Objectives: This study aimed to compare the efficacy of dietary intake of eicosapentaenoic acid (EPA; 20:5 ω-3) and docosahexaenoic acid (DHA; 22:6 ω-3) on very short sleep duration (<5 h/night) in adults.
Methods: The bootstrap method was used in the multinomial logistic regression to estimate the ORs and corresponding 95% confidence intervals (CIs) of very short sleep duration. We used rolling window method to analyze the effects of EPA and DHA dietary intakes on very short sleep durations in men and women over age. To illustrate the stability of the results for the selected window width, we built a shiny application.
Results: Compared to the first quartile, the mean ORs of EPA intake on very short sleep duration and the corresponding 95% CIs for the second, third and fourth quartiles of EPA intake among men under 32 years old were 1.50 (0.56, 3.44) mg, 1.55 (0.59, 3.48) mg, and 3.99 (1.15, 10.01) mg, respectively. Among women over 44 years old, the ORs for DHA intake were 1.12 (0.81, 1.52) mg, 0.94 (0.68, 1.29) mg, and 0.62 (0.38, 0.98) mg for the second, third and fourth quartiles, respectively.
Conclusions: The associations of EPA and DHA with very short sleep duration are sex- and age-dependent. In males under the age of 32, a significant positive correlation exists between dietary EPA intake and very short sleep duration. For women above 44 years of age, an increase in DHA intake can notably ameliorate issues of very short sleep duration.
Keywords: DHA; EPA; Very short sleep duration; bootstrap methods; differential effects; gender differences; middle-aged and elderly; omega-3 fatty acids.