Transition of nighttime sleep duration and sleep quality with incident cardiovascular disease among middle-aged and older adults: results from a national cohort study | Archives of Public Health

Table of Contents

Study design and participants

The CHARLS aims to collect high-quality data representative of the middle-aged and older population in China to analyze issues related to the aging population. The national baseline survey of CHARLS was conducted in 2011, employing a multi-stage probability sampling method to recruit more than 17,000 individuals from about 10,000 households across 150 county-level units and 450 village-level units. In the current analysis, we utilized data collected in the 2011, 2015, and 2018 waves. In the 2011 baseline survey, a total of 17,705 participants were recruited. We excluded participants who were under 45 years of age or lacked age information (N = 474), those with baseline CVD or missing CVD information (N = 7,129), and those without sleep measurement data (N = 1,199). Additionally, 998 participants who were lost to follow-up were further excluded. Ultimately, as shown in Figs. 1, 7,905 eligible participants were included in the current analysis. The ethics of CHARLS study was approved by the Biomedical Ethics Review Committee of Peking University (Approval number: IRB00001052-11014 and IRB00001052-11015). All participants have signed a written informed consent.

Exposure

Nighttime sleep duration and sleep quality data were collected in both the 2011 and 2015 waves of the survey. Participants were asked the following question: “During the past month, how many hours of actual sleep did you get at night (average hours for one night)?” Sleep duration was classified as optimal (6–8 h) or non-optimal (< 6 or > 8 h) [10]. Sleep quality was assessed by asking about the number of days of restless sleep (such as insomnia, waking up early, etc.) within the last week. The responses included four categories: (1) rarely or none of the time (< 1 day); (2) some or a little of the time (1–2 days); (3) occasionally or a moderate amount of the time (3–4 days); (4) most or all of the time (5–7 days), each assigned scores from 1 to 4. Sleep quality was further categorized as good (< 3 points) and poor (≥ 3 points). The changing patterns in nighttime sleep duration between 2011 and 2015 were used to identify four groups: consistent optimal, optimal to non-optimal, non-optimal to optimal, and consistent non-optimal. For sleep quality, the four groups were identified as: consistent good, good to poor, poor to good, and consistent poor.

Outcome

The primary outcome of the study was CVD, which included self-reported physician diagnoses of heart disease and/or stroke. The incidence of CVD was ascertained by asking the question in 2018 wave [11], “Did your doctor tell you that you have been diagnosed with a heart attack, angina pectoris, coronary heart disease, heart failure, or other heart problem?“, and “Did your doctor tell you that you were diagnosed with a stroke?“. If the participant gave at least one positive answer, they were considered to have CVD, and the time of diagnosis was also recorded.

Measurement of covariates

Sociodemographic and lifestyle information was collected using a standardized questionnaire. Current marital status was categorized into married and cohabiting, married but temporarily separated, and single. Residence was divided into rural and urban areas. Educational attainment was classified into three levels: primary school or below, middle school, and high school or above. Smoking status was categorized into never smokers and those with a history of smoking. Drinking status was divided into those who drink more than once a month, those who drink once a month or less, and non-drinkers. Household income was grouped into quartiles. Physical activity was defined as engaging in 30 min of moderate activity at least five times per week, or 20 min of vigorous activity at least three times per week [12]. Height was measured with the individual standing erect on the floor board of the stadiometer. Waist circumference was assessed while the individual was standing. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters. Systolic and diastolic blood pressures were measured using a digital blood pressure monitor (Omron™ HEM-7200, Dalian, China). Hypertension was diagnosed if the participant had a previous diagnosis of hypertension, a systolic/diastolic blood pressure ≥ 140/90 mmHg, or was taking antihypertensive medications. Diabetes was diagnosed if the participant had a previous diagnosis of diabetes, a fasting blood glucose level ≥ 7 mmol/L, a glycated hemoglobin level ≥ 6.5%, or was taking glucose-lowering medications.

Blood samples were stored at -70 °C. Bioassays were performed at national or county Centers for Disease Control that underwent uniform quality control. Blood glucose, creatinine levels, total cholesterol, triglycerides, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol were determined using the enzymatic colorimetric method. Hypersensitive C-reactive protein (hs-CRP) was measured using an immunoturbidimetric assay.

Statistical analysis

The study population was divided into two groups based on whether participants developed CVD. Descriptive statistics were expressed as mean ± standard deviation for normally distributed continuous data, and as median (interquartile range) for skewed continuous data. Differences between the groups were assessed using Student’s t-test or Kruskal-Wallis H test. Categorical data were expressed as frequency (percentage) and compared using the chi-square test.

A robust Poisson regression model was employed to ascertain the impact of baseline nighttime sleep duration and sleep quality, and their changing patterns on the incidence of CVD, reported as incidence risk ratios (IRR) with 95% confidence intervals (CI). Three multivariable models were utilized: Model 1, which adjusted for age, sex, and BMI; Model 2, which further adjusted for education, residence, smoking, alcohol consumption, and household income; Model 3, which additionally adjusted for hypertension, diabetes, and physical activity. In addition to analyzing the individual effects of nighttime sleep duration and sleep quality, the joint effect of baseline nighttime sleep duration and sleep quality on the risk of CVD was also assessed.

Initially, we analyzed the individual and joint effects of baseline nighttime sleep duration and sleep quality on the risk of CVD. Then, we explored the longitudinal changing patterns of nighttime sleep duration and sleep quality on the risk of CVD using two methods. First, a dose-response analysis was employed to assess the effects of transitions in nighttime sleep duration and sleep quality on CVD, using a restricted cubic spline model with four knots to quantitatively visualize the continuous relationship between these transitions and CVD events. Second, using ‘consistent optimal’ and ‘consistent good’ as reference categories, the association of the other categories with CVD, based on the changing patterns in nighttime sleep duration and sleep quality, was also explored. Subgroup analyses and effect modification were conducted in both predefined and exploratory subgroups across various factors, including age (< 65/≥65 years), sex (male/female), obesity status (BMI < 24/≥24 kg/m²), diabetes status (yes/no), hypertension status (yes/no), and physical activity (yes/no). We conducted several sensitivity analyses to test the robustness of the results. First, logistic regression methods were used to clarify the association between changing patterns in nighttime sleep duration and sleep quality and CVD. Second, analyses excluding individuals with depressive symptoms were conducted. Finally, we included the 10-year CVD risk scores as a covariate for adjustment in the analysis. The method of the last observation carried forward, or the means and medians were used to interpolate the missing data (Supplementary Table 1). Stata (StataCorp LLC, version 15.0) and R software (version 4.2.2) were used for data analyses.

link