Table 2. Poor sleep quality, dementia odds (Loge transformed) and all-cause mortality: Cox PH models, overall, by sex and by race, HRS 2006–2020a,e.
Overall | Male | Female | White | Black/Hispanic/Others | Psex | Prace | |
β ± SE | β ± SE | β ± SE | β ± SE | β ± SE | |||
Unweighted N | N = 6,718 | N = 2,854 | N = 3,864 | N = 5,440 | N = 1,278 | ||
Analysis A | |||||||
Reduced models | |||||||
Model 1A: Poor sleep quality | +0.022 ± 0.007c | +0.030 ± 0.010c | +0.016 ± 0.009 | +0.026 ± 0.007d | −0.009 ± 0.015 | 0.27 | 0.021 |
Model 1B: Hurd dementia | +0.085 ± 0.010d | +0.103 ± 0.005d | +0.082 ± 0.011d | +0.086 ± 0.011d | +0.082 ± 0.018d | 0.17 | 0.59 |
Model 1C: Expert dementia | +0.151 ± 0.009d | +0.144 ± 0.005d | +0.156 ± 0.010d | +0.156 ± 0.009d | +0.114 ± 0.014d | 0.69 | 0.001 |
Model 1D: LASSO dementia | +0.194 ± 0.011d | +0.180 ± 0.014d | +0.207 ± 0.014 d | +0.197 ± 0.011d | +0.174 ± 0.018d | 0.84 | 0.028 |
Full models | N = 6,368 | N = 2,653 | N = 3,715 | N = 5,204 | N = 1,164 | ||
Model 2A: Poor sleep quality | −0.033 ± 0.009c | −0.016 ± 0.013 | −0.044 ± 0.012c | −0.030 ± 0.010c | −0.057 ± 0.022c | 0.099 | 0.032 |
Model 2B: Hurd dementia | +0.098 ± 0.010d | +0.080 ± 0.015d | +0.114 ± 0.012d | +0.100 ± 0.011d | +0.084 ± 0.017d | 0.99 | 0.041 |
Model 2C: Expert dementia | +0.101 ± 0.009d | +0.075 ± 0.013d | +0.118 ± 0.011d | +0.108 ± 0.009d | +0.065 ± 0.016d | 0.56 | 0.002 |
Model 2D: LASSO dementia | +0.150 ± 0.013d | +0.121 ± 0.018d | +0.172 ± 0.016d | +0.153 ± 0.014d | +0.125 ± 0.024d | 0.74 | 0.048 |
Analysis B | |||||||
Reduced models | N = 6,718 | N = 2,854 | N = 3,864 | N = 5,440 | N = 1,278 | ||
Model 1A: Poor sleep quality | +0.072 ± 0.023c | +0.099 ± 0.034c | +0.052 ± 0.028 | +0.088 ± 0.023d | −0.026 ± 0.048 | 0.26 | 0.017 |
Model 1B: Hurd dementia | +0.675 ± 0.057d | +0.674 ± 0.080d | +0.684 ± 0.081d | +0.706 ± 0.058d | +0.563 ± 0.116d | 0.73 | 0.012 |
Model 1C: Expert dementia | +0.731 ± 0.055d | +0.716 ± 0.063d | +0.743 ± 0.078d | +0.745 ± 0.062d | +0.673 ± 0.095d | 0.90 | 0.16 |
Model 1D: LASSO dementia | +0.681 ± 0.058d | +0.708 ± 0.069d | +0.679 ± 0.083d | +0.696 ± 0.064d | +0.658 ± 0.092d | 0.36 | 0.069 |
Full models | N = 6,368 | N = 2,653 | N = 3,715 | N = 5,204 | N = 1,164 | ||
Model 2A: Poor sleep quality | −0.069 ± 0.027b | −0.011 ± 0.037 | −0.110 ± 0.033c | −0.056 ± 0.030 | −0.147 ± 0.067b | 0.036 | 0.037 |
Model 2B: Hurd dementia | +0.400 ± 0.065d | +0.304 ± 0.088d | +0.461 ± 0.088d | +0.428 ± 0.070d | +0.300 ± 0.117b | 0.78 | 0.055 |
Model 2C: Expert dementia | +0.457 ± 0.057d | +0.369 ± 0.069d | +0.510 ± 0.089d | +0.474 ± 0.066d | +0.422 ± 0.096d | 0.80 | 0.13 |
Model 2D: LASSO dementia | +0.415 ± 0.067d | +0.448 ± 0.073d | +0.410 ± 0.097d | +0.416 ± 0.077d | +0.437 ± 0.102d | 0.26 | 0.37 |
Abbreviations: HRS: Health and Retirement Study; PH: Proportional hazards; SE: Standard Error. aValues are β ± SE, with β representing LogeHR from Cox PH model for each exposure-outcome relationship. Cox PH models were conducted overall and across sex or race/ethnicity groups, taking into account 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)) and continuous “poor sleep quality” score; while Analysis B included binary dementia status (3 algorithms) and “poor sleep quality tertile”. bP < 0.05; cP < 0.010; dP < 0.001 for null hypothesis that β = 0. eReduced models (Models 1A–1D) were adjusted for age at 2006, sex and race/ethnicity. Full models (Model 2A-2D) further adjusted the reduced model by all covariates described under the Covariates section, including socio-demographic, socio-economic, lifestyle and health-related factors.