In a period of rapid change in heath care delivery, both managers and policy makers are sorely lacking evidence as to the impact of their actions on outcomes that can be objectively measured and valued. This project proposes to provide an important new gauge of the relationship between inpatient hospital resources and net value received, using the unique perspective of the employer/group purchaser. The large company, a portion of whose data on 350,000 employees will be studied, is both paying for direct heath care services (input side) and absorbing the indirect compensation and productivity costs of lost work time (output side). For the firm decrease in days or dollars of hospitalization for an illness may be offset by the incremental effect of a less rapid return to work. Alternatively, the effect of direct heath care cost containment action may be a faster recovery due to the avoidance of hospital induced complications. This research will provide empirical data by major DRG, among geographical areas and across different industries served by this firm, regarding the tradeoff between direct health care cost and its indirect effect on corporate value. These results also will be useful as indicators of the general effectiveness of the health system regarding other population segments. The methodology will utilize multiple regression to estimate coefficients of input (hospital days and payments) and a number of covariates (job classification, sex, age,probability of termination, and prior years utilization as an indicator of health status) and other dummy control variable (industry, location). These will be examined for significance and differences among subgroups. MANOVA also will be used to test the significance of such differences, and the importance of interaction effects. Aggregation techniques using both parametric and nonparametric transformation, will be used to control for differences across DRGs and thereby permit larger sample sizes for testing hypotheses. The end result will be specific information by critical sub areas of the system on the effectiveness of marginal heath care expenditures. Areas for attention in both policy and management will be highlighted.
