A Preliminary Analysis of Beneficiary Discharge Status and Post-Hospital Placement Before and After the Implementation of Medicare's Prospective Payment System

In recognition of the inherently inflationary nature of retrospective reimbursement, the Reagan Administration enacted legislation that substantially changed Medicare's hospital reimbursement system. The Prospective Payment System (PPS) mandated paying hospitals a fixed payment, set in advance, based on the patient's diagnosis rather than retrospectively paying for all services delivered to a patient. Critics contend that PPS introduces incentives for hospitals to conserve resources during the hospital stay and to shift care to less costly settings, both potentially affecting quality of care to the elderly. The question addressed by this dissertation is whether there were changes in the discharge health status and post-hospital placement of Medicare beneficiaries as a result of the implementation of PPS. Using a quasi-experimental time-series PRE/POST design, data was collected from the medical records of 2,619 Medicare beneficiaries (1,258 in the PRE-PPS period; 1,361 in the POST-PPS period) hospitalized between 1981 and 1986. Two large (300+ beds) and two medium-sized (100-300 beds) hospitals, representative of hospitals in the Portland metropolitan area, served as data collection sites. Medical records were selected from five Diagnosis Related Groups (DRGs): three medical DRGs (stroke, heart failure, and pneumonia) and two surgical DRGs (hip replacement and major joint pinning). Analysis of the data show that overall length of stay declined from 11.3 days in the PRE-PPS period to 8.6 days in the POST-PPS period, a reduction of 2.7 days and significant at the p = $<$.001 level and a significant increase in Dependency between the PRE and POST periods for four of the five DRGs studies (Stroke, Pneumonia, Heart Failure, and Hip Replacement). Finally, an analysis of differences in post-hospital placements shows a significant increase in POST-PPS placements to home alone (p = $<$.05), home health (p =.01), and for hospital transfers (p = $<$.001). Though limited in its generalizability, the data presented in this dissertation support the contention that Medicare patients are leaving the hospital sooner, in more dependent states of health than before PPS, and that greater numbers of potentially high care patients are being discharged to home and to home health.