Delayed Mid-Sleep Time Associated With Weight Gain While Controlling for Eating Behaviors and Attention Deficit Hyperactivity Disorder Symptoms During the COVID-19 Pandemic

Article information

Psychiatry Investig. 2023;20(8):768-774

Publication date (electronic) : 2023 August 11

doi : https://doi.org/10.30773/pi.2022.0326

Ali Kandeger¹, Omer Faruk Uygur², Seockhoon Chung^,3, Elif Yavuz¹, Yavuz Selvi¹

¹Department of Psychiatry, Faculty of Medicine, Selçuk University, Konya, Türkiye

²Department of Psychiatry, Faculty of Medicine, Atatürk University, Erzurum, Türkiye

³Department of Psychiatry, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea

Correspondence: Seockhoon Chung, MD, PhD Department of Psychiatry, Asan Medical Center, University of Ulsan College of Medicine, 86 Olympic-ro 43-gil, Songpa-gu, Seoul 05505, Republic of Korea Tel: +82-2-3010-3411, Fax: +82-2-485-8381, E-mail: schung@amc.seoul.kr

Received 2022 November 9; Revised 2023 April 12; Accepted 2023 May 31.

Abstract

INTRODUCTION

The coronavirus disease-2019 (COVID-19) pandemic was the cause of a stressful period worldwide due to many factors, such as the risk of infection and death, isolation, risky working conditions, financial/job loss, and grieving circumstances [1-3]. Published meta-analyses have shown that individuals’ lifestyles changed and that depression, anxiety, post-traumatic stress disorder symptoms, and many negative psychological outcomes increased in societies during the COVID-19 outbreak [4,5].

Consistent data have also been presented regarding the negative effects of eating attitude changes and weight control problems on individuals during the COVID-19 pandemic [6]. Recent meta-analyses have shown that 65% of eating disorders became symptomatic and 52% of obese individuals gained weight during the pandemic. In addition, the pandemic period was strongly associated with an increase in disordered eating attitudes and weight gain [7]. Previous studies have elicited that disordered eating attitudes (e.g., dieting, oral control, and bulimic behaviors) aimed at weight control or weight loss and which did not reach the diagnostic criteria of an eating disorder were more common in females [8]. Many researchers usually assess eating disorders with the Eating Attitude Test (EAT). Three main eating disorder behaviors were defined as subscales in EAT. These include dieting (e.g., “Am terrified about being overweight”), oral control (e.g., “Avoid eating when I am hungry”), and bulimic behavior (e.g., “Find myself preoccupied with food”) [8].

In addition to eating and weight issues, sleep-wake cycles and sleep habits were affected during the outbreak. A meta-analysis of 54,231 individuals reported that 35.7% of the general population had sleep problems [9]. The shortening of sleep duration and delayed sleep phase is associated with metabolic problems and weight gain. It has also been reported that individuals who go to bed late and wake up late (e.g., evening chronotype) have more disordered eating attitudes and are more prone to weight gain. Cumulative evidence unveiled that circadian misalignment reduces energy expenditure by approximately 3%, alters levels of appetite hormones, and promotes unhealthier food choices compared to adequate sleep [10]. The only study to date that examined the relationship between sleep phase and weight gain during the pandemic showed that having a late chronotype was associated with weight gain and obesity in 30,275 people [11]. Mid-sleep time is often operationalized as mid-sleep, that is, a midpoint between sleep onset and wake-up. It can be used to assess an individual’s chronotype. A later mid-sleep time on free days indicates an evening chronotype, and an earlier mid-sleep time on free days shows the morning chronotype [12]. We calculated using the following formula for the mid-sleep scores: mid–sleep=sleep onset+sleep duration/2. Therefore, we can estimate the effect of chronotypes on individuals’ weight changes during the pandemic by using mid-sleep time.

