HOSPITAL CARDIAC DEATH RATE HIGH; GRANADA HILLS, OTHERS STUDIED.

Byline: Paul Hefner Daily News Sacramento Bureau

Heart attack patients treated at Granada Hills Community Hospital were more likely to die than those at most other facilities in California, according to a statewide study to be made public today.

The study, based on patients treated between 1991 and 1993, provides a hospital-by-hospital report card on the survival of heart attack patients, which public health officials said can be a bellwether for the quality of care.

The study contained good news: The overall death rate from heart attacks in California has dropped steadily to a low of 13.9 percent in 1993, the most recent year studied.

But the study also singled out 31 hospitals, including Granada Hills and nine others in Los Angeles County, as scoring poorly, with higher-than-expected mortality rates.

Eight others in Los Angeles and Ventura counties scored well, with lower-than-expected rates, according to the study conducted for the Office of Statewide Health Planning and Development.

``It demonstrates that not all hospitals are average,'' said Harold Luft, director of the Institute for Health Policy Studies at the University of California, San Francisco. ``Some are really doing better than you would expect and some are doing worse.''

Hospital officials criticized the findings, which were based on patients treated between 1991 and 1993, as out of date.

``In medical care, that's almost ancient history,'' said Karen White, a Granada Hills hospital vice president.

Researchers acknowledged that their findings were based on dated records, but said the time was needed to obtain and analyze data from hospitals. A follow-up report including findings through 1996 should be out late next year, officials said.

What hospital officials and state regulators agree on is that the study should not be used alone in the selecting a hospital.

Cardiac treatment varies

California hospitals admit 40,000 heart attack patients a year, more than 5,000 of whom ultimately die within 30 days of admission. Altogether, the study found that 333 hospitals experienced death rates close to the state average and within an acceptable range, but 31 showed a significantly higher mortality rate.

Of 219 Granada Hills patients included in the study, 47 - more than one in five - were dead within a month of admission. Those are seven to 13 more deaths than researchers expected over a three-year period.

Granada Hills hospital officials said they have taken steps to reverse the trend. The hospital now participates in a number of programs designed to improve the care of cardiac patients, White said.

For example, the hospital now tracks the time between when a heart patient comes through the door to when anti-clotting agents are administered - considered a key factor in patient survival.

Other poor results

The Granada Hills hospital was the only institution in the San Fernando and neighboring valleys where researchers said they found significantly worse-than-expected death rates.

Higher than expected rates were found at nine other hospitals in Los Angeles County: Alhambra Hospital, Bay Harbor Hospital, Garfield Medical Center, Greater El Monte Community Hospital, Huntington East Valley Hospital, Memorial Hospital of Gardena, Pomona Valley Hospital Medical Center, Presbyterian Intercommunity Hospital and San Dimas Community Hospital.

No Ventura County hospitals posted significantly worse-than-expected death rates.

Eight facilities in the two counties scored well by posting exceptionally low mortality rates, including Columbia Los Robles Hospital and Glendale Adventist, Cedars-Sinai and University of California, Los Angeles medical centers.

Aggressive care works

State researchers did not attempt to isolate what caused the differences among specific hospitals, but separate research suggested that hospitals where death rates were low were quicker to provide a number of treatments, from bypass surgery to aspirin.

``In general, low-mortality hospitals treat heart attacks more aggressively than high-mortality hospitals,'' the study found.

That appeared to be the case at Glendale Adventist, where cardiologists have carried out for years what now is standard medical practice - tracking patients from the time of admission to treatment.

``With heart attacks, as the clock ticks, time is muscle,'' said Jeanette Abundis, the medical center's administrator-director of critical care, emergency and cardiology. ``We take that seriously.''

But she added that the patient - not the hospital - is most often to blame for delays in treatment.

``The classic sign is chest pain. The classic reaction is denial,'' Abundis said. ``People have reported waiting days before getting help.''

How it was done

In conducting the study, researchers say they have tried to account for the wide range of factors that affect the odds of any particular patient dying.

Some hospitals, on average, tend to admit sicker patients as a result of their location or other reasons. Without accounting for those differences, those hospitals would show a higher death rate than others, regardless of the quality of care.

To avoid that, researchers developed two formulas, factoring in the age of patients, the type of heart attack and whether the patient already had another chronic illness, such as cancer or diabetes.

Researchers used the two formulas to project ``risk-adjusted death rates'' - in laymen's terms, the percentage of patients who would have died at each hospital if every facility treated a hypothetical average patient.

In the case of Granada Hills, researchers calculated risk-adjusted death rates of 20.2 percent and 17.1 percent. In both cases, the rate was above the statewide average of 14.5 percent for the three years, though only one formula found the difference to be statistically significant.

The study found that some hospitals experienced slightly higher than expected rates, but they were not listed as such because the margin was too small to be statistically significant.

Granada Hills hospital's White said she wasn't on the staff at the 201-bed facility at the time the patients involved were treated, so she can't be sure what led to the relatively high percentage of deaths.

One reason contributing to Granada Hills' death rate, White noted, is that 10 of those patients had ``do not resuscitate'' orders in place. In addition, some patients died days or weeks after leaving the hospital, she said.

Researchers said they do not believe that patients with ``do not resuscitate'' orders should have been excluded from their research. ``DNR patients are not predestined to die,'' the study noted.

Caveats

White cautioned the public against using the study's results to draw any conclusions about the quality of care now offered at the hospital.

``I would say that overall I'm disappointed that we're trying to deal with information that is this old,'' she said. ``Unfortunately, we don't have a mechanism to be able to get the data out in a timely manner.''

Abundis agreed that the age of the cases studied gives the public little ability to judge the current quality of care at hospitals. She advised patients to talk with their doctors about the services available at local hospitals.

And she said that even patients who arrive at a hospital emergency room with chest pain shouldn't be shy about asking what kind of treatment they can expect - and consider going elsewhere if care might be delayed.

``They should ask, do you have cardiologist that can treat me now?'' Abundis said.

CAPTION(S):

Chart

Chart: (Color) HEART ATTACK REPORT CARD

Researchers took into account the age, severity of illness and prior treatment of patients to arrive at a ``risk-adjusted death rate'' for each hospital. Comparing the rate against the state average, they determined whether each facility's rate was significantly better, worse or as expected.

SOURCE: Statewide Health Planning and Development

Dionisio Munoz/Daily News