Post-Acute Care Referral and Outcomes for Patients with Alzheimer's Disease and Related Dementias

Harvard Medical School, Boston MA

Investigators

Linked publications & trials

Abstract

PROJECT SUMMARY Alzheimer's disease and related dementias (ADRD) constitute a common, morbid and costly set of conditions among older persons. Roughly 1.4 million individuals with ADRD are discharged annually from the hospital, with many of these individuals referred to a post-acute care (PAC) setting. Unfortunately, a paucity of knowledge exists about how individuals with ADRD are discharged to a PAC setting and which PAC setting leads to the best outcomes for this vulnerable patient population. Our overall goal is to understand the comparative effectiveness of PAC settings through a comprehensive understanding of the referral process, whether individuals with ADRD are discharged to lower quality skilled nursing facilities (SNFs) and home health agencies (HHAs), and which PAC settings are most appropriate for patients with ADRD. Being able to disentangle the appropriate PAC setting is critical in the context of the growing use of alternative value-based payments models which are designed to encourage a more efficient use of health care services. Our conceptual framework is motivated by the idea that ADRD patients might be particularly susceptible to provider-driven decision making in the discharge process due to their impaired cognitive abilities. Our underlying hypotheses are: i) ADRD patients will tend to utilize more intensive PAC settings, ii) ADRD patients will be discharged to lower quality SNFs and HHAs, and iii) for similar ADRD patients, lower intensity PAC settings can generate improved outcomes. The proposed study will link Medicare claims data and SNF and HHA assessment data across the 2007-2015 nine-year period to examine PAC outcomes for all ADRD patients and for ADRD patients with five tracer conditions: hip fracture, lower extremity joint replacement, coronary heart failure, chronic obstructive pulmonary disease, and stroke. In order to account for non-random selection across PAC settings, we will use an instrumental variable approach based upon the differential distances from the discharging hospital and beneficiary zip code to different PAC providers. The expected results of our proposed aims are a deeper understanding of how availability and PAC quality impact utilization patterns and patient outcomes. This knowledge has the potential for informing current and future Medicare PAC policies intended to improve post-hospital discharge outcomes for patients with ADRD.