Table 2.

Fully adjusted results of mixed-effect logistic regressions predicting the effect of family support on sleep problems

	Model 1: Sleep Dysfunction (≧1 sleep problem)	Model 2: Sleep Duration (< 6 h of sleep)	Model 3: Sleep Quality (poor sleep quality)
	Risk Ratio (95% CI)
Family support (z-score)	0.94 (0.90 - 0.98)**	0.88 (0.81 - 0.95)**	0.92 (0.87 - 0.98) *
Friend support (z-score)	0.98 (0.01 - 1.03)	0.95 (0.88 - 1.04)	0.99 (0.92 - 1.06)
Gender & sexuality
Homosexual man	Reference group
Homosexual woman	0.94 (0.79 - 1.09)	0.75 (0.52 - 1.07)	1.01 (0.78 - 1.30)
Bisexual man	0.95 (0.81 - 1.11)	1.09 (0.77 - 1.48)	1.00 (0.75 - 1.28)
Bisexual woman	1.15 (1.00 - 1.29)*	1.11 (0.83 - 1.47)	1.22 (0.97 - 1.50)
Other man	1.03 (0.86 - 1.21)	1.13 (0.78 - 1.57)	0.99 (0.73 - 1.30)
Other woman	1.15 (1.00 - 1.31)	1.52 (1.13 - 1.98) **	1.30 (1.01 - 1.62)*
Neighbourhood cohesion	0.94 (0.89 - 1.00)	0.82 (0.72 - 0.93)**	0.88 (0.88 - 0.97)*

All models above included additional controls for region-level fixed-effect, year fixed-effect, wave fixed-effect, household fixed-effect, ethnicity, marital status, age (in natural log units), personal income (in £1000), highest qualification, SF-12 PCS, chronic health problems, clinical depression, frequent use of sleep aid, and substance use

*p < 0.05, **p < 0.011