TABLE 2.
Multivariate Linear Regression Model With Subjective Sleep Quality in the Jackson Heart Sleep Study
| Variable | Unadjusted Model |
Adjusted Model |
||||
|---|---|---|---|---|---|---|
| β | SE | Partial R2 | β | SE | Partial R2 | |
| Insomnia symptomsa | 0.096 | (0.007)**,b | 0.20 | 0.092 | (0.007)**,b | 0.21 |
| RLS symptoms | 0.395 | (0.076)**,b | 0.03 | 0.402 | (0.081)**,b | 0.07 |
| Sleep apnea (REI > 15) | 0.138 | (0.069)* | 0.00 | 0.213 | (0.075)**,b | 0.04 |
| Daytime sleepiness | 0.015 | (0.008)*** | 0.00 | 0.014 | (0.008)*** | 0.04 |
| Insomnia Dx | 0.606 | (0.213)*,b | 0.01 | 0.558 | (0.244)*,b | 0.04 |
| Sleep apnea Dx | 0.576 | (0.225)*,b | 0.01 | 0.634 | (0.261)*,b | 0.04 |
| RLS Dx | 0.723 | (0.335)*,b | 0.00 | 0.842 | (0.381)*,b | 0.04 |
| Sleep efficiency | −0.022 | (0.007)**,b | 0.01 | −0.017 | (0.007)*,b | 0.04 |
| Sleep fragmentation index | 0.009 | (0.004)*,b | 0.01 | 0.010 | (0.004)*,b | 0.04 |
| Sleep duration | −0.001 | (0)*** | 0.00 | −0.001 | (0.001)* | 0.04 |
| Sleep midpoint | 0.001 | (0)*,b | 0.01 | 0.001 | (0.001)*** | 0.04 |
| WASO | 0.004 | (0.001)*,b | 0.01 | 0.002 | (0.002) | 0.04 |
| Sleep latency | −0.011 | (0.023) | 0.00 | 0.013 | (0.025) | 0.03 |
| Psychosocial factors | ||||||
| Depressive symptoms | 0.030 | (0.004)**,b | 0.06 | 0.026 | (0.005)**,b | 0.08 |
| Anxiety | 0.030 | (0.004)**,b | 0.07 | 0.027 | (0.004)**,b | 0.09 |
RLS = restless legs syndrome; REI = respiratory event index; Dx = diagnosis; WASO = wake after sleep onset; WHIIRS = Women’s Health Initiative Insomnia Rating Scale. Adjusted model includes age, sex, education, income, employment. Exposures are modeled in separate models. Higher scores indicate worse sleep quality.
*
p ≤ .05.
**
p ≤ .01.
***
p ≤ .10.
a
Modified WHIIRS.
b
Below the statistically significant threshold: p ≤ .0310 and p ≤ .0276 for unadjusted and adjusted estimates, respectively, based on the Benjamini-Hochberg adjustment for multiple comparisons.