U.S. News disputes Health Affairs study

U.S. News & World Report disputes a study published last week in Health Affairs, which found that hospital reputations and prices did not correlate with outcomes.

The study, led by Chapin White of the RAND Corporation, noted that expensive hospitals were more likely to be included in U.S. News' hospital rankings, despite the fact that cheaper hospitals outperformed the more expensive ones in 30-day readmissions and preventing blood clots or death among surgical patients.

In a letter to the editor of Health Affairs, U.S. News'  Ben Harder took issue with White and his team's claim that the publication's rankings are "largely based on reputation." Specifically, Harder writes, the journal uses whether or not a hospital appears in U.S. News' 50 Best Hospitals in cancer, cardiology, orthopedics or gynecology as an indicator of reputation. Health Affairs justifies this based on a 2010 study, but the study, Harder argues, does not come to the conclusion Health Affairs claims.

"While [the study] concluded that reputation contributed to 'the relative standings of the top 50 hospitals'--in other words, which hospital is No. 1 versus No. 2," Harder writes, there was no indication that hospital reputation contributed substantially to a hospital's ranking.

"In fact, U.S. News uses objective, primarily Medicare-derived quality measures such as mortality, volume and patient safety indicators to calculate 67.5 percent of each hospital's rank-determining score in the four specialties analyzed," Harder wrote. "Reputation accounts for only 32.5 percent and will soon be reduced to 27.5 percent. Many hospitals with negligible or no reputation are ranked among the top 50."

Furthermore, Harder writes, U.S. News has already refined its methodology to modify the significance of reputation, and researchers have yet to reproduce the 2010 study using the revised methodology. Harder also expressed concerns about other potential issues with the Health Affairs report. For example, he wrote, it "compares Medicare-derived outcomes to claims data from a non-Medicare population," which may not make for an exact comparison.

To learn more:
- read the full response