Methodological changes would increase the number of one-star hospitals from 112 to 284 while also increasing the number of five-star hospitals from 83 to 278, according to a contractor.


Oct. 2—Hospital advocates threw their support behind the federal government’s recent decision to pause the overall-quality star ratings for hospitals while considering changes.

The Centers for Medicare & Medicaid Services (CMS) recently announced in an email to hospitals that a planned October update to its hospital star-rating site will not occur while the agency reevaluates the methodology behind the system.

“CMS decided not to proceed with the October update to continue its examination of potential changes to the Star Rating methodology based on public feedback,” the agency stated in a Sept. 27 notice.

Biannual updates to the star rating were generally planned for July and December. After the delay of some data, CMS had planned for the mid-year update to occur in October.

Instead, star ratings released in December 2016 will remain on the ratings website until the next update, CMS said.

The decision followed ongoing concerns from hospitals that the ratings methodology was flawed and simplistic. The delay was greeted as a good sign for implementation of an overhaul.

“CMS made the right call in postponing the October update of star ratings, and we appreciate the agency's stated interest in further improving its approach,” Akin Demehin, director of policy for the American Hospital Association (AHA), said in a written statement to HFMA News. “As we have shared with CMS, the current design and execution of star ratings fails to create an accurate, representative picture of quality that patients and families can rely on to inform their healthcare decisions.”

Demehin said AHA supports transparency and will continue its “dialog” with CMS about “providing the public with accurate, meaningful information about quality.”

Understanding the Ratings

The overall star ratings were launched in July 2016 to aggregate quarterly data from existing quality measures that are publicly reported on the Hospital Compare website into a single star rating for each hospital. The website includes information on over 100 quality measures and more than 4,000 hospitals.

The rating initially incorporated 64 of the more than 100 measures, but CMS reduced that number to 57 in January.

Hospitals are assessed only on the measures for which they submit data; some hospitals may report as few as nine measures. CMS had said hospitals could improve their star ratings by improving quality across all reported measures—particularly those for outcomes and patient experience, which receive higher weighting. Hospitals cannot improve their ratings by selecting the measures to be reported, according to CMS. Ratings are based on reported measures across seven categories that meet the inclusion criteria.

Overall star ratings are based on relative performance. Hospitals that improve more than other hospitals are likely to achieve higher ratings, CMS said.

Hospitals can correct suspected errors or appeal their star rating by submitting a request along with their CMS certification number through the inpatient and outpatient question-and-answer tools.

A hospital’s star rating does not affect its payments.

Push for a Pause

The pause came two days after a Sept. 25 letter from the AHA to CMS that repeated the hospital association’s request for “CMS to suspend the reporting of overall star ratings until the methodology is improved.”

The AHA also had requested that CMS remove the star ratings from Hospital Compare and not republish them until it “corrects the errors and outside experts agree that the updated methodology is executed correctly.”

The AHA urged CMS to consider alternative approaches to an overall star rating, perhaps star ratings that cover specific topic areas such as patient safety, patient experience of care, and cardiac care.

Overhaul Efforts

The AHA’s comments came as part of a public comment period on changes that CMS was considering to its overall hospital star-ratings methodology.

Some of the proposed changes were supported by CMS’s Technical Expert Panel this summer. CMS also encouraged stakeholders to run their own analysis of the proposed methodology.

The AHA said the proposed changes addressed some “missteps” related to “errors in the execution of the chosen methodology.”

However, the hospital group noted that CMS’s own analysis showed that the changes would alter star ratings for nearly 700 hospitals, thus “amplifying our concern about the reliability and accuracy of the chosen methodology.”

Among projections compiled by Yale New Haven Health Services Corporation-Center for Outcomes (YNHHSC/CORE) Research and Evaluation, with which CMS contracted to conduct an evaluation of the star-rating system, was that the methodological changes would increase the number of one-star hospitals from 112 to 284 and the number of five-star hospitals from 83 to 278.

An analysisof the socioeconomic characteristics of hospitals’ home ZIP codes based on the number of stars they received found that one-star hospitals were in ZIP codes nearly “three rungs lower on the SES [socioeconomic status] ladder than five-star hospitals.” The analysis by Herb Kuhn, president and CEO of the Missouri Hospital Association, and colleagues was published on the New England Journal of Medicine’sCatalyst website.

The hospital groups’ concerns followed others that had been raised about the star-rating methodology. For example, Francis Crosson, MD, chairman of the Medicare Payment Advisory Commission, wrote a letter to CMS a year ago in which he stated that the rating system fails to account for factors such as the severity of conditions at a particular hospital.

“For example, at one-star hospitals an average of 78 percent of admissions were admitted through the emergency department (ED), while at five-star hospitals only 36 percent of admissions, on average, were admitted through the ED,” Crosson wrote. “One-star hospitals are treating a greater share of likely more severe cases from EDs.”

CMS asked YNHHSC/CORE to develop two parallel sets of data files for the December 2017 Hospital Compare update, with one using the current methodology and the second incorporating the proposed methodological changes. CMS said the approach would allow it to implement proposed changes as soon as December.

However, even if CMS can improve the methodology, the AHA said it would continue to have “significant concerns about the conceptual underpinnings of star ratings,” primarily because the measures included in the ratings were never intended to be used to create a single, representative score of hospital quality.


Rich Daly is a senior writer/editor in HFMA’s Washington, D.C., office. Follow Rich on Twitter: @rdalyhealthcare

Publication Date: Monday, October 02, 2017