The 130th Annual Meeting of APHA

Abstract #45769

Richard H. Fortinsky, PhD¹, Ramon Garcia, MS¹, Juliane R. Fenster, MS, MPH¹, Elizabeth Madigan, PhD, RN², and Susan Tullai-McGuinness, MSN, MPA, RN². (1) University of Connecticut Center on Aging, University of Connecticut Health Center, 263 Farmington Avenue, Farmington, CT 06030-5215, 860-679-8069, fortinsky@nso1.uchc.edu, (2) Frances Payne Bolton School of Nursing, Case Western Reserve University, 10900 Euclid Ave., Cleveland, OH 44106-4904

Hospitalization is considered a poor outcome for adults who receive Medicare-funded home care. Yet few studies have examined clinical, functional, and sociodemographic factors that may influence the likelihood of hospitalization among Medicare home care patients. Therefore, we studied patients (N=886; mean age=77.7 years; 67% female) from 22 Ohio home health agencies who completed an episode of Medicare home care in 1999 or 2000 and were either hospitalized (18.3%) or discharged to self-care at home. Data sources included the Outcome and Information Assessment Set for patient variables, and billing records for home health aide, physical therapy, and/or skilled nursing utilization. Logistic regression models estimated the likelihood of hospitalization; independent variables included age, gender, primary diagnosis, source of admission to home care, rehabilitation prognosis (good vs. guarded), severity of pain and dyspnea, and degree of dependence in ambulation and self care ADL. Along with these independent variables, three separate regression models were performed using home health aide, physical therapy, and skilled nursing utilization as predictors. Predictors associated with a greater likelihood (Wald test p value < .05) of hospitalization across all three regression models were: primary diagnosis of wound or skin problem (adjusted odds ratios (AOR)=2.7, 2.9, 3.0)); guarded rehabilitation prognosis (AOR=1.9, 1.9, 2.0); and dyspnea (AOR=1.5 in all models). Patients with a primary diagnosis of osteoarthritis were 92% less likely to be hospitalized in all models. Patients using home health aides were 34% less likely to be hospitalized. Implications for home care practice and policy will be discussed.

Learning Objectives:

• Explain how the Outcome and Assessment Information Set (OASIS) can be used to study Medicare home care patient outcomes.
• Discuss clinical, functional, and sociodemographic factors that affect likelihood of hospitalization after a Medicare home care episode.
• Understand how the likelihood of hospitalization is associated with home care clinical staff resource use.

Keywords: Outcomes Research, Home Care

Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.