A month into Sandeep Jauhar's medical internship at a prominent teaching hospital in New York City, he was asked to drain fluid from the belly of a patient who was HIV-positive. "I was trying to get out of the hospital to keep a dinner appointment," he recalls. "I was sort of rushing. I heard a snap and there was all this fluid leaking all over the floor." Jauher's gloves were too small, he hadn't assembled the tubes for the blood correctly, he was new, he was inexperienced and nobody was watching. "[The patient] was totally oblivious to the disaster, but it was a mess," he says. "These are the mistakes that new, green interns can make."
According to conventional wisdom, a patient's chances of encountering a mistake-prone rookie like Jauhar go way up in the summer. That's because July 1 is the start of the academic year for medical schools: In teaching hospitals around the country, medical students will replace interns, interns will replace residents and residents will move on to fellowships or to become full doctors.
This crucial and sometimes perilous training period can be incredibly difficult for medical students. As Jauhar writes in his recent book, "Intern, A Doctor's Initiation," incoming doctors are not only practicing on patients for the first time, they're also learning the often Byzantine workings of their respective hospitals, new technical language, new procedures and the tedious, yet critical, ways to fill out paperwork. All this learning is packed into 80-hour workweeks and overnight shifts in a busy hospital environment—a far cry from the academic environment they might be coming from. But is it really riskier to go into a teaching hospital during those first few weeks of intern training? Or is the "July phenomenon" a medical myth?
The number of mishaps related to newbie interns is hard to pin down. After all, most doctors may not be as forthcoming as Jauher is about his mistakes. However, no one disputes that hospital errors do occur and they do cost lives. In 1998, the Institute of Medicine found that up to 98,000 deaths annually are caused by medical mistakes, and in the decade since, that number has hardly improved, according to experts in patient safety. In fact, the Institute for Healthcare Improvement (IHI), a Massachusetts-based independent not-for-profit organization, estimates that 15 million incidents of medical harm, both deaths and other "adverse events," occur in the United States each year.
While research into whether those medical mishaps are related to the July phenomenon has been sparse and often ambiguous, one study found that the July medical-training period is associated with between 1,500 and 2,750 accelerated deaths every year. In a study of the July phenomenon from which initial findings were released in 2005 by the National Bureau of Economic Research, Harvard Business School health-care economists Robert Huckman and Jason Barro compared mortality rates in teaching and non-teaching hospitals around the country. They found that there are 4 percent more incidences of accelerated death in average-sized teaching hospitals in July and August.
That, say some experts, may be the unavoidable price of medical-training programs. "New physicians, just like new people in any profession, are going to make mistakes," Huckman says. "There's nothing we can do about that."
But other studies have not found a link between those early weeks of medical training and an increase in patient mortality. Huckman himself calls the 4 percent figure a little misleading. "If not for the turnover, these individuals might have survived hospitalization," he says. "But for people who die in hospitals, it's quite likely that they have a condition that's fairly serious to begin with. These are deaths that we observe that effectively occur earlier than they might otherwise occur. And when you look at it in respect to the other things that go on in hospitals that lead to patient mortality, I wouldn't say that this is a huge cost. It's important to put it in context."
Jay Bhattacharya, a professor at Stanford Medical School who also teaches on health care economics, says that up until the Harvard study, he considered the July phenomenon more medical myth than reality. Bhattacharya and others believe that the mistakes newbie doctors make by and large involve unnecessary tests and longer hospital stays, errors that aren't typically life-threatening. "But a lot of times, those kinds of things do snowball," he says. Bhattacharya says he "wouldn't paint an alarmist picture about it, but it's worth knowing."
Other research has found no increase in death rates during July. In a 2003 study, Doctors William A. Barry and Gary Rosenthal of the University of Iowa, looked at patient mortality rates in intensive care units and found no significant difference for the same period. "Fortunately, the results were boring," Dr. Rosenthal says. "What it suggests is that there's enough checks in the system that patients are not at undue risk. It's possible the attending physicians are a little more vigilant [in the summer]."
A bigger safety problem may be the routine patient handoffs that happen several times a day in hospitals. David Stevens, editor of Quality and Safety in Healthcare at the Dartmouth Center for Health Literacy, believes that the July turnover is just a large-scale example of those daily staff turnovers, which are also fraught with potential errors. At every shift change, doctors and nurses exchange information about each of their patients. These conversations are critical to patient care: they tell the incoming doctors what to look for, what's been observed and how to proceed. As one nurse explained it, a lot of that information is instinctive and can't be communicated through a chart.
"A huge gap occurs when you bring in a whole new cohort [of caregivers]," says Stevens. "But there's also a lot of micro-segments of that that occur. Every morning, every weekend, between when one cohort leaves and another comes on. These handoffs are a big deal in safety. We've only just begun to do it systematically." (The World Health Organization's Joint Commission on Patient Safety reports that gaps in communication at shift change can "cause serious breakdowns in the continuity of care, inappropriate treatment and potential harm for the patient." In 2007, the commission recommended standardizing handover procedures worldwide.)
The one element of continuity through both yearly and daily staff changes is the nursing staff. Most medical professionals will say that the RNs and LPNS are crucial to continuity of care and training. "Doctors rely on nurses, and there are a lot of nurses who have bailed out doctors on occasion," says Ann Williamson, Director of Nursing at the University of Iowa. But, according to Williamson, that doctor-nurse relationship is growing more strained in recent years. "There is more of a competitive drive," she says. "Some of the newer doctors coming out are embarrassed to rely on a nurse."
Bhattacharya agrees there is sometimes a reluctance on the part of new interns to ask for direction even in those first few months of training. "There's a culture in many medical centers where asking for help is a sign of weakness. You don't want to show that you don't know what you're doing."
However, interns are often acutely aware of their own limitations. Dr. Gregg Meyer, of the center for quality and safety at Massachusetts General Hospital, remembers feeling nervous and out of place when he began his training. "I think of my colleagues during orientation," he says. "The No. 1 question was, 'When are you going to figure out that I don't really belong here? That I'm not good enough?' The reality is that you're never good enough."
But ultimately Meyer, like every other doctor, nurse and researcher contacted for this story, said that he would have no hesitation about going to the hospital in July or sending a family member: "The truth of it is, yes, you're dealing with doctors who are inexperienced. But oversight is higher, too. There's never a good time of year to get sick."