A 5-star rating system for nursing homes and the unintended consequences on health care disparities

Information about the quality and performance of health care facilities can be confusing to consumers. Dozens of government organizations, trade groups and websites rate doctors, hospitals and long-term care facilities on all kinds of scales, from patient satisfaction to medical outcomes.

In 2008, the Centers for Medicare and Medicaid Services (CMS) attempted to simplify some of this data by creating a five-star rating system for nursing homes. The idea was that public reporting would drive improvement in care, helping nursing home residents and their families choose higher quality facilities, in turn encouraging nursing homes to improve quality to retain residents.

This data can be of limited use, however, for people whose decisions are constrained by insurance networks, cost and geography. People who are enrolled in both Medicare and Medicaid, often called “dual eligibles,” are particularly limited in their choices for long-term care. They are much more likely to have lower incomes, disabilities or cognitive impairment, and to receive low-quality health care in poor neighborhoods than other Medicare beneficiaries.

Tamara Konetzka, PhD

A new study in the May issue of Health Affairs by public health researchers from the University of Chicago, Harvard, and Penn confirms that despite best intentions, the new rating system exacerbated health disparities between this dual eligible group and non-dual eligible nursing home residents, i.e. those with better financial support. By 2010, two years after the system began, both groups lived in higher quality nursing homes overall, but non-dual eligible residents were more likely to actively choose a higher-rated nursing home. The gap between the two groups also increased: dual eligibles were still more likely to live in a one-star home, and less likely than non-dual eligibles to live in a top-rated home.

Tamara Konetzka, PhD, associate professor in the Department of Public Health Sciences at the University of Chicago and lead author of the study, said these gaps aren’t necessarily caused by the rating system, but rather reflect longstanding issues with health care disparities.

“It’s not the fault of the 5-star system that we’re seeing this problem with dual eligibles,” she said. “Instead it’s the very intractable problem of disparities in health care due to where people live. It isn’t caused by this ratings system, but the rating system doesn’t do much to solve it either. Knowing which nursing homes offer the highest quality doesn’t help if there are no high-quality homes in your neighborhood. ”

Data from the study illustrate how the two groups were able to respond differently to the new ratings. Non-dual eligible residents were 6.6 percent less likely to live in a one-star home by 2010; 5.4 percentage points could be attributed to their existing nursing home improving its score, while the remaining 1.2 percent was due to choosing a higher quality home. Dual eligible residents were better off overall too. They were 6.2 percent less likely to live in the lowest rated homes than in 2008, but only .3 percent of that improvement came from residents being able to move elsewhere.

But should it matter if someone ends up in a better nursing home because they moved, or because they one they were living in improved?


Geographic distribution of one-star and five-star Nursing Homes in the Chicago area. Authors’ analysis from Nursing Home Compare December 2008 rating data. Notes: Green squares are nursing homes with a 1-star rating. Dark blue squares are nursing homes with a 5-star rating. (Source: Health Affairs/Konetzka, et al)

Konetzka said it does, because many long-term care stakeholders are skeptical about the quality improvements shown by the lowest rated homes. When CMS introduced the rating system, it had three components: independent inspections, staffing levels and data on clinical outcomes. The last two components were self-reported, which left open the possibility for manipulation or simple changes in documentation instead of true quality improvement.

For instance, Konetzka and her colleagues worked with one nursing home that scored badly on pain levels reported by residents. To fix this, they began asking residents about their pain levels after they administered medications for the day instead of first thing in the morning. It didn’t change anything about the care patients received, and it wasn’t outright fraud, but it made a big impact on those pain scores.

CMS has since tweaked the ratings system to limit the impact of improvement on the self-reported domains, and Konetzka said she still believes it’s a useful tool.

“I’m a big believer in information and I think it can do a lot of good,” she said. “Public reporting is not going to solve the geography problem, but people respond to simplified information like the five-star system.

“There’s plenty of work to be done refining it, but people pay attention and react, both the facilities that care about their scores and consumers who choose nursing homes,” she said.

About Matt Wood (531 Articles)
Matt Wood is a senior science writer and manager of communications at the University of Chicago Medicine & Biological Sciences Division.
%d bloggers like this: