When I was a medical resident, I noticed that bad things kept happening to my older patients: many got confused and some fell and maybe even broke a hip. I wondered whether the problems they developed were related to the acute medical illness for which they were admitted or to the hospitalization itself. So I did a study in which I compared the experience of older patients to that of people under 70. By looking through patients’ hospital charts and sitting in on the nurses’ rounds every day, the time when they reported to the next shift what was really going on with their patients, I was able to determine who was confused, who fell, who stopped eating, and who was incontinent. Then I analyzed whether there was any conceivable relationship between their medical problems and the symptom they developed. For example, a person admitted with a stroke or meningitis (an infection of the lining of the brain) could be expected to be confused but not a patient with a stomach ulcer. What I found was that 40% of the older patients, compared to 9% of the younger ones, had one or more of these symptoms that couldn’t be explained by their admitting diagnosis. Moreover, as soon as patients had one of these problems, doctors intervened in some way—they ordered restraints for the patients who had fallen or a urinary catheter for those who were incontinent—and all those interventions in turn predisposed to new problems. In subsequent years, several other investigators documented the perils of hospitalization for older people and geriatricians introduced ACE (acute care for the elderly) units to minimize the risk of hospital-induced problems. These units have made a difference, but even in specialized units, older people are at risk of hospital-related complications. Today, there is a renewed interest in learning about the perils of hospitalization. One prominent researcher introduced the concept of “post-hospitalization syndrome,”arguing that older patients are at heightened risk of problems after discharge, problems related not only to the acute illness for which they were hospitalized, but also to the debilitating effects of having been in the hospital. Patients are often sleep-deprived, poorly nourished, and de-conditioned after a hospital stay, and it is these factors that may predispose to difficulties in the 30 days after discharge. According to this analysis, physicians and nurses need to pay more attention to making the hospital a better and safer place for patients. Now a new study picks up on the theme of the post-hospitalization syndrome, measuring the risk of adverse drug reactions during this period of heightened vulnerability. Pharmacists reviewed the records of 850 older people who collectively experienced 1000 hospitalizations and they identified 330 possible adverse drug events (injury from a drug and not the underlying disease) during the 45 days after discharge. Physicians looked through the list and agreed that 242 cases were truly adverse drug events, of which 2.5% were life threatening and another 21% were serious. They deemed just about one-third of these events preventable. Most of the drugs causing these problems were cardiovascular drugs or diuretics (fluid pills that are typically also used to treat heart disease); the next major class of offenders was narcotics. The authors conclude that doctors need to do a better job in the hospital (deciding on what medications a patient should be discharged) and afterwards (monitoring for side-effects). It seems that patients still get into trouble after hospitalization, particularly frail elders, just as they did 30 years ago when I published my study of iatrogenesis. Adverse drug reactions are yet another form of trouble. But what are the implications of these observations? We should try harder to make the hospital a safer place for frail old patients. We should watch assiduously every time an older patient starts a new drug, and people who are discharged from the hospital often go home with several new medications or new doses of old medications. Maybe we should also think about whether the patient should really have been admitted to the hospital in the first place. Perhaps his illness could have been prevented. More plausibly, perhaps we could treat the illness in a way that didn’t necessitate admission to a large, alien institution like a hospital. An older person cared for at home when he develops pneumonia or a worsening of his chronic heart failure won’t suffer from confusion induced by unfamiliar surroundings. He won’t have his sleep disrupted by monitors going off in the adjacent bed or nurses and doctors talking loudly in the hall. Of course he won’t have all the benefits of acute hospital care either, the sophisticated technology, the 24-hour nursing care. But maybe the risks aren’t always worth the benefits. Maybe we should design alternatives to hospital care that feature some of the benefits of the hospital but all the benefits of home.