Abstract MP60: Associations of Socio-Economic and Access to Care Factors With Inpatient and Outpatient Heart Failure. The Aric Study
BACKGROUND: Little is known about the burden of heart failure (HF) managed in outpatient settings and of the characteristics that influence the degree to which HF is diagnosed in outpatient venues predating a hospitalization. We hypothesized that access to care and socio-economic contextual factors relate to patient ascertainment of incident HF in outpatient vs inpatient settings.
METHODS: We created an open cohort of consecutive fee-for-service CMS Medicare beneficiaries (2003 - 2006) residing in the four epidemiologic surveillance areas of the Atherosclerosis Risk in Communities (ARIC) Study. Using a 24 month look-back period we classified incident inpatient HF (IP HF) diagnoses as the first observed HF hospitalization (ICD-9 code 428.xx in any position). Incident outpatient HF (OP HF) diagnoses were defined as two outpatient HF encounters within 365 days, with a 12 month look-back period. Three contextual factors were characterized at the level of zip code tabulation area: number of primary care physicians per 1,000 population (grouped by tertiles), the proportion of the population living in a medically underserved area (MUA) (any vs. none), and the proportion of the population living in poverty (high defined as 10% or more). Associations of contextual factors with incident IP HF or OP HF diagnoses were estimated as incidence rate ratios (IRR, 95% CI) using Poisson generalized linear models. We estimated separate models for each contextual factor while controlling for the other contextual factors, age, sex, race, and ARIC community.
RESULTS: The study population included 106,585 consecutive fee-for-service beneficiaries (40% male, median age 75 years, 11.5% non-white). We observed 3,348 incident IP HF diagnoses and 1,050 OP HF diagnoses over 310,689 person-years of follow-up (age-adjusted rates per 1,000 person-years: IP HF 15.8 (95% CI 15.0, 16.6); OP HF rate 4.7 (4.4, 5.2). The adjusted rates of IP HF diagnosis were similar across strata of physician density, while rates of OP HF diagnoses were higher in low physician density areas compared to high physician density areas (IRR 1.23, (1.17, 1.29). Rates of IP HF and OP HF diagnoses were higher in high poverty areas compared to low poverty areas (IP HF IRR 1.23 (1.10, 1.38); OP HF IRR 1.26 (1.03, 1.55)). In areas with any MUA population, IP HF rates were higher as compared to areas with no MUA (IRR 1.10 (1.01, 1.21)) but no difference was found for OP HF rates.
CONCLUSION: Contextual factors pertaining to low access to care and low socio-economic status were associated with differences in the rate of inpatient versus outpatient HF diagnoses among CMS Medicare beneficiaries in four ARIC Study communities. Further research is needed to understand the effect of observed differences in the clinical location of HF diagnosis on outcomes.
Author Disclosures: C. Cuthbertson: None. A. Kucharska-Newton: None. M. Patel: None. R. Camplain: None. R. Foraker: None. L. Wruck: None. A. Folsom: None. N. Puccinelli-Ortega: None. K. Matsushita: None. G. Heiss: None.
- © 2015 by American Heart Association, Inc.