Sleep and Sleep Hygiene of Adolescents With and Without ADHD During COVID-19

Objective: This study examined the effect of COVID-19 restrictions on the sleep and sleep hygiene of adolescents with ADHD and comorbid sleep problems and neurotypical adolescents (NT). Method: Four groups (two ADHD and two NT) of in total 100 adolescents (50 ADHD and 50 NT) were included. One ADHD and NT group were tested during many COVID-19 restrictions, the other during few. MANCOVAs were implemented with ADHD diagnosis and level of COVID-19 restrictions as independent and sleep outcomes (subjective and objective total sleep time (TST) and sleep onset latency (SOL), sleep and sleep hygiene problems) as dependent variables. Results: Both groups had a shorter objective TST during the week during many COVID-19 restrictions. Furthermore, adolescents with ADHD had a shorter subjective SOL during the weekend when there were many COVID-19 restrictions, while the SOL of the NT group stayed the same. Conclusion: COVID-19 restrictions are related to the sleep of adolescents with and without ADHD. However, causality and underlying mechanisms need further investigation.


Introduction
Restrictions related to the COVID-19 pandemic have significantly impacted the mental health of adolescents, including an increase of disorders such as post-traumatic stress disorder, depression, and anxiety disorders (Guessoum et al., 2020).Some groups were even more likely to experience negative outcomes; older adolescents, girls, and neurodiverse adolescents, including among others adolescents with attention deficit hyperactivity disorder (ADHD; Samji et al., 2022).Although underlying mechanisms are not yet fully understood, this might be due to the susceptibility of these groups to mental health problems.In neurodiverse adolescents specifically, more anxiety, depression, and behavior problems were reported, possibly due to their higher levels of psychological distress (Samji et al., 2022).Also, when directly comparing youth with and without ADHD, those with ADHD were found to be more likely to report an increase in ADHD-related symptoms, such as inattention, hyperactivity/impulsivity, and oppositional and defiant symptoms (Breaux et al., 2021).Possible factors that are responsible for these more exaggerated negative effects in adolescents with ADHD may be their difficulties with emotion regulation, as this has been found to predict mental health functioning during the pandemic (Breaux et al., 2021), their increased susceptibility to distress due to the loss of daily routines (Jefsen et al., 2021), and their loss of a peer network due to the restrictions regarding social gatherings and activities.
Another factor that was affected by the COVID-19 pandemic is sleep (Kronk et al., 2023), with the majority of children and adolescents reporting to have disturbed sleep during the COVID-19 pandemic (Becker & Gregory, 2020;Richter et al., 2023;Sharma et al., 2021).It has been suggested that COVID-19 restrictions have a negative effect on sleep hygiene (i.e., modifiable factors and behaviors influencing sleep; Kronk et al., 2023;Van Dyk et al., 2019), by for example excessive use of electronics, which may partially be due to parents letting go of stricter media device regulations during the pandemic, especially at bedtime (Bacaro et al., 2022).In adolescents with ADHD, the majority already experiences sleep problems (Langberg et al., 2020), significantly more so than their neurotypical peers (NT; Marten et al., 2023).Their ADHD symptoms might negatively impact their sleep hygiene, for example by getting easily distracted by external stimuli before and during bedtime, having difficulties with planning, and sticking to a bedtime-routine, which in turn is known to have a negative effect on sleep quality (Martin et al., 2020).Due to their difficulties with executive functioning, in adolescence most specifically planning (Boyer et al., 2018), the noted loss of daily routines during COVID-19 (Jefsen et al., 2021) may have impacted them even more.Of note, so far no study has examined the relation between COVID-19 related restriction measures and sleep hygiene in adolescents with ADHD.
A small number of studies did examine the relation between COVID-19 restrictions and subjectively measured sleep comparing adolescents with and without ADHD, using a variety of different questionnaires.One study found significantly more parent-reported sleep disorders, in children and adolescents with ADHD as compared to NT children and adolescents during COVID-19 related confinement (Navarro-Soria et al., 2021).Another study examined perceived changes in the sleep of children and adolescents with and without ADHD, comparing periods before and during lockdown (Bruni et al., 2021).According to their parents, the ADHD group had a later bedtime, a reduced total sleep time (TST; i.e., time between sleep onset and offset), and an increased sleep onset latency (SOL; time it takes to fall asleep) during the lockdown compared to the NT group.Additionally, both groups experienced increased difficulties in falling asleep and more daytime sleepiness.Another study prospectively examined the effect of COVID-19 restrictions on the sleep of adolescents with and without ADHD (Becker et al., 2021).Both groups experienced more difficulties falling asleep and maintaining sleep and had later bed-and waketimes during COVID-19 compared to before COVID-19.This was based on self-and parentreport and no differences between the groups were reported.Surprisingly, both groups also reported less daytime sleepiness and NT adolescents were found to have a longer TST during school nights when there were COVID-19 restrictions.In sum, these findings suggest that adolescents with and without ADHD experienced sleep alterations during COVID-19, based on subjective assessments with questionnaires.However, results are mixed with regard to whether these changes are the same in adolescents with and without ADHD.Thus, further investigation is needed.
The current study adds to the literature by focusing on multiple aspects that have not been included in the previous studies.First, objective sleep parameters are included.Sleep is a broad concept and to fully understand it, both subjective and objective sleep parameters should be considered, as they do not always overlap (Becker et al., 2019).This might add valuable insight in how COVID-19 is related to sleep.Second, the relation with sleep hygiene is examined, which might be an important factor influenced by COVID-19 restrictions.Third, we assess adolescent-report instead of parent-report.This is crucial to understand their point of view and experiences (Becker, 2020), and might be more informative, as during adolescence, parents are less aware of the sleep and sleep problems of their children (Short et al., 2013).Lastly, as the majority of adolescents with ADHD experiences sleep problems, significantly more so than NT controls, only adolescents with ADHD and comorbid sleep problems are included in the current study, resulting in a more homogeneous group.This allows us to examine the specific effect of COVID-19 on their sleep problems and might help to further disentangle the previous mixed results.
Taken together, the current study aims at examining the relation between COVID-19 restrictions and subjective and objective sleep of adolescents with and without ADHD.To achieve this aim, we first examine the effect of ADHD diagnosis on subjective and objective TST and SOL, and subjective sleep and sleep hygiene problems.We hypothesize that adolescents with ADHD experience more sleep difficulties, as they are recruited to have comorbid sleep problems.Second, we examine whether COVID-19 restrictions are related to the sleep of adolescents with and without ADHD, using the same measures as described above.We hypothesize that COVID-19 restrictions have a negative impact on both the sleep of adolescents with and without ADHD and that this effect is stronger in the ADHD group.