Attention deficit hyperactivity disorder (ADHD) is a childhood-onset neurodevelopmental disorder characterized by attention deficit, overactivity, and impulsivity. In addition to these core symptom clusters, it is known that individuals with ADHD experience emotional dysregulation, anxiety, depression, and sleep disorders more than the general population does [13]. Since the nature of ADHD is as a 24-hour disorder, individuals with ADHD have less sleep time, a delayed sleep phase, and are more prone to the evening chronotype. ADHD symptoms are related to sleep disturbances, circadian disruption, disordered eating attitudes, and obesity [14,15]. Accumulated evidence suggests a relationship between ADHD symptoms and eating problems by both genetic and nonshared environmental factors [16-18]. ADHD symptoms, especially impulsivity, are particularly associated with overeating, bulimic, and binge eating behaviors. Shared genetic factors explain an important variance in this association, especially in females [19]. Moreover, recent longitudinal studies found that ADHD has a causal role in obesity via contributing to weight gain [20].

As emphasized above, the pandemic period has led to an alteration in sleep-wake cycles and eating habits and an increase in insomnia, delayed sleep phase, weight gain, and obesity in the general population. One study associates delayed sleep phase with weight gain; however, there is no study in the literature to date that associates these factors with ADHD symptoms [11]. Our aim in this study was to examine the relationship of weight change in the last year with sleep and eating behaviors and ADHD symptoms in young women.

METHODS

This online study with a cross-sectional design was conducted among female university students at Selçuk University Central Campus between January 8, 2021 and January 11, 2021. The study protocol was approved by the Selçuk University Ethics Committee (Decision Number: 2021/369), and necessary permissions were obtained from the Selçuk University Rectorate and the Ministry of Health of the Republic of Türkiye for the conduct of the study.

RESULTS

The sample consisted of 578 female university students whose mean age was 21.65±3.12 years; their demographic characteristics are presented in Table 1. According to weight change last year, we divided all participants into three groups as participants who gained weight (WGs), participants whose weight did not change (nWCs), and participants who lost weight (WLs). Of all participants, 42.73% (n=247) were in the WGs group, 30.79% (n=178) were in the nWCs group, and 26.47% were in the WLs (n=153) group.

Table 1.

Demographic characteristics of participants

When comparing means and standard deviations of measurements of sleep/eating behavior and ADHD symptoms between groups based on weight change (Table 2), bed time (F=4.28; p=0.014), wake time (F=6.52; p=0.002), and mid-sleep time (F=6.41; p=0.002) were more delayed in the WGs group than in both the nWCs and WLs groups. WGs had the highest bulimic behaviors scores (F=4.59; p=0.011) and the lowest oral control behaviors scores (F=9.70; p<0.001).

Table 2.

Comparison of means and standard deviations of measurements of sleep/eating behavior and ADHD symptoms between groups based on weight change^†

In the correlation analysis, mid-sleep time was positively correlated with ADHD symptoms (r=0.08; p<0.05) and weight change (r=0.14; p<0.05). In addition, bulimic behaviors and dieting behaviors were positively correlated with both childhood and current ADHD symptoms. The Pearson product-moments correlation coefficients are presented in Table 3.

Table 3.

Pearson product-moments correlation coefficients

In the first step of the hierarchical regression analysis to determine factors associated with weight change (Table 4), childhood and current ADHD symptoms did not show an association with weight change. In the second step, sleep-wake parameters were added to the analysis, and mid-sleep time was a strong predictor of weight gain (β=4.71, t=2.18, p=0.03). In the third step, in which disordered eating behaviors were added to the analysis, bulimic behaviors (β=0.20, t=3.20, p=0.001) were associated with weight gain, and oral control behaviors (β=-0.11, t=-3.24, p=0.001) were associated with weight loss. Additionally, multicollinearity was evaluated in linear regression analysis and the fact that “VIF” values are less than 5 and “tolerance” values are not greater than 5 and not less than 0.2 indicate that there is no multicollinearity that may adversely affect the power of the regression model [32].

Table 4.

Hierarchical regression analysis of weight change (N=578)

DISCUSSION

The aim of this study was to examine the relationship of weight change with the sleep-wake cycle, disordered eating behaviors, and ADHD symptoms in young women during the COVID-19 pandemic. Bulimic behaviors were positively associated with weight gain, while oral control behaviors were negatively associated. Sleep time, mid-sleep time, and wake time were delayed significantly in WGs than in nWCs and WLs. Additionally, delayed mid-sleep time was associated with weight gain independent of disordered eating behaviors and both childhood and current ADHD symptoms.

