Congestive heart failure hospitalizations and survival in California: Patterns according to race ethnicity

Background Congestive heart failure (CHF) disproportionately affects Africa n Americans, but data are limited concerning CHF hospitalization patterns a mong Hispanic and Asian populations, the 2 fastest growing ethnic groups in the United States, and race/ethnic patterns of rehospitalization and survi val among patients with CHF are unknown. We conducted a study to assess rat es of CHF hospitalization, readmission, and survival among diverse populati ons in California. Methods and Results we used 2 study designs. First, we calculated the popul ation-based incidence of CHF hospitalization in California in 1991. Next we conducted a retrospective cohort study that identified patients initially hospitalized for CHF in 1991 or 1992 and followed these patients For 12 mon ths after their index hospitalization to determine their likelihood of reho spitalization or death. Data were analyzed with Cox proportional hazards mo dels. African Americans had the highest rate of CHF hospitalization. Age-ad justed hospitalization rates were comparable among whites, Latinos, and Asi an women and all lower than those in African American, whereas Asian men ha d the lowest rates. On adjusted analyses, African Americans were more likel y than whites and Asians to be rehospitalized (relative risk 1.07; 95% conf idence interval 1.04 to 1.10). However, they were less likely to die within the 12-month follow-up period (relative risk 0.86; 95% confidence interval 0.82 to 0.90). Whites, conversely, had the highest posthospitalization mor tality rates. Conclusions These findings demonstrate important racial-ethnic differences in CHF morbidity and mortality rates. The disparate findings of higher hosp italization and rehospitalization rates and lower mortality rates among Afr ican Americans than whites may represent differences in the underlying path ophysiology of CHF in these groups or differences in access to quality care . Further studies are needed to explain these seemingly paradoxical outcome s.