Participants
In total, 100 adolescents were included (50 ADHD and 50 NT).The two ADHD groups (M Age = 14.46;SD = 1.53) consisted of in total 20 boys and 30 girls and the NT groups (M Age = 14.98;SD = 1.51) also consisted of in total 20 boys and 30 girls.In the ADHD group, 42 adolescents used ADHD medication.Inclusion criteria were 1) 13 to 17 years old, and 2) IQ ≥ 80 (as established by the two subtests Matrix Reasoning and Vocabulary of respectively the WISC-V [Wechsler, 2014] or the WAIS-IV [Wechsler, 2008]).Additionally, the ADHD group needed to 1) have a previous ADHD diagnosis, of which diagnostic criteria were confirmed at the time of assessment by a semi-structured diagnostic interview, the schedule for affective disorders and schizophrenia for school-age children-present and lifetime version (K-SADS-PL; Kaufman et al., 2016), 2) have stable ADHD medication use for 4 weeks, and 3) experience self-reported sleep problems at the time of assessment. 1 This was established by a sleep interview based on criteria of the DSM-5 (American Psychiatric Association, 2013) and the ICSD-3 (American Academy of Sleep Medicine, 2014).Sleep problems were defined as either a SOL of at least 20 minutes, being awake for at least 30 minutes after sleep onset or having a TST of less than 7 hours.In the NT group this criterion was not applied for inclusion.Exclusion criteria for both groups were 1) medication use for depression and anxiety, 2) melatonin use, 3) comorbid autism spectrum disorder, and 4) medical disorders and medication possibly affecting sleep, as these might otherwise account for the sleep problems experienced by both groups.

