Association for Public Policy Analysis & Management

The Medicare Bundled Payment for Care Improvement (BPCI) program is a demonstration program with the stated goals of supporting better care coordination, improving clinical quality and reducing total episode cost. Previous experiments with similar payment models suggest that hospitals will reduce inpatient stays and shift patients to lower cost sites of post-acute care to succeed in this program. Roughly 25% of Medicare beneficiaries live in rural areas. Hospitals serving rural patients may face unique challenges in responding to this demonstration. As Medicare expands bundled payments for joint replacement surgery it is critical to understand the impact of the program across all types of hospitals.

This is a mixed methods case study of a single site participating in the Medicare BPCI program for total joint replacement surgery. Twenty-nine interviews with clinical staff and administrative leaders were conducted at two points in time. Interviews explored the organization’s bundled payment program response and staff's perceived barriers and facilitators to success. Key patient and hospital level outcomes of interest include: hospital length of stay, discharge destination, patient satisfaction scores and 30-day readmission rates. Outcomes were assessed via pre-post tests and interrupted time series analysis. Discharge destination, readmission rates and length of stay were internally collected by the hospital. Patient satisfaction scores were measured via HCAPHS scores. Patient level data was collected on patients undergoing joint replacement surgery at the study site between 2012 - 2015.

Hospital length of stay decreased significantly after the implementation of the bundled payment program (3.29 pre to 3.01 post, t: 8.51 p<0.05), though the interrupted time series analysis suggests this change may not be associated with participation in the bundled payment demonstration. The rate of patients discharged home remained stable (70%), as did the monthly 30-day readmission rate (2.5%). Qualitative findings reveal a number of important considerations. Patient education and engagement prior to surgery supported shorter length of stay and discharge home. Engagement of clinical staff in the bundled payment model was a barrier to success. Surgeons remained hesitant to shorten length of stay or shift discharge to home rather than a nursing facility particularly when patients were going home to a rural area with limited health care resources, and inpatient nursing staff reported similar challenges.

While hospital length of stay declined during the study period, it is unclear if this was the result of participation in the bundled payment program alone, or the result of improvement initiatives that predated the payment change. Broader success under the bundled payment program will require a higher level of engagement from the clinical staff, as well as resolution of the challenge of discharging patients home to rural locations with limited health care resources.

Rural residents comprise a substantial proportion (almost 25%) of Medicare beneficiaries. As Medicare expands bundled payments, it is critical to account for rural residential status of patients as a challenge for reducing length of stay and shifting discharge to home.