Lipid-lowering drug treatment and mortality among individuals ≥75 years without cardiovascular disease: a population-based cohort study

Abstract Aims Lipid-lowering drugs are commonly prescribed for prevention of cardiovascular disease (CVD), but their efficacy in people aged ≥75 years is less understood. This population-based cohort study aimed to examine the association of lipid-lowering drug treatment with all-cause and cardiovascular mortality among people aged ≥75 years and whether the association, if existent, varied with patient characteristics. Methods and results A nationally representative sample of 6409 adults aged ≥75 years and without CVD at baseline was drawn from the Third National Health and Nutrition Examination Survey (NHANES III) conducted between 1988 and 1994 and the 10 continuous NHANES cycles between 1999–2000 and 2017–18 in the USA. Among these participants, 1227 were on lipid-lowering drug treatment at baseline. All-cause and cardiovascular deaths were ascertained by linkage to National Death Index records through 31 December 2019. The association of lipid-lowering drug treatment with mortality was investigated through multivariable Cox regression analysis and expressed as hazard ratio (HR) with 95% confidence interval (CI), adjusting for major risk factors of CVD and mortality. Complete-case analysis, competing risk analysis, and exclusion of deaths occurring within the first year of the follow-up were conducted to examine the robustness of results. Subgroup analyses were conducted to detect potential interaction between lipid-lowering drug treatment and important patient characteristics including demographics, lifestyle factors, comorbidities, baseline CVD risk, and concurrent medications. During a median follow-up of 6.5 years (78 months), 4634 (72.3%) participants died and 1834 (28.6%) of them were attributed to cardiovascular causes. Use of lipid-lowering drugs was associated with lower risks of both all-cause mortality (adjusted HR 0.74, 95% CI 0.67–0.81, P < 0.001) and cardiovascular mortality (adjusted HR 0.64, 95% CI 0.54–0.76, P < 0.001). After adjusting for covariates, the overall survival of users was 1.6 years longer than that of non-users. For all-cause mortality, the adjusted HRs for ‘within 2.5 years’, ‘2.5–7.5 years’, and ‘over 7.5 years’ after baseline were 0.61 (95% CI 0.49–0.76), 0.72 (95% CI 0.63–0.83), and 0.82 (95% CI 0.71–0.94), respectively. For cardiovascular mortality, the corresponding results were 0.46 (95% CI 0.30–0.68), 0.59 (95% CI 0.47–0.76), and 0.78 (95% CI 0.62–0.97), respectively. Various sensitivity analyses yielded consistent results with the main analyses. The HRs for all-cause mortality were statistically significant in all subgroups defined by age (including the ‘≥85 years’ group), sex, self-rated current health status, body mass index, hypertension, diabetes, estimated 10-year cardiovascular risk at baseline, cancer, and concurrent use of common medications, favouring lipid-lowering drug users. In those without high cholesterol, with chronic kidney disease, or with pulmonary disease, the HRs were not statistically significant, with wide CIs, partly due to the small number of users. Similar patterns were observed for cardiovascular mortality. Conclusion In this study population of people aged ≥75 years and without CVD at baseline, use of lipid-lowering drugs was associated with significantly lower risks of all-cause and cardiovascular mortality.