Procedure
The Ethical Committee Research UZ/KU Leuven and the Social and Societal Ethics Committee KU Leuven approved the study (S64197/G-2020-2319-R3). Participants were recruited across Flanders via psychologists, psychiatrists, secondary schools, academic hospitals, general practitioners, social media, and the network of the researchers.Both the ADHD and the NT adolescent groups consisted of two subgroups, with one having been tested while there were many COVID-19 restrictions and the second one having been tested while there were few COVID-19 restrictions (see Table 1).Written informed parental consent and adolescent assent were collected prior to the 2-hour meeting with the adolescent and a parent.During this meeting, inclusion criteria were verified and questionnaires were filled in.Afterwards, adolescents had to fill in a sleep diary and wear an actigraph.

Measures
Sleep parameters were defined as TST and SOL, calculated separately for week and weekend nights.For the subjective parameters, adolescents had to fill in a sleep diary using m-Path (Mestdagh & Verdonck, 2019).Objective sleep parameters were registered using wrist actigraphy (Motionwatch, CamNtech Ltd).Participants filled in the sleep diary and wore the actigraph for as many nights as needed to have measurements of 4 week nights where adolescents physically went to school the next day and two weekend nights.These specific school nights were chosen to be able to examine their sleep as consistently as possible across participants, due to the fact that schools were allowed to decide themselves how they would organize the 50% remote classes (this included both the timing of the remote classes and their content).Thus, by only including nights where they physically went to school the next day, we could still examine the effect of COVID-19 restrictions and less structure in for example their social life on their sleep, but could be certain that these findings were not impacted by the differences in remote class organization.
Sleep deprivation (i.e., chronic insufficient sleep) was measured with the Chronic Sleep Reduction Questionnaire (CSRQ), a 20-item self-report questionnaire (Meijer, 2008), with good internal consistency, as validated in a sample of Dutch adolescents (α = .85;Dewald et al., 2012).Sleep hygiene problems were measured using the revised Adolescent Sleep Hygiene Scale (ASHSr), a 24-item selfreport questionnaire in which a higher score means less sleep hygiene problems.It has good internal consistency (α = .84;Storfer-Isser et al., 2013).

Statistical Analyses
In total, four participants had missing data as they either provided <4 nights of week night data or <2 nights of weekend data and two had missing data regarding the ASHSr.Three two-way MANCOVAs examined whether the relation between COVID-19 restrictions and sleep was different for adolescents with and without ADHD.ADHD diagnosis and level of COVID-19 restrictions were the two independent variables, while each of the sleep outcomes (subjective TST and SOL, objective TST and SOL, and sleep and sleep hygiene problems) was included as a dependent variable.Age and gender were included as covariates.All analyses were conducted using IBM SPSS Statistics (Version 28; IBM Corp, 2021) and as effect size partial eta squared (η p 2 ) was computed, with .01being a small effect, .06medium, and .14large (Cohen, 1988, as cited in Lakens, 2013).

Effect of ADHD Diagnosis on Sleep
There was a main effect of ADHD diagnosis on subjective sleep, F(89, 4) = 11.39,p < .001,Pillai's Trace = .34,which confirmed the inclusion criterion; adolescents with ADHD had a significantly longer SOL during the week (η p 2 = .2) and the weekend (η p 2 = .31).During the week, their SOL was 44 minutes longer and during the weekend 41 minutes.No effect on subjective TST was found, neither during the week nor during the weekend.There also was a main effect of ADHD diagnosis on objective sleep, F(87, 4) = 4.29, p = .003,Pillai's Trace = .17,confirming the subjective inclusion criterion of more disturbed sleep in the ADHD group.Adolescents with ADHD had a significantly longer objective SOL both during the week (η p 2 = .07)and the weekend (η p 2 = .07)compared to adolescents without ADHD.During the week their SOL was 19 minutes longer and during the weekend their SOL was 17 minutes longer.ADHD diagnosis did not affect objective TST, neither during the week nor during the weekend.Lastly, a main effect of ADHD diagnosis on sleep deprivation and sleep hygiene problems was found, F(91, 2) = 17.36, p < .001,Pillai's Trace = .28.Adolescents with ADHD reported more sleep deprivation (η p 2 = .16)and more sleep hygiene problems (η p 2 = .24)compared to NT adolescents.

