Association of sleep patterns and cardiovascular disease risk is modified by glucose tolerance status

Abstract Aims To investigate whether the association between sleep patterns and cardiovascular disease (CVD) risk differs according to glucose tolerance status. Materials and Methods This prospective study included 358,805 participants initially free of CVD from the UK Biobank. We created a sleep score based on five sleep factors (sleep duration, chronotype, insomnia, snoring, and daytime sleepiness) with one point for each unhealthy factor. Cox proportional hazards models were used to examine the association between sleep and incident CVD, including coronary heart disease (CHD) and stroke, according to normal glucose tolerance (NGT), prediabetes, and diabetes. Results During a median follow‐up of 12.4 years, 29,663 incident CVD events were documented. There was a significant interaction between sleep score and glucose tolerance status on CVD ( P for interaction = 0.002). Each 1 point increment in sleep score was associated with a 7% (95% confidence interval 6%–9%), 11% (8%–14%), and 13% (9%–17%) higher risk of CVD among participants with NGT, prediabetes, and diabetes, respectively. Similar interaction patterns were observed for CHD and stroke. Among the individual sleep factors, sleep duration and insomnia significantly interacted with glucose tolerance status on CVD outcomes (all P for interaction <0.05). All five unhealthy sleep factors accounted for 14.2% (8.7%–19.8%), 19.5% (7.4%–31.0%), and 25.1% (9.7%–39.3%) of incident CVD cases among participants with NGT, prediabetes, and diabetes, respectively. Conclusions The CVD risk associated with a poor sleep pattern was exacerbated across glucose intolerance status. Our findings emphasise the importance of integrating sleep management into a lifestyle modification programme, particularly in people with prediabetes or diabetes.