Insomnia symptoms and risk of cardiovascular diseases among 0.5 million adults: A 10-year cohort

Abstract

Objective: To examine the associations of individual insomnia symptoms with risks of incident cardio-cerebral vascular diseases (CVD) and possible moderating factors among Chinese adults.

Methods: The China Kadoorie Biobank is a prospective cohort study that recruited participants from 10 areas across China. Data from 487,200 adults 30 to 79 years of age who were free of stroke, coronary heart disease, and cancer at baseline were analyzed. Three insomnia symptoms were assessed with self-reported difficulties in initiating or maintaining sleep, early morning awakening, and daytime dysfunction for at least 3 d/wk at baseline. Incidences of CVD were followed up through disease registries and national health insurance databases until 2016.

Results: During a median of 9.6 years of follow-up, 130,032 cases of CVD were documented. Cox regressions showed that 3 insomnia symptoms were associated with increased risk of total CVD, with respective adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) of 1.09 (95% CI 1.07-1.11), 1.07 (95% CI 1.05-1.09), and 1.13 (95% CI 1.09-1.18). Participants with individual symptoms also had higher risks of ischemic heart disease (IHD; HR 1.13, 1.09, and 1.17) and ischemic stroke but not hemorrhagic stroke. Participants with all 3 symptoms were at an 18%, 22%, or 10% higher risk of CVD, IHD, or ischemic stroke compared to nonsymptomatic adults. Associations between 3 symptoms and CVD incidence were consistently stronger in younger adults or those without baseline hypertension (p for interaction <0.05).

Conclusions: Individual and coexisting insomnia symptoms are independent risk factors for CVD incidence, especially among young adults or adults who have not developed hypertension.

Figure 1. HR (95% CI) of total and subtypes of CVD incidence according to number of insomnia symptoms

Models were adjusted for age; education level; annual household income; marital status; alcohol consumption; smoking status; tea consumption; physical activity level; intake frequencies of red meat, fresh fruits, and fresh vegetables; family history of heart attack and stroke; body mass index; prevalent hypertension and diabetes mellitus; frequent use of sleep aid medications; frequency of snoring; and depression or anxiety symptoms. Models were stratified by sex, 10 study areas, and baseline age in 5-year intervals. The p values for trend were estimated by modeling the number of insomnia symptoms as a continuous variable. CI = confidence interval; CVD = cardio-cerebral vascular disease; HR = hazard ratio; IHD = ischemic heart disease.

Figure 2. HR (95% CI) of total CVD incidence according to insomnia symptoms stratified by baseline characteristics

All p values for interaction were <0.05 except for the interaction between daytime dysfunction (DDF) and body mass index (BMI). Model 3 was adopted to adjust for confounding factors. CI = confidence interval; CVD = cardio-cerebral vascular disease; DIMS = difficulties in initiating or maintaining sleep; EMA = early morning awakening; HR = hazard ratio.