Objectives A rapidly increasing number of retirements is projected in most developed countries, with the oldest baby-boomer generation now reaching 65 years of age. We sought to determine whether retirement affects medication adherence among hypertensive men and women.

Methods A prospective cohort study using linked national registers for 1825 hypertensive men and women (mean age 59.3 years at retirement, 71% female, 100% White) (the Finnish Public Sector Study). Trajectories of antihypertensive medication adherence from 3 years before, to 3 years after, retirement were estimated based on reimbursed prescriptions and World Health Organization defined average maintenance doses. Year 0 refers to the year of retirement.

Results Multilevel longitudinal analysis of 11 588 person-years, adjusted for sex, age at retirement and calendar year, showed a 30% increase in days not treated in post-retirement years (years +1 to +3) compared to pre-retirement years (years -3 to -1) (rate ratio 1.29, 95% CI: 1.07 to 1.56). The post-retirement drop in medication adherence was observed in men and women, and it was not dependent on the type of retirement or length of time being treated by antihypertensive medication. No such drop was seen in the subgroup of retirees who had a serious co-morbid chronic condition requiring continuous drug treatment, such as diabetes mellitus, coronary heart disease or cerebrovascular disease.

Conclusions Retirement is associated with a deterioration in medication adherence among hypertensive men and women free of co-morbid conditions.