Long-chain omega-3 fatty acids, selenium, and mercury in relation to sleep duration and sleep quality: findings from the CARDIA study.

Abstract

To examine the associations of long-chain omega-3 polyunsaturated fatty acids (LCω3PUFA) intake with sleep quality and duration in a cohort of American young adults, and to explore whether the associations of interest are modified by selenium (Se) and/or mercury (Hg) status. The study sample consisted of 3964 men and women from the longitudinal Coronary Artery Risk Development in Young Adults (CARDIA) study, aged 25.0 ± 3.6 at baseline. Intake of LCω3PUFA was assessed using an interviewer-administered dietary history questionnaire at baseline (1985-1986), Y7 (1992-1993), and Y20 (2005-2006). Toenail Se and Hg concentrations were quantified at Y2 (1987-1988). The outcomes were self-reported sleep quality and sleep duration measured by one question for each at Y15 (2000-2001) and Y20. Generalized estimating equation was used to examine the association between cumulative average intake of LCω3PUFA and sleep measures. Restricted cubic spline was performed to explore the potential non-linear associations of interest. Se and Hg were dichotomized by their median values to examine the potential effect modification of Se and/or Hg. We did not observe any significant associations (linear or non-linear) of LCω3PUFA intake with either sleep quality or duration. Also, no significant association was observed in any subgroup classified by toenail Se and/or Hg concentrations. Similarly, sensitivity analysis indicated that the null associations between LCω3PUFA intake and sleep quality or duration persisted across subgroups classified by race, gender, obesity, or having small children. Findings from this longitudinal analysis did not support the hypothesis that LCω3PUFA intake is associated with sleep quality or sleep duration.