The long-term objectives of the proposed research are (1) to make a significant contribution to the understanding of the relationship between volume (both physician-and hospital-specific) and patient outcomes, and (2) to methodologically advance the study of small area variations.
The specific aims of the research are to measure and analyze: (1) the attraction of patients to hospitals with better outcomes, (2) the relationship between provider volumes and patient outcomes, and (3) the influence of outcomes on hospital admissions rates. Using 1980-1988 hospital discharge abstract data from upstate New York zip code-specific demographic data and patient outcome data from HCFA MEDPAR files, the investigators will answer the following major research questions: 1. What is the influence of risk-adjusted outcomes on hospital market share for specific procedures after controlling for distance from patient residence to the hospital, case-mix adjusted-charges, hospital characteristics, proximity of physicians admitting to the hospital, and hospital acceptance of the poor? 2. What is the effect of physician and hospital volume and characteristics on outcomes once one controls for patient characteristics and the possibility that hospitals and physicians with better than expected outcomes attract more patients? 3. How important are local physician and hospital supply factors, adjusted hospital charges, and case mix-adjusted outcomes in explaining small area admission rates after controlling for the population's age, gender, race, income, and insurance? The investigators' extensive experience in volume-outcome and hospital market analysis provides the foundation for the proposed research. Tested methods for calculating risk adjustments, modeling conditional choice of hospitals, determining hospital-zip clusters, and refining patient-outcome measures will be applied to the New York data set.
