• Bruce Leff, Christine Ritchie, Sarah Szanton, Oren Shapira, Amanda Sutherland, Andrew Lynch, Brian W Powers, Mona Siddiqui, and Katherine A Ornstein.
• Division of Geriatric Medicine and Gerontology, Department of Medicine, Center for Transformative Geriatric Research, Johns Hopkins University School of Medicine; Department of Health Policy and Management, Johns Hopkins University Bloomberg School of Public Health; and Department of Community and Public Health, Johns Hopkins School of Nursing, Baltimore, Maryland (B.L.).
• Ann. Intern. Med. 2024 Sep 1; 177 (9): 119912081199-1208.

BackgroundInterest in home-based care is increasing among Medicare Advantage (MA) plans. The epidemiology of homebound MA beneficiaries is unknown.ObjectiveTo determine the prevalence, characteristics, predictors, health service use, and mortality outcomes of homebound beneficiaries of a large national MA plan.DesignCross-sectional.SettingNational MA plan.ParticipantsHumana MA beneficiaries in 2022 (n = 2 435 519).MeasurementsHomebound status was assessed via in-home assessment using previously defined categories: homebound (never or rarely left home in the past month), semihomebound (left home with assistance, had difficulty, or needed help leaving home), and not homebound. Demographic, clinical, health service use, and mortality outcomes were compared by homebound status.ResultsIn 2022, the overall prevalence of homebound beneficiaries was 22.0% (8.4% of beneficiaries were homebound, and 13.6% were semihomebound). In adjusted models, female sex (odds ratio [OR], 1.36 [95% CI, 1.35 to 1.37), low-income status or dual eligibility for Medicare and Medicaid (OR, 1.56 [CI, 1.55 to 1.57]), dementia (OR, 2.36 [CI, 2.33 to 2.39]), and moderate to severe frailty (OR, 4.32 [CI, 4.19 to 4.45]) were predictive of homebound status. In multivariable logistic regression, homebound status was associated with increased odds of any emergency department visit (OR, 1.14 [ CI, 1.14 to 1.15]), any inpatient hospital admission (OR, 1.44 [CI, 1.42 to 1.46]), any skilled-nursing facility admission (OR, 2.18 [CI, 2.13 to 2.23]), and death (OR, 2.55 [CI, 2.52 to 2.58]).LimitationThe study period overlapped the tail end of the COVID-19 pandemic, and data were derived from a single national MA plan, which limits generalizability.ConclusionOverall homebound prevalence in a national MA plan was 22.0% and was independently associated with increased health service use and mortality. Study findings can inform strategic initiatives to identify and manage care for homebound beneficiaries.Primary Funding SourceHumana, under a collaborative research agreement with Johns Hopkins University.

24 keywords...

hide…