Abstract MP082: Sleep Duration and Insomnia as Risk Factors for Coronary Heart Disease among Postmenopausal Women: Findings from the Women's Health Initiative

Abstract

Background Several prospective studies have reported a positive association for both long and short sleep duration with coronary heart disease (CHD) risk. Insomnia may also be a CHD risk factor. Our aims were to identify whether sleep duration and insomnia symptoms are associated with greater incident CHD among postmenopausal women. Methods Participants were 86,329 postmenopausal women, aged 50-79 at enrollment into the Women's Health Initiative (WHI) observational study, who reported information regarding sleep habits at baseline (1993-1998) and were followed for first occurrence of a coronary event. Cox proportional hazards models were developed using sleep duration and insomnia as the primary exposures. Person-years of follow up were calculated from the screening visit through August 2009. Results Compared to mid-range sleep duration (7-8 hours) women reporting shorter sleep (≤5 hours) and longer sleep (≥10 hours) had higher incident CHD (HR=1.21, 95% CI 1.09-1.33; HR=1.51, 95% CI 1.08-2.11), respectively, after adjusting for age and race. Only longer sleep remained significantly associated with incident CHD in fully adjusted models (HR=1.53, 95% CI 1.09-2.14). Women that scored high (≥9) for insomnia on the WHI Insomnia Rating Scale (WHIIRS) demonstrated the highest risk of CHD (HR=1.35, 95% CI 1.24-1.47, age- and race-adjusted; HR=1.18, 95% CI 1.08-1.30, fully adjusted models). When the analyses are stratified by the WHIIIRS, among those with high insomnia scores (≥9), long sleep almost doubled the risk of CHD (HR=1.97, 95% CI 1.18-3.30) versus mid-range sleep duration, whereas among those with lower WHIIRS scores (<9), long sleepers have a relatively lower increased risk of CHD (HR=1.49, 95% CI 1.08-2.06). Conclusions Women reporting ≥10 hours of sleep per night with high insomnia scores had a higher risk for incident CHD. A significant interaction between sleep duration and insomnia was observed (p<0.02), and both long sleep duration and a high insomnia score were associated with greater risk of CHD. Table 1. Cox proportional hazards models - sleep duration and incident CHD * among WHI participants stratified by level of insomnia Cases N High level of perceived insomnia (WHIIRS ≥ 9) Model adjusted for age, race Model fully adjusted † HR 95% CI P-value HR 95% CI P-value Sleep time ≤5 h 347 4580 1.16 1.02, 1.32 0.03 1.05 0.93, 1.18 0.46 6 h 679 9610 1.08 0.97, 1.20 0.14 1.00 0.91, 1.10 0.98 7-8 h (ref) 734 10919 REF REF REF REF REF REF 9 h 48 517 1.38 1.03, 1.85 0.03 1.26 0.97, 1.65 0.08 ≥10 h 12 114 1.98 1.12, 3.51 0.02 1.97 1.18, 3.30 0.01 Cases N Low level for perceived insomnia (WHIIRS < 9) Model adjusted for age, race Model fully adjusted † HR 95% CI P-value HR 95% CI P-value Sleep time ≤5 h 130 2209 1.12 0.96, 1.31 0.16 0.92 0.77, 1.09 0.34 6 h 768 13,416 0.99 0.92, 1.07 0.88 0.94 0.87, 1.02 0.12 7-8 h (ref) 2418 41347 REF REF REF REF REF REF 9 h 171 2934 1.06 0.93, 1.21 0.38 0.98 0.85, 1.13 0.73 ≥10 h 23 357 1.67 1.22, 2.28 0.01 1.49 1.08, 2.06 0.02 * Outcome includes myocardial infarction, CHD death, percutaneous transluminal coronary angioplasty, coronary artery bypass grafting or hospitalized angina † Model adjusted for age, race, education, income, smoking, BMI, physical activity, alcohol intake, depression, diabetes, high blood pressure, cholesterol medication, comorbid conditions