Healthy sleep score and incident cardiovascular diseases: the Paris Prospective Study III (PPS3)

Abstract

Abstract Background/Introduction Most studies on the association between sleep habits and cardiovascular disease (CVD) have focused on one single sleep dimension, essentially sleep duration and sleep apnea. Purpose To examine the joint effect of several dimensions of sleep habits with incident CVD in a community-based prospective cohort. Methods Between 2008 and 2011, 10,157 men and women aged 50 to 75 years were recruited in a preventive medical center. They underwent a standard physical examination coupled with standard biological tests, and provided information related to lifestyle, personal and family medical history, current health status, and medication use on questionnaires. Sleep habits were self-reported on validated questionnaires that assess sleep duration and insomnia complaints (Pittsburg questionnaire), early chronotype, sleep apnea (Berlin questionnaire) and subjective daytime sleepiness (Epworth questionnaire). Each sleep dimension was assigned 1 point if optimal and 0 point otherwise. A healthy sleep score ranging from 0 to 5 (the higher the better) and reflecting the number of optimal sleep dimensions was computed: early chronotype, sleep duration of 7–8 h/day, never/rarely insomnia, no sleep apnea, and no frequent excessive daytime sleepiness. The occurrence of incident CVD events including coronary heart disease and stroke was followed every two years up to September 2020, and events were validated after review of the medical records. The multivariable association between higher healthy sleep score and CVD events was examined in proportional hazard Cox regression analysis. Population-attributable fractions were calculated to estimate the proportion of CVD cases that could be prevented by healthier sleep habits. Results This study included 7203 participants (62% of men, mean age: 59.7 years±6.2) who were free of CVD at baseline and had complete data on sleep habits and covariates. Among them, 6.9% had a poor sleep score (healthy sleep score of 0 or 1), and 10.4% had an optimal sleep score (score= 5). After a median follow-up of 8 years, 275 participants had incident CVD events. After adjustment for age, sex, total alcohol consumption, socioprofessional categories, smoking status, body mass index, physical activity, family history of heart diseases, LDL and HDL cholesterol, and diabetes status, the risk of CVD decreased by 22% (HR=0.78 [95% CI: 0.71–0.86]) per 1 point increment in the healthy sleep score, and there was a 74% risk reduction in CVD risk (HR=0.26 [0.13–0.51]) between participants with the highest (score of 5) and those with the lowest (score of 0–1) healthy sleep score (Table 1). Under the hypothesis that all the participants would achieve an optimal sleep score of 5, 70.8% of incident CVD could be potentially avoided each year. Conclusion(s) In this community-based prospective cohort, a higher healthy sleep score combining 5 sleep dimensions was associated with a lower risk of CHD or stroke. Funding Acknowledgement Type of funding sources: Public Institution(s). Main funding source(s): The National Research Agency (ANR), The Region Ile de France (Domaine d'Intérêt Majeur)