Voluntary early retirement and mortality in patients with and without chronic diseases: a nationwide Danish Registry study

Abstract

ObjectiveThis study explores how the choice of voluntary early retirement (VER) affects mortality in a population where VER is available 5 years before regular retirement age. Study designThis retrospective cohort study uses a registry-based follow-up design with access to Nationwide Danish Registry Data. MethodsThe study includes all Danish individuals who between 2000 and 2015 were part of an unemployment insurance fund and working at the time of their 60th (P60) or 62nd (P62) birthday. Those alive 1 year from their 60th or 62nd birthday were included in the mortality analysis. Individuals were registered as VER recipients if they chose the benefit within 1 year from P60 or P62. Three-year mortality likelihood following the first year from inclusion was explored for both cohorts separately. Multiple subgroups were explored in the mortality analysis, including individuals with chronic obstructive pulmonary disease (COPD), heart failure, and diabetes. ResultsP60 included 627,278 individuals, and VER was chosen by 22.5%. P62 included 379,196 individuals, and VER was chosen by 33.4%. The likelihood of VER in the P60 was lower in healthy individuals (odds ratio [OR] 0.87, confidence interval [CI] 0.85–0.88) and higher in COPD (OR 1.15, CI 1.07–1.22) and heart failure patients (OR 1.15, CI 1.05–1.25). Three-year mortality was significantly higher in those choosing VER in P60 (OR 1.28, CI 1.22–1.34), which was also found for all health subgroups (healthy, OR 1.18, CI 1.07–1.30; COPD, OR 1.55, CI 1.16–2.07; heart failure, OR 1.42, CI 1.02–1.98; diabetes, OR 1.36, CI 1.12–1.65). The increased mortality risk was not found in the P62 cohort. ConclusionThe choice of VER is more likely in patients with COPD and heart failure. VER in the P60 cohort is associated with an increased mortality likelihood, which was not found in the P62 cohort, which may be explained by health selection bias.