The methodology CMS uses for hospital star ratings on the Hospital Compare website could be improved if the agency makes five key adjustments, suggests David Levine, MD, FACEP, senior vice president for advanced analytics and informatics with Vizient.

Levine worked with Rush University Medical Center in Chicago to analyze how CMS weighs patient safety measures in the star ratings. He recommends these changes:

1. Use smarter hospital groupings.

Rather than combining community hospitals, large health systems, specialty hospitals, and teaching/academic hospitals on many key metrics, CMS should group hospitals by subtype. This could mean grouping facilities that treat advanced cancers, neurosurgical cases, and cardiology cases, for instance.

2. Group primary procedures by specialty.

The current system does not allow consumers to seek the best hospital for a specific need, like cardiac care or joint replacement.

3. Use better data, and use it faster.

Data come only from Medicare patients who don’t represent the overall patient population of a hospital, and Vizient’s analysis found that 64% of an overall star rating score is based on data that are at least two years old.

4. Create data transparency.

More transparency about the data used for star ratings would help hospitals focus their quality improvement resources. Today, the ratings are more like “report cards” that provide only an overall score.

5. Revise the weighting of quality measures.

CMS uses four major criteria to calculate star ratings: patient safety, patient experience, readmission, and mortality risk. Those criteria account for 80% of the total score. Other criteria should be incorporated to provide a more accurate picture of hospitals’ overall quality.