Objectives This study aimed to examine the association between statin exposure and dementia risk in individuals with hypercholesterolaemia using data from the NHIS‐HEALS database between 2002 and 2015. Methods Subjects were classified into statin exposure and statin nonexposure groups according to medication possession ratio. Dementia was defined as those with primary diagnostic dementia codes such as F00‐F03, G30, G31.1, G31.9 or G31.82. Cox proportional hazards regression models were adopted after stepwise adjustment for confounders to investigate the prospective association between statin exposure and dementia risk. Results During the follow‐up period (median follow‐up 11.7 years), 711 cases of dementia occurred, accounting for 11.5% of the total study population (statin exposure group, 8.2%; statin nonexposure group, 12.9%). Compared to the statin nonexposure group, fully adjusted hazard ratios (HRs) (95% confidence intervals CIs) for overall dementia in the statin exposure group were 0.63 (0.43–0.91) and 0.62 (0.50–0.78) in men and women, respectively. Compared to the statin nonexposure group, the HRs (95% CIs) for Alzheimer’s disease and related dementia, vascular dementia and other types of dementia in the statin exposure group were 0.54 (0.32–0.91), 2.45 (0.69–8.68) and 0.59 (0.32–1.07), respectively, in men and 0.53 (0.38–0.73), 1.29 (0.42–3.96) and 0.70 (0.51–0.96), respectively, in women. Conclusions Hypercholesterolaemic individuals exposed to statin had a lower risk of overall dementia and Alzheimer’s disease and related dementia in both sexes, and a lower risk of other types of dementia in women, than subjects who were not exposed to statins.