Sex differences in the association between smoking status and increased incidence of Alzheimer’s dementia

Abstract

AbstractBackgroundGrowing evidence indicates that smoking, a late life risk factor for Alzheimer’s dementia (AD), is significantly associated with increased prevalence of AD. However, the relationship between sex, smoking, and incident AD has not been well explored. The current study examined the relationship between sex, smoking status, and time to clinical diagnosis of AD.Method387 participants from UK Biobank were included in the study (mean±SE age=67.7±1.5 years, 50.9% female). Participants were over the age of 65 at their baseline visit and subsequently developed AD. Participant smoking status was determined from a baseline lifestyle questionnaire (8.5% current smoker, 45.7% previous smoker). Cox proportional‐hazard models were used to test the association between smoking and conversion. An interaction term and stratification by sex were used to assess the moderating role of sex in the association. Baseline age, APOE‐e4 status, sex, and Townsend deprivation index were included in the analyses.ResultMedian time to conversion was 6.39 years (interquartile range (IQR): 2.75) among never‐smoker participants, 6.36 years (IQR: 2.5) among previous smokers, while current smokers had a median time to conversion of 5.56 years (IQR: 2.59). Never‐smoker men had a shorter median time to conversion than never‐smoker women (6.20 (2.55) vs 6.57 (2.96) years). Current smoker men had a shorter median time to conversion than current smoker women (5.41 (2.69) vs 5.81 (1.87)). Previous smoker men, however, had a longer median time to conversion than previous smoker women (6.47 (2.46) vs 6.0 (2.58)). The hazard ratio (HR) for the current smokers relative to never‐smokers was 1.52 (confidence interval (CI): (1.03, 2.24), p=0.04), indicating that smoking increases the risk of conversion to AD by 52%. Smoking status and sex interaction (HR=0.65, CI: (0.42,0.99), p=0.05) revealed a significant difference between previous smoker men (HR=0.73, CI: (0.53,1), p=0.06) and previous smoker women (HR=1.11, CI: (0.83,1.5), p=0.48). Cessation of smoking had an opposing effect on the relative risk between men (reduced by 27%) and women (increased by 11%).ConclusionCurrent smoking status is associated with increased incidence of AD and differs by sex. Further analyses are needed to understand the causal mechanisms by sex underlying these associations.