Effect of COVID-19 Restrictions on Sleep
There was a significant effect of COVID-19 restrictions on subjective sleep, F(89, 4) = 4.87, p = .001,Pillai's Trace = .18.However, the significant interaction effect, F(89, 4) = 3.13, p = .02,Pillai's Trace = .12,showed that this effect was stronger in adolescents with ADHD compared to NT adolescents.Adolescents with ADHD experienced a shorter subjective SOL during the weekend when there were many COVID-19 restrictions (η p 2 = .07).Their SOL was 30 minutes shorter during many COVID-19 restrictions, while the SOL of the NT group increased slightly by 2 minutes.There was no effect of COVID-19 restrictions on subjective SOL during the week and also not on subjective TST during week or weekend.A main effect of COVID-19 restrictions on objective sleep was also found, F(87, 4) = 2.59, p = .04,Pillai's Trace = .11,with shorter TST during the week when there were many restrictions (η p 2 = .05).COVID-19 restrictions were not related to objective TST during the weekend or SOL, neither during the week nor during the weekend.There was no significant interaction effect.There was no effect of COVID-19 restrictions on sleep deprivation and sleep hygiene problems and no interaction effect (Table 2 for findings of all outcomes).

Discussion
This is the first study including adolescents with ADHD and comorbid sleep problems that not only looked at subjective sleep parameters, but also included objective measurements and sleep hygiene problems when exploring sleep differences during few and many COVID-19 restrictions.The first hypothesis, stating that the ADHD group with comorbid sleep problems had significantly more sleep difficulties compared to the NT group, was supported.The second hypothesis was partially supported.During few COVID-19 restrictions, adolescents slept longer during the week.However, in contrast to our hypothesis, adolescents with ADHD reported that they fell asleep faster during many COVID-19 restrictions.
Subjective SOL in the ADHD group was found to be shorter during the weekend when there were many COVID-19 restrictions, while it stayed the same in the NT group.
These changes are partially in line with previous studies.Adolescents with ADHD were reported to experience an increase in SOL (Bruni et al., 2021), while NT adolescents reported either no changes (Ramos Socarras et al., 2021) or an increase in SOL during COVID-19 (Jamieson et al., 2021).Although speculative, a factor that could be responsible for the decrease in SOL in the ADHD group is diminished rumination.Adolescents with ADHD have been found to show a stronger reaction when experiencing cues of social rejection compared to their NT peers (Kohls et al., 2009), which might result in rumination in the evenings and consequently more sleep problems (Yu et al., 2022).Considering that during many COVID-19 restrictions, adolescents had significantly less social interactions and fewer stimuli on the weekend, these might have resulted in the ADHD group not being overstimulated.Consequently, a hypothesis may be that they were ruminating less in the evenings and could fall asleep faster.It is however, still interesting to note that the SOL of the ADHD group did stay longer than clinically acceptable (longer than 20-30 minutes) independent of restrictions, while the SOL of the NT group was always in the clinically acceptable range.Consequently, one might suggest that a clinically elevated SOL, also called initial insomnia, might be a defining feature of adolescents with ADHD and sleep problems.One underlying biological mechanism that has been proposed is a delayed circadian phase, reported in individuals with ADHD across the lifespan (Bijlenga et al., 2019).
Additionally, objective week TST was shorter during many COVID-19 restrictions in both groups.Findings of previous studies have been inconclusive.In NT adolescents TST increased (Becker et al., 2021;Bruni et al., 2022;Weingart et al., 2021), but in adolescents with ADHD either no change (Becker et al., 2021) or a decrease has been reported (Bruni et al., 2021).In contrast to our study, the previous studies examined sleep during fully remote classes.This might have resulted in delayed wake times and consequently, longer TST, as adolescents did not need to transfer to school (Weingart et al., 2021).The current study examined sleep during school nights where adolescents physically attended school the next day, resulting in a more valid measure and reference point.Still, there were changes across schools in remote and physical schooling (e.g., half week remote/ half week physical).Thus, it could be suggested that due to the loss of structure, including the changes in school starting times (combination of remote and