Table 2.
Independent and joint associations of sleep duration and physical activity with cardiovascular disease prevalence.
| Exposure | Model 1: Unadjusted | Model 2: Multivariable Adjusted | Model 3: IPTW Analysis |
|---|---|---|---|
| OR (95% CI) | OR (95% CI) | OR (95% CI) | |
| A. Independent associationsa | |||
| Sleep duration | |||
| Normal (6–8 h) | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Short (<6 h) | 1.58 (1.54–1.63) | 1.26 (1.21–1.30) | 1.26 (1.22–1.30)b |
| Long (>8 h) | 1.83 (1.76–1.90) | 1.19 (1.14–1.24) | 1.17 (1.13–1.21)b |
| Physical activity | |||
| Active | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Inactive | 2.14 (2.09–2.19) | 1.08 (1.05–1.11) | 1.11 (1.09–1.14)c |
| B. Joint associationsd | |||
| Sleep and physical activity | |||
| Normal sleep and active | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) |
| Normal sleep and inactive | 2.14 (2.08–2.20) | 1.10 (1.06–1.13) | 1.12 (1.09–1.15) |
| Short sleep and active | 1.56 (1.50–1.63) | 1.28 (1.22–1.35) | 1.28 (1.23–1.34) |
| Short sleep and inactive | 3.45 (3.30–3.61) | 1.33 (1.26–1.41) | 1.38 (1.31–1.44)e |
| Long sleep and active | 1.87 (1.78–1.96) | 1.23 (1.16–1.30) | 1.18 (1.12–1.24) |
| Long sleep and inactive | 3.82 (3.63–4.03) | 1.24 (1.17–1.32) | 1.29 (1.22–1.36) |
CI, confidence interval; CVD, cardiovascular disease; IPTW, inverse probability of treatment weighting; OR, odds ratio; Ref, reference group.
Model 1 (Unadjusted): Crude associations accounting only for complex survey weights.
Model 2 (Multivariable Adjusted): Adjusted for age, sex, race/ethnicity, education level, annual household income, and marital status.
Model 3 (IPTW Analysis): A doubly robust estimator model. Weights were derived from a propensity score model predicting physical inactivity conditional on all covariates (age, sex, race, education, income, marital status, BMI, smoking, binge drinking, insurance, general health, depression, diabetes, kidney disease, COPD, asthma, arthritis, and cancer). The outcome model was further adjusted for these covariates to ensure robustness against weight estimation errors.
aIndependent associations were estimated from separate models where sleep duration and physical activity were mutually adjusted for each other.
bEstimates for sleep duration in Model 3 represent associations within the IPTW-weighted pseudo-population balanced for physical activity propensity. The IPTW procedure was applied to physical inactivity only. CI derived from: Short Sleep (Estimate = 0.2289, SE = 0.0170); Long Sleep (Estimate = 0.1538, SE = 0.0184).
cEstimate represents the average treatment effect (ATE) of physical inactivity on CVD prevalence in the weighted population. CI derived from: Inactive (Estimate = 0.1053, SE = 0.0116).
dJoint associations categorize participants into six mutually exclusive groups based on sleep duration and physical activity status.
eThis joint exposure category represents the highest-risk group and is the primary focus of the main analysis.