Sleep duration and cancer risk in women

Susan Hurley, Debbie Goldberg, Leslie Bernstein, Peggy Reynolds

The objective was to conduct an analysis of sleep duration and risk of selected site-specific and groups of cancer among a large prospective cohort of California women. The study population was comprised of 101,609 adult females participating in the California Teachers Study. All sites of invasive cancer prospectively diagnosed from baseline (1995-1996) through 2011 were identified through linkage to the California Cancer Registry (n = 12,322). Site-specific analyses focused on the following cancers: breast (n = 5,053), colorectal (n = 983), lung (n = 820), melanoma (n = 749), and endometrial (n = 957). Additionally, we evaluated a group of estrogen-mediated cancers consisting of breast, endometrial, and ovarian cancer (n = 6,458). Sleep duration was based on self-report of average time sleeping in the year prior to baseline. Cox proportional hazard models were used to calculate adjusted hazard ratios and 95 % confidence intervals (HRs, 95 % CI). Point estimates for all sites and site-specific cancers generally were near or below one for short sleepers (<6 h/night) and above one for long sleepers (10+ h/night); confidence intervals, however, were wide and included unity. Compared to average sleepers (7-9 h/night), long sleepers had an increased risk of the group of estrogen-mediated cancers (HR 1.22, 95 % CI 0.97-1.54, p (trend) = 0.04). These analyses suggest that longer sleep may be associated with increased risks of estrogen-mediated cancers. Further studies with more refined measures of sleep duration and quality are warranted.