Abstract
Aims/hypothesisSleep duration is a risk factor for incident diabetes mellitus and CHD. The primary aim of the present study was to investigate, in sex-specific analyses, the role of incident diabetes as the possible biological mechanism for the reported association between short/long sleep duration and incident CHD. Considering that diabetes is a major risk factor for CHD, we hypothesised that any association with sleep duration would not hold for cases of incident CHD occurring before incident diabetes (‘non-diabetes CHD’) but would hold true for cases of incident CHD following incident diabetes (‘diabetes-CHD’).MethodsA total of 6966 men and 9378 women aged 45–73 years from the Malmö Diet Cancer Study, a population-based, prospective cohort, who had answered questions on habitual sleep duration and did not have a history of prevalent diabetes or CHD were included in the analyses. Incident cases of diabetes and CHD were identified using national registers. Sex-specific Cox proportional hazards regression models were stratified by BMI and adjusted for known covariates of diabetes and CHD.ResultsMean follow-up times for incident diabetes (n = 1137/1016 [men/women]), incident CHD (n = 1170/578), non-diabetes CHD (n = 1016/501) and diabetes-CHD (n = 154/77) were 14.2–15.2 years for men, and 15.8–16.5 years for women. In men, short sleep duration (< 6 h) was associated with incident diabetes (HR 1.35, 95% CI 1.01, 1.80), CHD (HR 1.41, 95% CI 1.06, 1.89) and diabetes-CHD (HR 2.34, 95% CI 1.20, 4.55). Short sleep duration was not associated with incident non-diabetes CHD (HR 1.35, 95% CI 0.98, 1.87). Long sleep duration (≥ 9 h) was associated with incident diabetes (HR 1.37, 95% CI 1.03, 1.83), CHD (HR 1.33, 95% CI 1.01, 1.75) and diabetes-CHD (HR 2.10, 95% CI 1.11, 4.00). Long sleep duration was not associated with incident non-diabetes CHD (HR 1.33, 95% CI 0.98, 1.80). In women, short sleep duration was associated with incident diabetes (HR 1.53, 95% CI 1.16, 2.01), CHD (HR 1.46, 95% CI 1.03, 2.07) and diabetes-CHD (HR 2.88, 95% CI 1.37, 6.08). Short sleep duration was not associated with incident non-diabetes CHD (HR 1.29, 95% CI 0.86, 1.93).Conclusions/interpretationThe associations between sleep duration and incident CHD directly reflect the associations between sleep duration and incident diabetes. Incident diabetes may thus be the explanatory mechanism for the association between short and long sleep duration and incident CHD.
Highlights
A suggested novel risk factor for incident diabetes mellitus is sleep duration; insufficient or excessive amounts of sleep may contribute to the development of diabetes, with prospective studies reporting that short but not long sleep duration [1,2,3,4], or both short and long sleep duration, are associated with incident diabetes [5,6,7]
Incident diabetes may represent a change in the risk of developing future CHD, yet no study to date has been able to conclusively demonstrate or refute whether only those who are diagnosed with incident diabetes during follow-up constitute a specific group in which short and long sleep duration are associated with incident CHD
The associations between sleep duration and incident CHD directly reflect the associations between sleep duration and incident diabetes, and when taken together with the strong association between sleep duration and diabetes-CHD, our study convincingly demonstrates that incident diabetes is the most probable explanatory biological mechanism for the positive associations found between sleep duration and CHD
Summary
A suggested novel risk factor for incident diabetes mellitus is sleep duration; insufficient or excessive amounts of sleep may contribute to the development of diabetes, with prospective studies reporting that short but not long sleep duration [1,2,3,4], or both short and long sleep duration, are associated with incident diabetes [5,6,7]. Cardiometabolic risk factors, including prevalent type 2 diabetes, have been discussed as possible mediators for the association between sleep duration and CHD [11, 14], with prospective studies adjusting for prevalent baseline cases of diabetes in their analyses [11,12,13,14,15]. This may be insufficient as the same studies do not take into account the incident cases of diabetes occurring during follow-up. Incident diabetes may represent a change in the risk of developing future CHD, yet no study to date has been able to conclusively demonstrate or refute whether only those who are diagnosed with incident diabetes during follow-up constitute a specific group in which short and long sleep duration are associated with incident CHD
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
