Short sleep duration associated with the incidence of cardio-cerebral vascular disease: a prospective cohort study in Shanghai, China

Abstract

Importance: Sleep duration plays an important role in predicting CCVD incidence, and have implications for reducing the burden of CCVD. However, the association between sleep duration and predicted cardio-cerebral vascular diseases (CCVD) risk remains to be fully understood.

Objective: To investigate the effects of sleep duration on the development of CCVD among Chinese community residents.

Design: A prospective cohort study. The baseline survey was conducted from January 2013 to July 2013. The cohort has been followed until December 31, 2016 using a combination of in-person interviews and record linkages with the vital registry of Pudong New Area, Shanghai, China.

Subjects: A total of 8245 Chinese community residents were initially enrolled in the cohort. Of those, 6298 underwent the follow-up examination.

Exposure: Self-reported sleep duration and sleep quality were obtained via the questionnaire. Sleep duration was divided into five categories: ≤5, 6, 7, 8, or ≥ 9 h per day.

Main outcome(s) and measure(s): CCVD, Coronary heart disease (CHD) and Stroke occurrence, Hazard ratios (HRs) and 95% confidence intervals (95% CIs) were calculated using Fine-Gray proportional subdistribution hazards models.

Results: During a median follow-up of 3.00 years (IQR 2.92-3.08), we observed 370 participants have had incident CCVD events, of whom 230 had CHDs, 169 had strokes, and 29 had both. After adjustment for relevant confounders, short sleepers (≤ 5 h) had 83% higher risk of total CCVD incidence (HR: 1.83; 95% CI: 1.32-2.54), 82% higher risk of CHD incidence (HR: 1.82; 95% CI: 1.21-2.75), and 82% higher risk of stroke incidence (HR: 1.82; 95% CI: 1.12-2.98) in contrast to the reference group (7 h). Some of these U-shaped relationships varied by age, and were more pronounced in individuals aged < 65 years. Individuals who slept ≤ 5 h per day with baseline hypertension had the highest risk of CCVD incidence (HR: 3.38, 95% CI 2.08-5.48), CHD incidence (HR: 3.11, 95% CI 1.75-5.53), and stroke incidence (HR: 4.33, 95% CI 1.90-9.86), compared with those sleep 7 h and without baseline hypertension.

Conclusions: Short sleep duration is independently associated with greater incidence of CCVD, CHD and stroke.

Keywords: Cardio-cerebral vascular disease; Coronary heart disease; Prospective cohort study; Sleep duration; Stroke.

Fig. 1

Diagram of this study

Fig. 2

Multivariable adjusted spline curves for association between sleep duration and risk for (A) CCVD incidence, (B) CHD incidence, and (C) stroke incidence, adjusted for age, sex, marital status, current smoking, alcohol consumption, tea consumption, physical activity, hypertension, type 2 diabetes, dyslipidemia, BMI, CRP, and sleep quality

Fig. 3

Hazard ratios (95% CI) for incident CCVD(A), CHD(B), and Stroke(C) in subjects above and below 65 years according to sleep duration. The adjusted covariates included age, sex, marital status, current smoking, alcohol consumption, tea consumption, physical activity, hypertension, type 2 diabetes, dyslipidemia, BMI, CRP, and sleep quality, with sleep duration of 7 h as the reference group

Fig. 4

Joint association of sleep duration and baseline hypertension status with (A) CCVD incidence, (B) CHD incidence, and (C) stroke incidence. All HRs and 95% CIs were computed with sleep duration of 7 h and absence of hypertension as the reference group, adjusted for age, sex, marital status, current smoking, alcohol consumption, tea consumption, physical activity, hypertension, type 2 diabetes, dyslipidemia, BMI, CRP, and sleep quality