Table 4.
Multiple logistic regression analyses with AD-ND positivity as a dependent variable
| Sleep quality | Sleep duration | |||||||
|---|---|---|---|---|---|---|---|---|
| Model 1 | Model 2 | Model 1 | Model 2 | |||||
| aOR (95% CI)a | p | aOR (95% CI)a | p | aOR (95% CI) | p | aOR (95% CI) | p | |
| Young adulthood | 1.428 (0.426–4.786) | 0.564 | 1.423 (0.423–4.781) | 0.569 | 0.826 (0.628–1.088) | 0.174 | 0.823 (0.625–1.084) | 0.166 |
| Midlife | 1.989 (0.714–5.539) | 0.188 | 1.999 (0.708–5.641) | 0.191 | 0.679 (0.516–0.894) | 0.006 | 0.678 (0.514–0.893) | 0.006 |
| Current | 0.693 (0.236–2.036) | 0.504 | 0.677 (0.227–2.018) | 0.484 | 0.938 (0.745–1.182) | 0.588 | 0.939 (0.745–1.184) | 0.594 |
Model 1: adjusted for age, gender, education, APOE4 status, vascular risk score, use of sleep pill, Hamilton Depression Rating Score. Model 2: adjusted for covariates of model 1 + global Aβ deposition. Global Aβ deposition was natural log-transformed to achieve normal distribution
AD-ND Alzheimer’s disease signature region neurodegeneration, aOR adjusted odds ratio, CI confidence interval, Aβ beta-amyloid, APOE4 apolipoprotein E ε4
aGood sleep quality was taken as a reference, so aOR represents the likelihood of being AD-ND positive for bad sleep quality compared with good one