Wide gap seen between
best and worst hospitals

Patients 52 Percent Less Likely to Die by Choosing Best Hospitals

October 15, 2009
by Brendon Nafziger, DOTmed News Associate Editor
Choosing the right hospital could lower your risk for death by 52 percent, according to a report released this week.

For its 12th year, hospital-ranking service HealthGrades handed out stars to the country's 5,000 hospitals -- with one star meaning the worst, and five the best -- by comparing risk-adjusted death and complication rates across 27 procedures and diagnoses, such as bypass surgery or stroke. They further broke down the results into 17 items to compare in-hospital and even regional differences in death and complication rates.

HealthGrades found that while your chance of dying from the procedures studied at a 5-star hospital was 52 percent lower than at a typical U.S. hospital, it was 72 percent lower than at a 1-star hospital. For complications from the orthopedic procedures looked at, the gap is even greater: you're 80 percent less likely to have in-hospital problems if you choose a top-ranked department in a good health center.

To do the study, HealthGrades trawled through almost 40 million Medicare records reaching back over three years. The analysts looked at so-called risk-adjusted mortality rates, which means they compared the predicted death rate for patients, taking into account age, co-morbidities and other factors, with the actual death rate.

"What we're seeing is that the chasm in quality between top performing hospitals and all others is not changing," Scott Shapiro, a spokesman for HealthGrades, tells DOTmed News.

In fact, the gap is getting worse: while all U.S. hospitals saw an 11 percent reduction in risk-adjusted mortality rates from 2006 to 2008, the best hospitals saw a 20 percent more rapid reduction than the worst-performing ones.

And behind these numbers are lives. HealthGrades claims that if the 1-star hospitals were as good as 5-star ones, over 200,000 people could potentially have been saved over this period.

Intriguingly, hospital quality varies geographically. Midwestern states, such as Illinois, Ohio and Michigan, earned the highest grades, while South Central states, such as Mississippi and Alabama, fared the worst. Mountain states showed the most improvement over the past few years, along with Texas.

Reasons for geographical variation are murky, and Shapiro says the study was not designed to look at what causes performance differences between the hospitals.

However, one aspect left out of the study could, potentially, provide a clue: Shapiro says HealthGrades does not take into account the socioeconomic status of patients. "It's not part of the risk-adjustment," he says, and though he thinks the risk-adjustment procedure probably controls for any factors poverty brings, he says it could be an issue "to the extent that that impacts health and is not captured by co-morbidity a patient presents with."

As for improving low-performing hospitals, Shapiro says it's a lot of work.

"Top performing hospitals don't get there by accident," he says. "They have a culture of quality that starts with the administration, with investments, with hiring the right staff, and putting in processes that not only monitor quality, but provide foundations for continuously improving it."