At the data collection process of our study, COVID-19-related home confinement conditions were being implemented in Türkiye, and all of the universities, including Selçuk University, had transitioned to a distance education system for over one year. Some of the participants (42.7%; n=247) reported gaining weight in the last year. Although studies conducted during the COVID-19 outbreak have reported a wide range of weight gain rates (11.1%–72.4%), the rate of participants reporting weight gain decreased as the participants’ mean age increased [6]. In a recent study conducted in Türkiye, 35% of 1,036 participants with a mean age of 33 years reported that they gained weight [33]. More participants in our study may have reported weight gain since all of the participants were female and younger (mean age: 21.65 years).

When the participants were divided into three groups according to the weight change they reported in the last year, the WGs had a more delayed sleep phase (sleep, midsleep, and wake time) than did the other two groups. Moreover, delayed mid-sleep time but not total sleep time was associated with the weight gain score even when all possible associated variables were controlled in the regression analysis. Cumulative evidence showed that the late sleep phase predicts weight gain and obesity with many mediating factors (e.g., impulsivity, disordered eating behaviors, and metabolism) in both children and adults, even at constant calorie intake [34-37]. However, it may be more important to investigate this relationship during stressful periods, such as a pandemic, when even the general population is affected. The only study examining this relation during the COVID-19 period found an association between delayed sleep phase and weight gain independent of physical activity and meal changes; our findings support this relationship [11]. When disordered eating attitudes were included in the analyses, not surprisingly, weight gain was positively associated with bulimic behavior and was negatively associated with oral control behavior. These associations were elucidated in the literature previously; however, this is the first time ADHD symptoms were included in the analysis.

Studies between ADHD core symptoms and disordered eating attitudes have not shown consistent associations with attention deficit, although the relationship between hyperactivity/impulsivity and binge eating has been elucidated. A recent study found that the ADHD core symptoms, attention deficit and hyperactivity/impulsivity, were associated with binge and restrictive eating, especially if the individual was in a negative mood [38]. This is notable because ADHD symptoms have been shown to be associated with bulimic and dieting behaviors as in our findings [39]. These disordered eating attitudes were reported to be mediators in the relationship between ADHD and weight gain; however, this relation could not be demonstrated in this study [20]. One possible reason for this may be that the study only investigated weight change over the past year in a cross-sectional design. Additionally, participants were not examined for ADHD diagnosis. Thus, participants’ high ADHD scores did not warrant that they had ADHD neurobiology. A longitudinal examination of weight changes in individuals with ADHD may yield stronger findings about the relationship between ADHD and weight gain and obesity.

This online survey study that prevents missing data was conducted with a large and homogeneous sample, the hypothesis was tested clearly, and important findings contributing to the literature were yielded. However, there were some limitations. The first limitation of our study was its cross-sectional design, which cannot clearly reveal temporal associations. Second, objective data collection methods were not used in our study, and all of the data, including weight change, were collected as self-reported answers via an online survey. Third, anxiety and depression scores, which may affect people’s sleep/eating behaviors and weight changes, were not included in the analysis. These symptoms can be considered confounding factors in the relationships elicited here. Fourth, this study did not consider lifestyle changes and psychological factors as confounders related with the COVID-19 outbreak. Fifth, the population of this study include only young female adult, it is not easy to generalization. Finally, we cannot confirm whether the results of this study were direct effects of the pandemic. We need to consider that the association should remain significant regardless of the pandemic.

In conclusion, WGs had a more delayed sleep phase than did nWCs and WLs in the one-year period during the COVID-19 outbreak. ADHD symptoms were positively correlated with diet and bulimic behaviors in the correlation analysis; however, they were not associated with weight change in the regression analysis. Bulimic behaviors were associated with weight gain, oral control behaviors were associated with weight loss, and the most prominent finding was that mid-sleep time was a strong predictor for weight gain, independent of all variables. Chronotherapeutic approaches that regulate sleep-wake rhythm may facilitate weight control, especially in stressful periods such as the COVID-19 outbreak.

Article information Continued