Skip to main content
. 2007 Oct 1;30(10):1245–1253. doi: 10.1093/sleep/30.10.1245

Table 7.

Joint Effects of Sleep Length and Sleep Quality in 1981 on Total Mortality: Hazard Ratios (95% Confidence Intervals) for Mortality in 1982–2003. Hazard Ratios with 95% Confidence Intervals Given. Percentage of Subjects in Each Sleep Behavior Combination Given for Men and Women

Sleep Length* Sleep Quality: Sleeping Percentage
Fully-Adjusted Model Hazard Ratios Men N = 9529 Fully-Adjusted Model Hazard Ratios Women N = 10265
Men Women
Short Well 5.0 3.7 1.32 (1.03, 1.68) 1.35 (1.02, 1.78)
Short Fairly well 6.3 5.6 1.45 (1.20, 1.76) 1.17 (0.94, 1.47)
Short Fairly poorly/Poorly 4.0 4.1 1.33 (1.05, 1.68) 1.08 (0.85, 1.36)
Average (reference) Well 30.8 27.1 1.00 1.00
Average Fairly well 34.8 33.7 1.09 (0.94, 1.26) 0.93 (0.79, 1.11)
Average Fairly poorly/Poorly 4.5 4.4 1.25 (0.99, 1.58) 0.97 (0.76, 1.25)
Long Well 7.5 10.4 1.42 (1.17, 1.73) 1.16 (0.93, 1.44)
Long Fairly well 6.3 10.0 1.26 (1.02, 1.56) 1.08 (0.88, 1.32)
Long Fairly poorly/Poorly 0.9 1.0 1.18 (0.80, 1.73) 1.39 (0.92, 2.09)
*

short = < 7 hours, average = 7–8 hours, and long = > 8 hours.

Covariates measured in 1981 (in addition to age) in this fully-adjusted model: education, marital status, working status, social class, BMI, smoking status, binge drinking, grams of alcohol consumed daily, conditioning physical activity, life satisfaction, and use of hypnotics and/or tranquilizers.