Table 3. Dementia odds (Loge transformed) and all-cause mortality across sleep quality tertiles: Cox PH models, HRS 2006–2020a.
T1 | T2 | T3 | Psleepf | |
β ± SE | β ± SE | β ± SE | ||
Analysis A | ||||
Unweighted N | N = 2,286 | N =3,055 | N =1,377 | |
Reduced modelse | ||||
Model 1A: Hurd dementia | +0.082 ± 0.012d | +0.090 ± 0.018d | +0.080 ± 0.015d | 0.37 |
Model 1B: Expert dementia | +0.158 ± 0.016d | +0.153 ± 0.010d | +0.132 ± 0.016d | 0.018 |
Model 1C: LASSO dementia | +0.204 ± 0.018d | +0.200 ± 0.015d | +0.164 ± 0.019d | 0.009 |
Full modelse | N =2,146 | N =2,930 | N =1,292 | |
Model 2A: Hurd dementia | +0.119 ± 0.016d | +0.097 ± 0.013d | +0.066 ± 0.019c | <0.001 |
Model 2B: Expert dementia | +0.122 ± 0.015d | +0.093 ± 0.012d | +0.081 ± 0.019d | <0.001 |
Model 2C: LASSO dementia | +0.173 ± 0.020d | +0.144 ± 0.017d | +0.122 ± 0.028d | <0.001 |
Analysis B | ||||
Unweighted N | N =2,286 | N =3,055 | N =1,377 | |
Reduced models e | ||||
Model 1A: Hurd dementia | +0.801 ± 0.101d | +0.673 ± 0.068d | +0.505 ± 0.117d | 0.003 |
Model 1B: Expert dementia | +0.836 ± 0.094d | +0.746 ± 0.081d | +0.557 ± 0.107d | 0.003 |
Model 1C: LASSO dementia | +0.717 ± 0.097d | +0.767 ± 0.083d | +0.479 ± 0.103d | 0.014 |
Full modelse | ||||
Model 2A: Hurd dementia | +0.574 ± 0.097d | +0.397 ± 0.070d | +0.195 ± 0.136 | 0.001 |
Model 2B: Expert dementia | +0.649 ± 0.077d | +0.409 ± 0.097d | +0.291 ± 0.123b | 0.001 |
Model 2C: LASSO dementia | +0.559 ± 0.086d | +0.483 ± 0.081d | +0.136 ± 0.130 | 0.001 |
Abbreviations: HRS: Health and Retirement Study; PH: Proportional hazards; SE: Standard Error. aValues are β ± SE, with β representing Loge (HR) from Cox PH model for each exposure-outcome relationship. Cox PH models were conducted overall and across sex or race/ethnicity groups, accounting for sampling weights and sampling design complexity in multiple imputed data. All covariates are measured in 2006 unless stated otherwise. Analysis A included continuous forms of dementia probability (Loge (odds of dementia)); while Analysis B included binary dementia status (3 algorithms). T1 corresponded to the first tertile of the “poor sleep quality” score (Mean ± SD: 0.49 ± 0.50, range: 0–1); T2 corresponded to the second tertile: (Mean ± SD: 2.90 ± 0.81, range: 2–4); T3 corresponded to the third tertile (Mean ± SD: 6.13±1.19, range: 5–9). bP < 0.05; cP < 0.010; dP < 0.001 for null hypothesis that β = 0. eReduced models (Models 1A–1C) were adjusted for age at 2006, sex and race/ethnicity. Full models (Model 2A–2C) further adjusted the reduced model by all covariates described under the Covariates section, including socio-demographic, socio-economic, lifestyle and health-related factors. fP-value associated with 2-way interaction term between poor sleep quality tertile and the main exposure dementia odds (Loge transformed), in a model not stratified by “poor sleep quality” tertile.