DaVita® Medical Insights

CMS Five-Star Quality Ratings: Correlation with Reduced Hospitalizations and Mortality

While there have been improvements in mortality over the past few decades, mortality and morbidity remain high among end stage renal disease (ESRD) patients. A new quality paradigm, the DaVita Patient-focused Quality Pyramid, addressed the lack of a unified framework to encourage an approach to improve the lives of patients with kidney disease. To achieve the ultimate goal to improve the lives of patients with kidney disease––the top of the pyramid––it is necessary to improve survival, decrease hospitalizations and optimize the patient’s experience with care.

These outcomes are evaluated by the CMS Five-Star Quality Rating System, which implemented a change in its evaluation methodology in calendar year (CY) 2015 in a way that allows centers to improve their star rating regardless of their performance relative to the industry. Under this new approach, 41 percent of centers were rated four or five stars, which is 11 percent more than allowed by the previous methodology. The increase in the proportion of four- and five-star centers across the industry would suggest that the patient is the real winner, assuming these ratings accurately capture excellence in the outcomes on the highest tier of the pyramid.

Given the focus on improving the quality of life for patients with advanced kidney disease, this article will focus on two measures that contribute to quality of life in the Five-Star Quality Rating System—standardized hospitalization ratio (SHR) and standardized mortality ratio (SMR)—to better understand how they are associated with five-star ratings. This article also will assess a measure under consideration for the Five-Star Quality Rating System—the Consumer Assessment of Healthcare Providers and Systems (CAHPS)—designed to measure the experience of patients receiving in-center hemodialysis care, although currently only available for approximately half of the industry.

The association between mortality, hospitalizations and five-star ratings

The relationship between star ratings, SMR and SHR was assessed to understand if mortality and hospitalization were lower at centers with higher star ratings. In theory, we would expect to see better performance in the SHR and SMR and a better reported patient experience associated with higher star ratings. Despite the notable contribution of the SHR and SMR to the star rating score, high performance in the other measures in the Five-Star Quality Rating System could compensate for poorer performance in the SHR and SMR, allowing a center to achieve an average or above-average star rating (three, four or five stars).

Previous research showed that in CY 2013 and CY 2014, higher star ratings were associated with lower SHR, SMR and Standardized Readmission Ratio (SRR) (p-trend<0.001 for each). Additionally, centers that improved their star rating from CY 2013 to CY 2014 had a dose dependent association of change in SHR, SMR and SRR. Since the publication of this abstract, CY 2015 star rating and CAHPS data were released. Given the recent methodology change, we sought to understand whether this association would remain in CY 2015.

Using data obtained from the CMS Dialysis Facility Compare website, 6,032 centers received a star rating in CY 2015. Of these centers, 6,020 had values for SHR, 5,954 for SMR and 3,349 for CAHPS. Across the industry, the mean SHR was 0.996 and SMR was 1.006, meaning that on average there were as many hospitalizations and as much mortality as expected. However, centers with higher star ratings performed better in the SHR and SMR than those with lower ratings (see Figures 1 & 2). In fact, the median hospitalization and mortality ratios decreased as star ratings increased from one to five-stars, such that for every higher star rating, the hospitalization and mortality ratio was lower (p<0.001).

The association between patient experience and five-star ratings

Turning to patient experience, CAHPS performance is a composite of six measures, which were averaged to obtain an overall score. CY 2015 data suggest that median scores in this overall CAHPS score increase as star ratings increase (see Figure 3). However, while centers assigned a rating of five stars have higher CAHPS scores than one-star centers (p<0.001), there is substantial overlap in the performance ranges at each level of star rating. For example, while the median CAHPS score is higher at a five-star center than a one-star center, many one-star centers had the same CAHPS score as a five-star center. As such, further analyses are required to understand how the patient experience, as defined by CAHPS, may or may not be related to patient quality of life or other clinical outcomes.


To achieve the goal of improving the lives of patients with kidney disease, it is necessary to improve survival, decrease hospitalizations, and optimize the patient’s experience with care. These measures are on the highest tier of the Patient-Focused Quality Pyramid. The analysis outlined here indicates that, in CY 2015, there are more four- and five-star centers across the industry, and, among these centers, there are fewer hospitalizations and lower mortality. Additionally, analysis of the CAHPS results for these centers suggests that patients may have a more positive experience at higher-rated centers, but further analysis of this correlation is needed. Therefore, the patient may be the real winner, and the CMS Five-Star Quality Rating System appears to be effective at communicating the differences in the quality of care among centers across the industry.

Note: Some of this content has been repurposed, with permission, from Nephrology News & Issues.

Andrea Besharat, MPH

Andrea Besharat, MPH

Andrea Besharat joined DaVita Inc. in 2016 and leads the analytics for CMS’ ESRD Quality Incentive Program (QIP) as well as the Five-Star Quality Rating System. Previously, she worked in research and focused on two domains: the study of prescription drug misuse, abuse and diversion, and inherited cancer syndromes. Her written works have been published extensively in both of these domains. She received her MPH in epidemiology from the University of South Florida.