reform that were provoked by Libby Zion’s death.
The second reform concerned how long residents work. Until the Zion case, it was commonplace for residents to be on the job 100 to 120 hours per week and to be on call every third or fourth night (during the 1930s and 1940s, they were on call every second night). Once or twice a week, each resident worked a thirty-six hour shift (from 7 AM the first day through 7 PM the second day, returning the next morning at 7 AM). The most obvious result was that residents did almost anything to cut down on their night-time tasks simply to get some sleep. A harrowing account of their working lives was given in 1986 by the sociologist Terry Mizrahi in her book Getting Rid of Patients, which described the variety of strategies residents used to avoid a “hit” (their term for a patient admission) or, failing that, to ease their workload. But so far as I know Mizrahi’s book had no effect on medical training.
Although virtually everyone in charge of medical schools and hospitals recognized the drawbacks of a 120-hour week, they not only accepted the arrangement but strongly defended it. Since patients needed “continuity of care,” so the argument went, it was better to have a tired doctor who knew the case than a rested one who had only read the chart. Critical illnesses, it was said, often develop over several twelve-hour periods; so shorter hours would mean that residents would miss a vital training experience. Most important, residents had to learn that medicine was a calling, not a job, and that one’s responsibility to one’s patients took precedence over all other considerations, including sleep. Put residents on a fixed schedule and they would come to adopt a lax “shift mentality.”
None of these propositions, however, was ever actually tested. In a profession that prides itself on rigorous trials for every new drug and procedure, few hospital directors experimented with different schedules or examined charts to see if the theory of twelve-hour periods had any bearing on how illnesses really developed. No one challenged the idea that sleeplessness instilled a professional ethos in interns or wondered why learning how to avoid seeing patients was so important a part of a young physician’s education. When the occasional critic noted that, apart from those in military service in wartime (and, some would add, new mothers or writers and editors facing deadlines), no one else regularly worked under conditions of sleep deprivation, the medical community insisted that a different system would be worse by far.
Ironically, as I have mentioned, the residents who treated Libby had not lost sleep. But once the facts of residency training came into the open, demands for change became inevitable. To the grand jury that reviewed the Libby Zion case, it seemed obvious that if the Federal Aviation Association said that pilots had to have “nine consecutive hours of rest for less than eight hours of scheduled flight time,” residents should not be treating patients for thirty-six hours straight. The grand jury also found that interns were not adequately supervised and recommended that hospitals change the prevailing arrangements for overseeing their work.
In response, David Axelrod, then head of the New York State Department of Health, appointed a commission chaired by Professor Bertram Bell of the Albert Einstein College of Medicine in 1986. After numerous hearings and over strenuous objections from doctors, the Bell Commission recommended several major reforms which were officially adopted in 1989 as Section 405 of the New York State Health Code. The work week of medical interns cannot exceed an average of eighty hours; they are not to work more than twenty-four consecutive hours and must have one twenty-four-hour period free each week. Every hospital emergency room has to have an attending physician present and on duty, and every hospital has to have an attending physician on call should residents require a consultation. The state provided hospitals with additional funds to defray the added costs of these reforms, including payments to auxiliary workers to do “scut work” and an increase of about 10 percent in the number of residency slots.
New York’s precedent has transformed residency training throughout the country. Although no other state has issued similarly specific regulations, practically every group that accredits residency programs now insists on a less crushing schedule (except in training for surgery, which has successfully resisted all pressures). The new requirements for residents in internal medicine, for example, are that they should spend, on average, no more than eighty hours a week in patient care; that they should be on call no more than one night out of three; and that they should have one day a week free. The administrators of hospital residency programs have adopted similar provisions—perhaps because they agree that they are necessary, but more likely because they must remain competitive with other programs. Had they not modified their demands, New York hospitals would have enjoyed a huge recruiting advantage over hospitals in other states.
The general consensus about the need to change residency programs has not produced any clear agreement on the value of the changes that have so far been made. As might be expected, the residents themselves are overwhelmingly in favor of them. Although the letter of the law is frequently violated (the hour limits are neither consistently followed nor enforced), workloads are generally lighter. Residents in internal medicine at Albert Einstein College of Medicine told researchers that “the regulations had reduced their fatigue on the wards and lessened emotional stress…. They now provided better patient care.”3 Residents in obstetrics and gynecology at the Columbia-Presbyterian Medical Center said that, with no diminution in the quality of patient care, there was “a marked improvement in the quality of their personal lives, the amount of sleep they achieved, and their level of stress.”4 Even surgical residents are convinced that “long consecutive hours impair quality of patient care.”5
On the other hand, senior physicians complain bitterly that residents have, as they had feared, adopted a shift mentality and that patient care has suffered. Residents, they say, leave tests uncompleted and ignore anyone who challenges them when they head for home. Most residents deny the charge, maintaining that they are not clock watchers and have not lost their sense of professional responsibility. Most attending physicians say they do not believe them.
Only one study has attempted to measure by objective standards the effects of the new rules on patient care. At New York Hospital, of all places, investigators reviewed patients’ charts immediately before and after the changes; they found that the new system increased delays in obtaining test results. The study also measured the changes in “in-hospital complications,” which include ignoring abnormal test results and making errors in prescriptions and dosage. It found no change had occurred in the total number of tests performed, the length of time patients spent in the hospital, the number of transfers to intensive care, or, most important, mortality rates.6 But the study has its weaknesses—the category of “in-hospital complication” is very loosely defined—and its findings might not hold true for other hospitals.
Whatever further investigations may show, there is no going back to the 120-hour workweek, in part because the public and the medical residents will not tolerate it, and in part, too, because residency in the 1990s, even after these changes, is a far more demanding and stressful experience than it was twenty years ago. Contrary to previous practice, Medicare and HMO regulations require that only people with very serious illness be admitted to a hospital. Patients also remain in hospitals for much shorter periods, from an average of twelve days in the early 1980s to under eight days now, owing to new and more complex technology for diagnosis and treatment as well as to constantly rising costs. It was one thing to work thirty-six hours when residents only had to read an EKG or listen to a stethoscope. It is quite another when they may be prescribing fast-acting cardiac drugs or performing more intricate and invasive diagnostic procedures, such as inserting a catheter into the heart to measure its output of blood.
In any case, the private doctor, who acted both as an entrepreneur and as a heroic figure to his patients, is giving way to the team member who is part of an HMO or a physician-run practice plan, an interchangeable part of a larger medical system. More and more he is forced by the schedules and rules of group practice to hand off a patient to colleagues and to try to do so in a way that ensures continuity of treatment. The once traditional question “Who is your doctor?” is changing to “Which health care plan do you belong to?” How many people now in their twenties are likely to develop a close personal relation with, and loyalty to, one particular doctor? In view of the direction in which the medical system seems to be heading, a carefully designed system of shifts may be precisely what doctors require.
Changes in medicine are beginning to mirror the “restructuring” that is taking place within other organizations. The doctor-patient relationship is increasingly formal and bureaucratic; more and more it is an exchange between strangers based on increasingly sophisticated methods of testing and treatment. In many cases medical practice has become more efficient, but it may also leave some patients feeling stranded, without the intimate observation of their symptoms and the sympathetic advice they formerly expected. This new situation is not what Sidney Zion had in mind when he began his campaign; but it may help to explain the degree to which he succeeded in changing the way doctors work in America.
Graduate Medical Education Directory, 1995, pp. 62–63.↩
Joseph Conigliaro, William Frishman, Eliot Lazar, and Lila Croen, "Internal Medicine Housestaff and...the Impact of New York State Section 405 Regulations...," Journal of General Internal Medicine, 8 (1993), p. 505.↩
Amalia Kelly, Frances Marks, Carolyn Westhoff, and Mortimer Rosen, "The Effect of the New York State Restrictions on Resident Work Hours," Obstetrics and Gynecology, 78 (1991), p. 468.↩
Steven Ruby, Lisa Allen, Peter Fielding, and Peter Deckers, "Survey of Residents' Attitudes Toward Reform of Work Hours," Archives of Surgery, 125 (1990), p. 765.↩
Christine Laine, Lee Goldman, Jane Soukup and Joseph Hayes, "The Impact of a Regulation Restricting Medical House Staff Working Hours on the Quality of Patient Care," Journal of the American Medical Association, 269 (1993), pp. 374–378.↩
The Libby Zion Case: An Exchange May 9, 1996
Graduate Medical Education Directory, 1995, pp. 62–63.↩
Joseph Conigliaro, William Frishman, Eliot Lazar, and Lila Croen, “Internal Medicine Housestaff and…the Impact of New York State Section 405 Regulations…,” Journal of General Internal Medicine, 8 (1993), p. 505.↩
Amalia Kelly, Frances Marks, Carolyn Westhoff, and Mortimer Rosen, “The Effect of the New York State Restrictions on Resident Work Hours,” Obstetrics and Gynecology, 78 (1991), p. 468.↩
Steven Ruby, Lisa Allen, Peter Fielding, and Peter Deckers, “Survey of Residents’ Attitudes Toward Reform of Work Hours,” Archives of Surgery, 125 (1990), p. 765.↩
Christine Laine, Lee Goldman, Jane Soukup and Joseph Hayes, “The Impact of a Regulation Restricting Medical House Staff Working Hours on the Quality of Patient Care,” Journal of the American Medical Association, 269 (1993), pp. 374–378.↩