Pennsylvania's healthcare data council last week released what experts believe to be the most comprehensive statewide analysis of heart attack treatment ever produced.
"It is the only document of its kind in the world at the moment," said David B. Nash, M.D., director of health policy and clinical outcomes at Thomas Jefferson University in Philadelphia.
Nash, who chaired the Pennsylvania Health Care Cost Containment Council's technical advisory group, said the three-volume report-covering the western, southeastern, and the central and northeastern regions of the state-is more detailed in scope and content than comparable initiatives in California, Michigan and Minnesota. It's also more sweeping than previous studies on coronary artery bypass graft surgery published by Pennsylvania's quasi-independent cost-containment council.
"Focus on Heart Attack" is based on 39,259 hospital admissions in 1993-the latest available data when the initiative began 18 months ago, said John Malcolm, M.D., the council's physician representative and a member of the board of trustees of the Pennsylvania Medical Society.
The report provides risk-adjusted mortality statistics and charges for heart attack treatment provided by 189 hospitals and 5,033 physicians.
With 93.7% of hospitals and 98.2% of physician group practices providing care resulting in mortality at or below expected rates, providers "are doing a good job treating heart attack patients," the council said. However, on closer inspection, the analysis sends up a few red flags.
For example, the report clearly shows variations in hospitalization and mortality rates by county and community. Malcolm pointed to Northumberland County, where hospital admissions are 34% higher than expected. While the report doesn't attempt to explain the difference, Malcolm said the depressed economy, older population and health status of former coal-mining residents may be factors.
The report shows differences in patient outcome by payer. In the western region, Medicare patients' mortality rates were within the expected range, but death rates for Medicaid patients were higher than expected, and rates of death among HMO and PPO patients fell below the expected range.
The payer data also show regional differences. While HMOs and PPOs in western, central and northeastern Pennsylvania had significantly lower-than-expected rates of mortality, HMOs and PPOs in the southeastern part of the state significantly exceeded the expected range of mortality. Referring to the discrepancy, Nash said, "I'd like to understand that more."
He hopes the report will be used by policymakers as a starting point for addressing unexplained clinical variations; by providers as a benchmark for clinical performance; and by purchasers as a tool to become "savvier shoppers."