Hospital Death Rates

In an August 20, 2008 article, USA TODAY reported that the U.S. Centers for Medicare and Medicaid Services (CMS) made hospital death rates for heart failure, heart attacks and pneumonia available to consumers for comparison. Prior to this, death rates for these conditions, which are widely considered yardsticks of a hospital's overall performance, were closely guarded secrets.

Anyone with access to the Internet can now compare these death rates for a local hospital with those for one across town or for medical institutions nationwide:

• Compare nearby hospitals' death rates for heart failure, heart attack and pneumonia
• Compare hospitals in your ZIP code
• Additional measures of how well hospitals meet their patients' needs
• Hospitals with the highest and lowest death rates

Previous attempts to make hospital death rates more transparent have often garnered criticism from hospital industry representatives. An effort by Medicare to report hospital death rates, for instance, faltered in the early 1990s under relentless criticism that death rates do not take into consideration a hospital's mix of patients, such as how sick, old and/or poor they were.

But the new analysis gets around this by tallying mortality rates for common life-threatening conditions, rather than the hospitals' overall death rates. This has made it more difficult to argue with proponents of the CMS comparison tools who suggest that they allow hospitals to assess and improve their quality of care and gives consumers more power to influence it.

One issue with the CMS' method of analysis, however, is that it makes the vast majority of hospitals look average. Only a handful of hospitals stand out as significantly better or worse than the norm. Some critics argue that this is by design in order to deflect blowback from the hospital industry.

Leapfrog, a consortium of corporations endeavoring to lower employee health costs by improving the quality of medical care, hopes to have its own rating system by fall, 2008. The new formula factors in the hospital's patient mix and the number of deaths expected in a hospital's particular patient base. And by capturing deaths that occur within 30 days of admission, the new rating system would make it harder for a hospital to game the numbers by shipping risky cases elsewhere.

Imperfections with the new system will doubtlessly remain, but it promises to be a useful tool that will arguably improve hospital care.

Related articles:

• Preventable medical errors (never events)
• Hospital infections (nosocomial infections)
• Emergency room errors