Change in the prevalence of discontinuation^a of statin therapy in the patient subgroups during the 5-year exposure period.

<p>Repeated measures log-binomial regression analyses adjusted for age at retirement.</p><p>^a No purchases during a calendar year.</p><p>^bThe prevalence ratios of discontinuation are derived from contrasting the prevalence in the last year of follow-up to the prevalence in the first year</p><p>^c Secondary prevention: previous in- or outpatient hospital visits due to coronary heart disease or cerebrovascular diseases in any year before retirement.</p><p>Change in the prevalence of discontinuation<a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0130901#t003fn002" target="_blank">^a</a> of statin therapy in the patient subgroups during the 5-year exposure period.