Abstract

The Prospective Payment System (PPS) introduced for Medicare patients in the USA in 2000 created strong financial incentives to reduce service provision to home health care clients. This study focuses on the initial impact of PPS on patient outcomes, given this reduction in service provision. Overall patient outcomes changed very little during introduction of PPS, and the rate of hospitalizations and emergent care needs also declined. This relatively unchanged effect on patient outcomes, combined with reductions in hospitalization and emergent care, suggest that the PPS may improve overall system efficiency. Further analysis is needed to understand the long-term impact on all patients. Background: A shift from a traditional fee-for service system to a Prospective Payment System (PPS) took place between 1998 and 2000 to cover health care costs of Medicare patients receiving home health care in the United States. The purpose of this new payment system was to constrain health care costs incurred by government by awarding a single payment to each health care agency managing a client at the beginning of the care episode. The level of payment was adjusted to account for the severity of the patient. As a result, the PPS created strong financial incentives to reduce service provision to clients, in the form of reduced visits. Authors hypothesized that this reduced service provision could result in worse patient outcomes, and this study focuses on the initial impact of PPS on patient outcomes.

Methods: National outcomes databases and Medicare claims data were used to analyze more than 164,810 health episodes of Medicare beneficiaries aged over 65 years receiving home health care between 1999 and 2001. Patient health outcomes were defined as a change in health status between the start of the care episode and discharge, capturing a range of functional, physiological, cognitive and behavioural states. Authors controlled for patient case mix between providers. Total home health visits were tracked for each patient, and were further divided into skilled nursing, therapy (e.g., physical, occupational, speech, medical social services) and aide (e.g., bathing, dressing, eating, etc) visits.

Findings: During the initial PPS period, the overall composition of this group of patients shifted toward being more functionally dependent and clinically complex. This may reflect more accurate reporting by agencies, or 'gaming' practices, because agencies now had strong financial incentives to demonstrate that each patient required more complex care. Total visits per episode (patient care period) declined 16.6% upon introduction of PPS. Greatest service reduction occurred for aide visits, while therapy visits increased 8.4%. This is consistent with expectations, given that more complex patients receive higher payments. Overall patient outcomes changed very little during introduction of PPS. Despite the drastic reduction in aide services, few changes were noted in patients' ability to manage activities of daily living. Overall, more patients' improved their ability to dress, groom, toilet, and ambulate while fewer patients improved their ability to transfer, eat, and use the telephone. Many physiologic outcomes remained unchanged although urinary incontinence improved in fewer patients. Under the PPS, the rate of improvement of cognitive abilities of patients, including cognitive functioning, confusion frequency, anxiety level and behavioural problem frequency, was lower. Finally, hospitalizations decreased, discharges to the community increased, and requirements for emergent care also decreased. Although many outcomes improved overall, the analysis also showed that certain subgroups of patients were negatively affected. Future studies need to look at outcomes in chronically ill patients.

Conclusions: The relatively unchanged effect on patient outcomes, combined with reductions in hospitalization and emergent care, suggest that the implementation of the new payment system has improved efficiency because similar outcomes are achieved at a lower cost. However, the results are based only on the first year of PPS, and longer-term analysis is necessary. For instance, the increasing self-reliance of patients in activities of daily living (ADLs) and instrumental activities of daily living (IADLs) imposed by the decline in aide visits may have long-term impacts on overall health that would not have been captured in this initial period. Possible applications to Canada involve moving away from a fee-for-service system towards capitation for home health care patients.

Reference: Schlenker, RE, Powell, MC, Goodrich, GK. "Initial Home Health Outcomes Under Prospective Payment." Health Services Research, 2005, 40(1): 177-193.