live classes), their bedtime was later than usual.This is in line with previous studies (Becker et al., 2021;Bruni et al., 2021).However, in the current study they could not compensate the later bedtimes by later wake times.Consequently, their TST could have been shorter.Possibly, especially adolescents with executive functioning deficits might have had difficulties with the loss of structure, as executive functioning has been found to predict impairments in daily life (Barkley & Fischer, 2011), but further research regarding this is needed.According to Cortese et al. (2020), teaching behavioral strategies to be able to deal with this loss of routine may be particularly important for adolescents with ADHD.
No effects of COVID-19 restrictions were found on sleep deprivation and sleep hygiene problems.Although no study so far has examined the effect of COVID-19 on sleep hygiene problems broadly, there were a few studies focusing on specific sleep hygiene variables.Sleep hygiene factors such as screen time in the general public (Drumheller & Fan, 2022) and social media use and gaming in children and adolescents with ADHD (Sciberras et al., 2022) were found to increase during COVID-19 and a relation with increased sleep difficulties has been reported (Drumheller & Fan, 2022), but assessing sleep hygiene more broadly we did not find such effects.
A limitation in the current study was that power to detect small effects was limited as the sample compared four groups of approximately 25 adolescents; more longitudinal research examining sleep in adolescents with ADHD and comorbid sleep problems is needed.Furthermore, although causality cannot be assumed, we found that the amount of COVID-19 restrictions was associated with subjective and objective sleep parameters.Another limitation linked to COVID-19 related restrictions is that this is a broad variable with different regulations across countries/schools.Consequently, we tried to have a more homogeneous assessment of its effects by only looking at nights where adolescents physically went to school the next day.Also, the measurements included should be interpreted with their limitations in mind.All subjective measurements were self-report.Parent-report next to self-report could have given some additional insight, as adolescents with ADHD have sometimes been found to be poor informants of their symptoms (Sibley et al., 2019).Additionally, although actigraphy has reasonable validity and reliability (Sadeh, 2011), polysomnography is still the gold standard and should therefore be considered in future studies (Rundo & Downey, 2019).Lastly, this study focused specifically on a homogeneous group of adolescents with ADHD and comorbid sleep problems, and is therefore not generalizable to all adolescents with ADHD.However, as approximately 72% of adolescents with ADHD experience sleep problems (Langberg et al., 2020), it might still give some idea how their sleep is affected.Our results may also not be representative for the entire population of adolescents with ADHD, as we excluded those with comorbid autism spectrum disorder and the sleep of adolescents with autism spectrum disorder is also affected by COVID-19 restrictions (Kronk et al., 2023).Future research should focus on specific mechanisms, such as the dimensions of executive functioning, underlying the possible effects of COVID-19 restrictions or comparable restrictions on the sleep of adolescents, as previous research has reported different results regarding the directionality of the effects.This is specifically important, as sleep is crucial for both physical and mental wellbeing of adolescents.Poor or little sleep increases negative affect, which might lead to more internalizing and externalizing difficulties (Becker & Gregory, 2020).As adolescents with ADHD are already more at risk for these comorbid difficulties (Reale et al., 2017), their sleep problems might have additional detrimental effects on their wellbeing.
In sum, this study, using a multi-method approach to assess sleep and sleep problems, suggests that COVID-19 restrictions are related to shorter sleep duration in both groups, but also to falling asleep faster in the ADHD group.Translating this to clinical implications, it shows that each adolescent might react differently to the COVID-19 restrictions and that it might not solely have detrimental effects on sleep.However, it might be important to teach them behavioral strategies to help them cope with comparable situations and to also monitor their mental wellbeing (Cortese et al., 2020).

Table 1 .
Overview of Timing and Few and Many COVID-19 Restrictions.

Table 2 .
Descriptive Statistics of the ADHD and NT Groups During Few and Many COVID-19 Restrictions.