Rehabilitation Processes in Skilled Nursing Facilities in the Context of Shifting Policy

Abstract

The purpose of this dissertation is to explore relationships between Medicare policy, rehabilitation practice, and patient outcomes within the fragmented and complex Skilled Nursing Facility (SNF) industry. United by a conceptual framework that connects Medicare policy with the structures, processes, and outcomes of rehabilitation in SNFs, the articles included in this dissertation were motivated in part by the implementation of a new Medicare reimbursement model, the Patient Driven Payment Model (PDPM), in October 2019. In response to rising intensity and costs of what Medicare considered clinically unnecessary rehabilitation, PDPM drastically shifted incentives for physical and occupational therapy provision in SNFs. However, the design of the PDPM reimbursement system does not account for potential policy impacts on patient outcomes. The first article, “Is More Always Better? Financially motivated therapy and patient outcomes in Skilled Nursing Facilities,” is a pre-PDPM secondary analysis of 2018 patient data from SNFs in the United States. The purpose of the study was to examine whether a financially motivated therapy billing practice known as thresholding was associated with patient functional improvement and community discharge outcomes. Thresholding, which occurred when SNFs provided ten or fewer minutes of therapy above weekly reimbursement thresholds under the previous payment system, is a unique metric that allows for specific study of financially motivated therapy that can be examined separately from intensive therapy, which may be clinically indicated. Results of this study indicate that extra minutes of therapy received by patients who experienced thresholding were associated with small positive effects on functional improvement and community discharge rates, even when controlling for overall therapy intensity. While PDPM was designed to disincentivize both thresholding and intensive therapy overall, these results emphasize the importance of Medicare payment policy designed to promote, not disincentivize, potentially beneficial rehabilitation services for patients. Using a similar cohort and study design, the second article, “Some But Not Too Much: Multiparticipant Therapy and Positive Patient Outcomes in Skilled Nursing Facilities,” establishes pre-PDPM relationships between multiparticipant therapy and patient outcomes. Provision of multiparticipant therapy – including two or more patients per therapy provider per session– is known to have increased over 300-fold immediately after PDPM implementation, as SNFs reduced intensive individualized therapy and utilized more multiparticipant sessions to reduce therapy staffing costs. Until post-PDPM data become available, this article used 2018 data to help predict possible implications for patients experiencing drastic shifts from individualized to multiparticipant therapy in SNFs. Models compared different levels of multiparticipant therapy provision as a proportion of total therapy. Compared to patients receiving no multiparticipant therapy, we found positive associations between low (below the median of 5%) and medium (median to <25%) multiparticipant therapy levels and outcomes. However, associations disappeared with high (≥25%) levels of multiparticipant therapy. Results indicate that providing up to 25% of physical and occupational therapy in multiparticipant sessions may be both efficient and beneficial for patients. The final article, “Variability in Therapy Staffing Changes in Skilled Nursing Facilities Under the Patient Driven Payment Model,” used 2019-2020 publicly available SNF staffing data to test whether changes in physical and occupational therapy staffing under PDPM varied by SNF organizational characteristics. Using longitudinal models, we detected larger relative staffing declines in for-profit SNFs and facilities with more rural and Medicare patients, SNFs employing more therapy assistants, and SNFs providing more intensive therapy prior to policy change. Facilities serving more racially diverse patients lost more skilled therapists than SNFs with less diverse populations. Contract therapy declined more than in-house therapy in non-profit and high-quality facilities. Results indicate that SNFs that engaged in profit-maximizing behaviors under the previous payment model were more responsive to PDPM in terms of implementing larger therapy staffing reductions. The organizational characteristics identified in this article should be specifically targeted in quality monitoring and policy evaluation efforts to better understand the impacts of PDPM on patient outcomes, especially for vulnerable groups. Together, these articles provide information for key stakeholders, including clinicians, SNF administrators, and policymakers to help inform practice, guide future research on the impacts of PDPM on patient outcomes, and inform patient-centered policy.