Hospital readmissions and the unfortunate predictability of it all

Readmissions to hospital, especially within 30 days of discharge, are on the radar of all health care organizations in the United States. This is due in no small part to stiff financial penalties that are now imposed on the worst performers by the federal government. On the surface, it may seem like a reasonable thing to penalize hospitals that don't successfully “get their patients better enough” to avoid readmission. However, the problem is highly complicated, and such a crude way of thinking about the problem (i.e. blaming it all on the hospital and the doctors) obscures from many of the real issues behind high readmission rates. These include factors such as an ageing population, lack of social support for many elders, the rising incidence of chronic incurable morbidities—all against the sad backdrop of the inevitability of sickness in old age. Battling nature can indeed be hard.

Most statistics suggest that approximately 1 in 5 Medicare patients (20%) are readmitted within 30 days. As shocking as this number may sound, it may even be an underestimate, because it doesn't include many patients under the age of 65 who are also stuck in the revolving door of frequent readmissions.

One of the key areas of focus has got to be better community care and follow-up, but how do we go about identifying these high-risk patients in the first place? Frontline hospital physicians, including myself, are all too familiar with the fact that a huge proportion of patients we admit to the hospital have been discharged in the not too distant past, sometimes as soon as the day before. It's the first thing we see when we scroll through the records of our new admission—what we call a “bounce back”. Equally, most physicians (and for that matter nurses and case managers) will instinctively know as soon as they discharge a patient, who is likely to be back very soon.

Over the last several years, there have been lots of clinical tools developed by academics to try to predict which patients are at the highest risk for readmission. Some in the “health care innovation” world have also tried to get computers and information technology in on the act. But at the end of the day, one can develop clinical tools till the cows come home, but rarely does it outstrip the good common sense of the people working at the frontlines.

Patients at a higher risk of readmission can easily be predicted by their recent history, clinical state, and social situation. The world of healthcare should identify these patients early on and immediately plough resources into working out what can be done upon discharge to keep these patients healthy in the community. There also needs to be better communication between the hospital and primary care teams, the patient's family, and also all of the specialists that may be involved in follow-up. And although we are focused on 30 day readmissions, because that's what the system uses as a measure, frequent admissions even within 60 or 90-day periods are also just as much of a problem and shouldn't be ignored. By zeroing in on these high-risk patients and keeping them healthy and out of the hospital, we are being the best doctors possible.

Suneel Dhand, MD, ACP Member, is a practicing physician in Massachusetts. He has published numerous articles in clinical medicine, covering a wide range of specialty areas including; pulmonology, cardiology, endocrinology, hematology, and infectious disease. He has also authored chapters in the prestigious "5-Minute Clinical Consult" medical textbook. His other clinical interests include quality improvement, hospital safety, hospital utilization, and the use of technology in health care. This post originally appeared at his blog.