Reducing readmissions: what might it take?

Introduction

Why do some patients discharged from a hospital come back whereas others do not? This is a question that has been vexing clinical and policy leaders over the past 5 years. Spurred by the Hospital Readmissions Reduction Program, hospitals have begun to focus on ensuring that patients are not readmitted soon after they are discharged. Efforts include improving discharge planning, setting up more effective follow‐up, and improving clinical quality so that patients are not readmitted for complications of care. These efforts have worked to an extent. Five years after public reporting of readmission rates went into effect, we have started to see the frequency with which patients are readmitted decrease. The latest data suggest that readmission rates have fallen by 1 percentage point, from approximately 19% of all discharges to 17.8% of all discharges in the Medicare population.[1]

Although this is an improvement, we are not nearly done. There is emerging consensus that some drivers of readmissions are accountable to the hospital providing the initial acute care, but a large component of readmissions may be driven by patient socioeconomic factors or community‐level factors. It is in that context that the study by Brown and colleagues published in this issue of JAHA provides important empirical insight. Using national Medicare data, they find that hospitals with more medical admissions, more physicians, and greater intensity of health care utilization had higher readmissions.[2] What do these insights tell us?

First and foremost, we know that admissions and readmissions are linked. Readmissions are, at their simplest form, an admission for a patient who was discharged within the last 30 days. I trained in San Francisco. When I moved to Boston, I was surprised to find that we would admit people who would have been treated as outpatients in the Bay Area. However, it was unclear which place provided better care—it was simply only different care. Some hospitals have a lower threshold for admitting someone and that translates into both lower admission and readmission rates, a relationship that has been consistently born out in Medicare literature.[3]

Second, whereas the notion that having a higher per capita rate of primary care physicians is associated with higher readmission rates will strike many casual observers as contradictory, it is quite consistent with the broader literature. Primary care is supposed to prevent these readmissions. However, empirical data suggest that our mental model for why people are readmitted may need updating. In many instances, having more primary care physicians—and, indeed, more primary care visits—may simply increase the chances that someone will pay close attention and ensure that people who are ill get additional care.[4] It is unsurprising that greater clinical attention—and greater clinical intensity—is associated with a higher likelihood of returning to the hospital.

All of this gets to the broader point about readmissions and what they represent. They are, ultimately, a narrow way to gauge the quality and costs of care for patients discharged from the hospital. Some patients, because of poor hospital care or poor transitions, may die or spend weeks in rehabilitation hospitals or nursing homes. None of those count as a readmission and therefore get little scrutiny. However, the patient who returns for a short readmission 4 weeks after discharge is seen as a clinical failure. These are the consequences of a narrowly tailored measure. Though the readmission rate as a utilization measure occasionally signals quality failures, we can and need to do better. One approach is to look at the totality of care that occurs in the days to weeks after discharge. We know that the risk of readmission remains elevated out to nearly 90 days after the patient leaves the hospital.[5] A better way to gauge and pay hospitals might be to create a longer bundle—ask hospitals to be accountable for a broader swath of services (not just readmissions) and build strong quality metrics that ensure that we are not simply denying access to critical services as a way to save money.

Our nation has embarked on a series of important new reforms that seek to change the way we deliver and pay for care. This is a good thing—the old way of doing things was not working well financially or for our patients. The article by Brown et al. is an important reminder that the story of readmissions is a complex one—affected not just by discharge planning, but also by the broader milieu of care that exists outside the hospital. If we want hospitals to be responsible for improving care and reducing costs, we will have to think beyond admissions and readmission.