Background: Although there is strong evidence of the benefits of antihypertensive treatment, the high prevalence of this important cardiovascular risk factor and its complications, as well as the low control rates of hypertension observed in many studies justify the investigation of these relationships in population studies. The objective was to investigate the ratio of cardiovascular disease mortality between hypertensives (non-treated, controlled and uncontrolled) and non-hypertensives in a cohort of a population sample of adults living in Ilha do Governador, Rio de Janeiro state, Brazil, who were classified in a survey conducted in 1991 and 1992 and whose death certificates were sought 19 years later.
Methods: A cohort study was performed on probabilistic linkage between data from an epidemiological study of hypertension performed in Ilha do Governador, in Rio de Janeiro, Brazil (1991 to 1992) and data from the Mortality Information System of Rio de Janeiro (1991 to 2009). The survey aimed to estimate the prevalence of hypertension and other cardiovascular risk factors in 1,270 adults aged 20 years or older selected through a probabilistic sampling of households at three economic levels (low, middle and high income). We performed a probabilistic record linkage of these databases and estimated the risk of cardiovascular death using Kaplan-Meier method to plot survival curves and Cox proportional hazards models comparing hypertensive subjects all together, and by hypertension subgroups: untreated, controlled, and uncontrolled hypertensives with non-hypertensive ones.
Results: A total of 170 deaths occurred, of which 31.2 % attributed to cardiovascular causes. The hazard ratio for cardiovascular death was 6.1 times higher (95 % CI 2.7 - 13.7) in uncontrolled hypertensive patients relative to non-hypertensive patients. The hazard ratios for untreated hypertensive and controlled hypertensive patients were 2.7 times (95 % CI 1.1 - 6.3) and 2.1 times (95 % CI 0.38 - 11.5) higher than for normotensive patients, respectively.
Conclusion: The present study demonstrated a higher cardiovascular death risk among hypertensive than among non-hypertensive ones that is not associated uniquely to treatment, because uncontrolled hypertensives demonstrated a greater risk than untreated ones. Although the subgroups of hypertensive individuals were susceptible to changes in their classification over the 19 years of the study, the baseline classification was consistent with a worse prognosis in these individuals.