Late life socioeconomic status and hypertension in an aging cohort: the Atherosclerosis Risk in Communities Study

Abstract

Objective: To investigate the association between individual and area-level socioeconomic status and hypertension risk among individuals later in life.

Methods: We used Cox proportional hazards models to examine the association of socioeconomic status with incident hypertension using race-specific neighborhood socioeconomic status, median household income, and education among 3372 participants (mean age, 61 years) from the Atherosclerosis Risk in Communities Study at Visit 4 (1996-1998). Incident hypertension was defined as self-reported diagnosis or reported use of antihypertensive medications.

Results: Over a median follow-up time of 9.4 years, there were 1874 new cases of hypertension (62.1 per 1000 person-years). Overall, being in high as compared with low socioeconomic status categories was associated with a lower risk of developing hypertension in late life, with hazard ratios (95% confidence intervals) of 0.87 (0.77-0.98) for high neighborhood socioeconomic status tertile, 0.79 (0.69-0.90) for high individual income, and 0.75 (0.63-0.89) for college education after adjustment for traditional risk factors. These findings were consistent and robust whenever accounting for competing risks of all-cause mortality. No significant interactions by race and age (dichotomized at age 65) were observed.

Conclusion: Among participants free of hypertension in midlife, high neighborhood and individual socioeconomic status are associated with a decreased risk of incident hypertension. Our findings support population-level interventions, such as blood pressure screening at senior centers and faith-based organizations, that are tailored to shift the distribution of blood pressure and reduce hypertension health inequalities among older adults.

FIGURE 1

Distribution of neighborhood socioeconomic status score by study site and race.

FIGURE 2

Kaplan-Meier plots examining cumulative incidence of hypertension by neighborhood socioeconomic status tertile for (a) all participants (P=0.0001), (b) whites (P = 0.0002), and (c) blacks (P=0.48). nSES, neighborhood socioeconomic status.