Heart Failure in the Southern Community Cohort Study: Incidence, Mortality and Role of Anthropometric and Socioeconomic Factors

Abstract

We sought to investigate: a) heart failure (HF) incidence and post-HF survival by race and sex, b) the association between neighborhood deprivation and HF risk, and c) the dose-response relationship between surrogate measures of adiposity and both HF risk and post-HF survival among low-income adults in the southeastern US. Participants were 27,078 white and black men and women enrolled in the Southern Community Cohort Study (SCCS) who had no history of HF and were receiving Centers for Medicare or Medicaid Services (CMS). Incident HF diagnoses were ascertained using ICD-9 codes 428.x via linkage of the cohort with CMS research files. Cox models were used to investigate the association between candidate predictors – neighborhood deprivation index (NDI), empirically-derived weight-height indices (W/Hn), body mass index (BMI) and waist circumference – and both HF risk and post-HF mortality. Over a median follow-up of 5.2 years, 4,341 participants (16%) were diagnosed with HF and among these, 952 (21.9%) died. The HF incidence rate was 32.8/1000 PY, remarkably higher than previously reported in other studies. Racial and sex differences in HF incidence were minimal. Men had significantly lower post-HF survival. Neighborhood deprivation was associated with increased HF risk [Hazard Ratio for 1 interquartile range increase in NDI = 1.12; 95% CI: 1.07-1.18] in fully adjusted models. For the prediction of HF risk and post-HF survival, utilizing log W/Hn (versus log BMI) yielded better model fit statistics. The plot of the log relative hazard of HF vs. BMI was J-shaped with the nadir of the curve for blacks occurring at higher BMI values (30 versus 25kg/m2 among whites). The SCCS provided novel inferences on HF incidence, post-HF survival and the role of socioeconomic and anthropometric factors and it could serve as a useful resource for investigating determinants of HF risk in a segment of the population underrepresented in other existing cohorts.