Cary Submits NIH Grant Application

Kudos to Michael Cary, associate professor, and his entire team for the resubmission of his NIH R01 application entitled: "Successful Community Discharge Following Inpatient Rehabilitation for Older Adults." This proposal requests funds for a 5-year period with a start date of July 1, 2020.

Nearly 1 in 8 of the 339,000 Medicare beneficiaries who receive post-acute care (PAC) in Inpatient Rehabilitation Facilities (IRFs) is rehospitalized within 30 days; with 1-year rehospitalization rates as high as 60%. Current models are not accurate enough to identify high-risk patients for additional interventions. Many of these models incorporate comorbidities and prior-year hospital use. Few consider key drivers of readmission among IRF patients such as social or functional status variables. Use of health services following IRF discharge could also improve model predictability, but the significance of such factors has not been widely studied. Without this knowledge, designing effective interventions to reduce rehospitalizations among older adults treated in IRFs is not possible. The purpose of this study is to develop an innovative prediction model for successful community discharge for older adults treated in IRFs.

Using 2014-2018 data from the Uniform Data System for Medical Rehabilitation (UDSMR) and CMS (>200,000 patients treated within 1,100 IRFs), and analytic strategies including multivariate logistic and survival models, the following aims and sub-aims have been proposed: Aim 1. Quantify the impact of functional limitations (mobility, self-care, cognition) and geriatric syndromes (incontinence, pressure ulcer, depression, delirium, and falls) on successful community discharge (no rehospitalization or death within 31-days) following inpatient rehabilitation for older adults. Subaim 1a. Quantify the impact of these factors on 1-year outcomes. Multilevel analyses using linear mixed and survival models adjusting for facility-level variation will be conducted. Subaim 1b. Develop and validate models that can be used in real time to calculate 1-year risks for IRF patients in clinical practice. Subaim 1c. Build web-based tool of this validated model to be implemented in real-world IRFs. Aim 2. Quantify the impact of health services use (primary care, specialty care, therapy type and setting, and home health) on successful community discharge following inpatient rehabilitation for older adults. Subaim 2a. Quantify the impact of these factors on 1-year outcomes. Multilevel analyses using linear mixed and survival models adjusting for facility-level variation will be conducted; an innovative combination of propensity score and instrumental variable approach will be used to account for selection bias. Our national study will be the first to build a predictive model of successful older adult community discharge for clinical use in IRFs. From Aim 1, we will develop prediction models comprised entirely of clinical information collected from patients during their IRF stay; we will also validate and build a web-based application to estimate 1-year readmission risk for IRFs in real-time. From Aim 2, we will identify the optimal combination of health services that best support patients’ successful return to the community, a finding that will inform care coordination across settings and new payment models. Together, these aims will inform our next study: a pilot study of a transitional care intervention targeting IRF patients at high risk for rehospitalization.