In a previous post, I examined the four main sources of publicly reported hospital quality data on the web, and discussed the main limitations of these sites. HealthGrades, one of the four sites, have recently published a study comparing quality outcomes in the nation’s best and worst hospitals and concluded that wide variations still exist.

Similar to the system used in grading hotels, HealthGrades has developed a star-rating system for hospitals based on common patient outcomes such as mortality and complication rates. Five-star hospitals have the “best” performance, three-star hospitals have “as expected” performance, and one-star hospitals have “poor” performance.

Comparing these hospitals on 17 different mortality-based procedures and diagnoses (such as bowel obstruction, heart failure, pneumonia, respiratory failure and coronary bypass surgery) and complication-based procedures (back and neck surgery, total knee and total hip replacement for example), the study concluded the following:

  • Between 2006 and 2008, mortality rates at 5-star hospitals improved at a more rapid rate (11.89%) than a 1- or 3-star hospitals (10.14% and 10.72%, respectively)
  • Patients had a 71.64% lower chance of dying in a 5-star hospital compared to a 1-star hospital (on all 17 procedures and diagnoses).
  • 224,537 Medicare lives could have been saved in two years if all hospitals performed at the 5-star level.
  • Patients had a 79.69% lower chance of experiencing a complication during an orthopedic procedure in a 5-star hospital compared to a 1-star hospital.
  • 110,687 Medicare orthopedic complications could have been avoided in two years if all hospitals performed at the 5-star level.

While keeping in mind that these data may be inconsistent with other sources on other sites, patients wondering how their hospital of choice faired can visit HealthGrades.