Sleep and physical activity as predictors of working life expectancy at the age of 50 years

Abstract Background Retirement ages have been set to increase in many European countries due to rapid population ageing. It is essential to understand different factors that may impact labour market participation and the duration of working lives at older ages. The aim of this study was to examine sleep and physical activity as predictors of working life expectancy (WLE) between ages 50 and 68. Methods Data of individuals ≥50 years were drawn from the Finnish Public Sector Stud. The study population consisted of 70,339 public sector employees (80% women) aged at least 50 years in 2000-2016. Participants reported their usual sleep duration (short: <6.5h, mid-range: 7-8.5h, long: ≥9h), sleep disturbances (none, moderate, and severe), and leisure-time and commuting physical activity (inactive, low, moderate, high, and vigorous). WLE from age 50 to 68 years was ascertained from a nationwide register of pensionable earnings. Multistate life table models were used to estimate WLE for categories of sleep and physical activity and stratified by occupation (categorized into professional, intermediate, and routine) and sex. Results On average, WLE at the age of 50 years was 13.15 years (95% CI 13.13-13.18). Long sleep duration was associated with 0.60-0.65 years shorter WLE compared to short and mid-range sleep duration among both men and women in all occupational classes at 50 years of age. A steep decline towards shorter WLE was also observed with more severe sleep disturbances and lower physical activity among all occupational classes and both men and women. Those in professional occupations with short or mid-range sleep duration, no sleep disturbances, or vigorous physical activity had the longest WLE, and no major sex differences were observed. Conclusions Supporting regular physical activity and sleep of recommended amount and good quality among older workers can help to promote longer working lives.