Health & Family

Medicare posts hospital death rates

To help consumers make better health care decisions, Medicare has for the first time published hospital-by-hospital death rates for patients treated for three common conditions: heart attack, heart failure and pneumonia.

Results show most Charlotte-area hospitals fall into the national average for the percentage of Medicare patients who die within 30 days of hospital admission.

Only one – Carolinas Medical Center-Union in Monroe – ranked better than the national average, based on deaths of pneumonia patients.

Two area hospitals – Frye Regional Medical Center in Hickory and Springs Memorial Hospital in Lancaster, S.C.– ranked worse than the national average for pneumonia deaths.

Medicare's Web site,, has offered hospital comparison data since 2005. Last year it included mortality data in several disease categories for the first time. But this is the first year the agency made public specific death rates for each hospital.

Area hospital officials praised Medicare's effort to make information about medical outcomes more transparent. But they emphasized that mortality data is only one measure of quality.

“Patients are interested in whether hospitals are safe, effective, efficient and pleasing and have good service,” said Dr. Roger Ray, chief medical officer for Carolinas HealthCare System, the largest hospital group in the Carolinas. “They're interested in all those attributes. This one measure (of mortality) is a piece of that, but it's not the whole pie.”

Ray and other hospital officials said mortality statistics can be misinterpreted.

For example, they said, Medicare's death rates don't take into account that some patients and their families reject aggressive treatment.

Of 20 pneumonia deaths at Springs Memorial in the latest Medicare report, more than half were in hospice or had signed “do not resuscitate” orders, according to Janice Dabney, assistant CEO for Springs Memorial.

“There's just a lot more to the number than people often understand,” she said.

Although all hospitals are compared the same way, those with smaller numbers, such as Springs Memorial, can suffer in that type of comparison, said Dr. Stephen Wallenhaupt, chief medical officer for Winston-Salem-based Novant Health, which owns Presbyterian Healthcare in Charlotte.

He and other hospital officials pointed out that the 30-day mortality data include deaths that occur outside the hospital for whatever reason.

If a patient were treated for pneumonia and discharged, then died in an accident within 30 days, “that would show as a death from pneumonia” for the hospital, Wallenhaupt said. “It's a little bit confusing.”

All three Presbyterian hospitals in Charlotte, Matthews and Huntersville ranked among average for mortality, but “I'm not sure (knowing that) is very useful to the public at this point,” he said.

Medicare's Web site also provides other comparisons, such as whether hospitals followed proper procedures in treating patients with certain conditions. It shows, for example, whether patients with heart attacks were given orders for aspirin on arrival and discharge and whether pneumonia patients received the correct antibiotics as well as pneumonia and influenza vaccines.

This “process” information could be more telling than the mortality data, some hospital officials said.

Despite its “worse than average” showing on pneumonia deaths, Springs Memorial scored better than the national average on several of those treatment measures. For example, 90percent of patients received pneumonia vaccines, compared to 77percent nationally.

Officials at Frye Regional declined to be interviewed, but sent a prepared statement making some of the same points.

“Pneumonia mortality rates do not reflect the complete evidenced-based measures related to pneumonia care,” the statement said.

Many who have used the Medicare Web site, including employees in the Medicare communications office, find it's not exactly user-friendly.

“It's getting a lot better,” said Christine Van Dusen, a quality measures specialist with Premier, a Charlotte-based alliance of 2,000 U.S. hospitals.

“They're making an effort. But you don't want to look at just mortality rates to make any determination of the quality of care that a hospital is providing. That's one part of it. But you have to look at all of it.”