Associations between sleep duration and dietary quality: Results from a nationally-representative survey of US adults

Abstract

Insufficient sleep duration is a recognized determinant of cardiometabolic disease, with poor diet quality as a likely intermediate. Yet, inconsistencies in findings on sleep duration and diet quality among adult populations remain, particularly regarding a potential non-linear relationship. Thus, within a nationally representative survey of US adults (2011–2016 National Health and Nutrition Survey), we evaluated cross-sectional associations between sleep duration and diet quality. Self-reported typical sleep duration was categorized into four groups (≤5 h per night, 6 h, 7–8 h, and ≥9 h per night). Diet was assessed with two 24-h recalls (averaged), and diet quality was measured with the Healthy Eating Index-2015 (score range 0–100), which is comprised of 13 dietary components. Regression analyses that accounted for complex survey weights were used to compute multivariable-adjusted models of sleep duration categories and HEI-2015 overall scores, and individual HEI components. Within this nationally-representative sample of adults ≥18 years, 10% reported 5 or fewer hours of sleep, and 20% reported 6 h, while 12% obtained ≥9 h (long sleep duration). Compared to adults with 7–8 h of sleep, those with ≤5 h had 1.7 lower multivariable-adjusted HEI-2015 score (95% CI 0.6, 2.7), and those with ≥9 h had 1.2 lower adjusted HEI-2015 score (95% CI 0.2, 2.2). Upon examination of individual HEI-2015 components, short sleepers had lower consumption of healthier components (total fruits, whole fruits, total vegetables, greens and bean, whole grains, dairy, total protein, and seafood), while long sleepers had higher consumption of some unhealthier components (refined grains and added sugar) as well as some healthier components (dairy and total protein). In conclusion, within a nationally-representative US survey, a reverse U-shaped association was observed between sleep duration and dietary quality.

Introduction

Insufficient sleep is prevalent in the US adult population, with over one-third estimated to obtain <7 h per night (Liu et al., 2016). The high prevalence of inadequate sleep duration has implications for cardiometabolic disease, as a robust literature now links short sleep duration with obesity, type 2 diabetes, and metabolic syndrome(St-Onge et al., 2016a). Diet is one of the likely intermediate factors between short sleep duration and cardiometabolic disease(Jansen et al., 2018). Small experimental studies show that when adults are subject to restricted sleep (typically 5 or less hours), appetite hormone levels are altered in addition to higher total caloric intake and greater preference for highly-palatable foods(Hanlon et al., 2016; Schmid, Hallschmid, Jauch-Chara, Born, & Schultes, 2008; Spaeth, Dinges, & Goel, 2013). Nonetheless, experimental conditions represent short-term and often extreme conditions (e.g. complete sleep deprivation), necessitating the examination of this association over the longer-term in free-living adult populations.

Epidemiological investigations of sleep duration and diet conducted within US populations have reported associations between short sleep duration and higher total energy intake (Galli et al., 2013; Grandner, Kripke, Naidoo, & Langer, 2010; Patterson et al., 2014; Stamatakis & Brownson, 2008; Stern et al., 2014), higher fat and carbohydrate intake (Grandner et al., 2010; Grandner, Jackson, Gerstner, & Knutson, 2013; Kant & Graubard, 2014), and lower fiber intake(Grandner et al., 2013). A few other studies have focused on the diet quality rather than individual macronutrients or foods, which can provide clearer information on how sleep affects the diet as a whole. In the US, the Healthy Eating Index (HEI) is a widely used measure of diet quality that measures the extent to which an individual's diet aligns with national dietary recommendations (2015 – 2020 Dietary Guidelines for Americans, 2015). Studies in Hispanic Americans (Mossavar-Rahmani et al., 2017) and postmenopausal women (Stern et al., 2014) showed that short sleep duration was associated with lower HEI. In contrast, a subgroup analysis using nationally representative data of women within 5 years of childbirth (Xiao et al., 2016) reported no association between short sleep duration and diet quality. However, compared to adequate sleepers, women with long sleep duration (≥9 h) had lower diet quality. An association between long sleep duration and poor diet quality is not necessarily surprising, since long sleep duration has been associated with cardiometabolic disease risk (Jike, Itani, Watanabe, Buysse, & Kaneita, 2018). Yet, the mechanisms behind this association are even more unclear than the mechanisms for short sleep duration, with only limited evidence for a pathway through diet. In particular, it is unclear whether an association between long sleep duration and lower dietary quality extends to the entire representative survey of US adults or only to postpartum women. Thus, the aim of the present study was to examine the association between sleep duration and dietary quality (HEI-2015 and its components) in a nationally-representative survey of US adults.