Examining the associations between attention‐deficit/hyperactivity disorder, sleep problems, and other mental health conditions in adolescents

Adolescents with attention‐deficit/hyperactivity disorder (ADHD) often experience greater sleep difficulties compared to those without. However, findings are mixed, and other mental health conditions are often overlooked. This study aimed to examine the relationship between sleep problems, ADHD, and other mental health conditions in a sample of adolescents. Data from 373 adolescents aged 10–19 years was used as part of the wider ‘Healthy Brain Network’ study, which targets children and adolescents experiencing mental health and neurodevelopmental difficulties. Mental health conditions were assessed via a comprehensive assessment. Sleep was measured by self‐ and parent‐report, as well as via up to a month of actigraphy data. Actigraphy data were analysed using mixed‐methods modelling, while subjective sleep data were analysed using multiple regression. Subjectively‐reported sleep was generally worse in adolescents who had ADHD and other mental health conditions compared to those with ADHD but no other conditions. There were no associations between ADHD status and objective sleep measures or self‐reported measures, but a significant association was found between ADHD status and parent‐reported sleep difficulties, even when accounting for other conditions. Parent‐reported sleep problems were associated not only with ADHD, but also with anxiety, depression, and externalising disorders. The strength of association between ADHD and sleep problems is potentially not as strong as previously thought when considering the role of other mental health conditions. Clinicians should consider the role of other mental health conditions when sleep problems are present, and vice versa. The study also highlights the importance of comprehensive, multi‐informant assessment of mental health conditions, including sleep.


| INTRODUCTION
Attention-deficit/hyperactivity disorder (ADHD) is a neurodevelopmental disorder characterised by symptoms in the broad domains of hyperactivity, inattention, and/or impulsivity (American Psychiatric Association, 2013).In adolescents, ADHD is further associated with impacts in areas such as educational achievement, friendships, quality of life, and overall mental health (Birchwood & Daley, 2012;DuPaul & Stoner, 2014;Klassen et al., 2004;Wehmeier et al., 2010).Sleep problems are reported to be more prevalent in adolescents with ADHD compared to those without (Becker et al., 2019); however, there appear to be inconsistent findings regarding sleep problems and ADHD depending on whether subjective (e.g., surveys) or objective measures (e.g., actigraphy) of sleep are used.Considering sleep in adolescents with ADHD, in particular, is of interest as adolescence is also a time of transition in many areas (Carskadon, 2011), but particularly in the domain of sleep (Colrain & Baker, 2011).Poor sleep in adolescence can contribute to a number of negative outcomes, such as poorer emotional regulation (Baum et al., 2014), increased risk-taking behaviours (Shochat et al., 2014), academic challenges (Cusick et al., 2018), and poorer mental health (Hestetun et al., 2018;Short & Louca, 2015).
Although there are many different definitions of sleep problems, most night-time sleep problems can be broadly conceptualised as insufficient sleep duration, poor sleep quality, or inappropriate timing of the sleep period (i.e., delayed sleep phase; Mindell & Owens, 2015).This study focuses on sleep duration and sleep quality.Sleep can be broadly measured in one of two ways: using objective (i.e., polysomnography [PSG]; actigraphy) or subjective (i.e., sleep diaries; questionnaires) measurement.
Subjective measures can either be self-reported or can be reported by another who has close contact with the individual in question (usually a parent).
Both methods offer distinct advantages and disadvantages.The main advantages of objective methods compared with subjective include greater objectivity, accuracy, and richness of data.Disadvantages include cost and feasibility.Within objective measurement methods, PSG is generally considered to be the most accurate in terms of measurement of more biologically based sleep problems in particular (e.g., obstructive sleep apnea, restless legs syndrome), but may not be feasible in research studies.Actigraphy, meanwhile, is considered to be less accurate than PSG in measuring sleep characteristics such as sleep onset and offset, but has a much lower participant burden, and is significantly more cost-effective (Acker et al., 2021).
Additionally, actigraphy enables higher ecological validity, as the actigraph can be worn for long periods of time, in an individual's usual home environment, which can improve measurement validity (Van Someren, 2007).However, given that actigraphy is based on movement, participants could be potentially coded as being asleep when they are lying in bed awake but relatively still.Subjective methods, such as sleep diaries and sleep questionnaires, are cheap and easy to administer, but require participants to recall their sleep patterns over a specific period of time (e.g., over the past 7 nights, over the last month), which can lead to over-or underestimation of sleep problems (Ji & Liu, 2016).Subjective measures may be particularly good at understanding characteristics associated with sleep that actigraphy cannot measure such as sleep-related anxiety.In children and adolescents, sleep diaries are moderately to highly correlated with actigraphy for estimating total sleep duration (Arora et al., 2013;Mazza et al., 2020;Tremaine et al., 2010).However, individuals tend to overestimate their sleep duration on other subjective measures, such as questionnaires, primarily through an underestimation of night awakenings (Tremaine et al., 2010).Likewise, parents also tend to overestimate their child's sleep duration (Mazza et al., 2020).It is also important to note that mental health issues can affect subjective perception of sleep difficulties, with anxious children reporting significantly more sleep problems than their peers without anxiety, despite no differences being detected on objective parameters (Alfano et al., 2015).In other words, despite objective measures not detecting sleep difficulties, the subjective experience of the individual may still be one of insufficient or poor quality sleep.For this reason, it is important that a combination of measures is used.
Studies using subjective measures of sleep have generally found that adolescents with ADHD have poorer sleep than their peers without ADHD (Lunsford-Avery et al., 2016).In terms of recent studies in the area, Sanabra et al. (2021) found that adolescents with ADHD had greater sleep onset latency (i.e., trouble getting to sleep) and poorer sleep efficiency than adolescents without ADHD based on selfreported measures.Similar results were reported by Becker et al. (2019), who found that adolescents with ADHD were more likely to report obtaining insufficient sleep and reported more subjective sleep issues than adolescents without ADHD.
However, these associations are often not replicated when examining sleep assessed via actigraphy.Becker et al. (2019), who detected relationships between subjectively reported sleep issues and ADHD, found little evidence for these relationships using actigraphy.
Although adolescents with ADHD had shorter time in bed and earlier waking times assessed via actigraphy, these effects were small in magnitude.Additionally, key actigraphy variables such as sleep efficiency and wake after sleep onset (WASO) did not differ between ADHD and non-ADHD groups.Similar findings were reported by Sanabra et al. (2021), who found no differences between adolescents with and without ADHD on actigraphy variables.Prehn-Kristensen et al.
(2011), who measured sleep using PSG with a small sample (N = 24), found that adolescents with ADHD had longer sleep onset latency and lower sleep efficiency than adolescents without ADHD, but interestingly, had longer rapid eye movement (REM) sleep.However, again, these associations were relatively small in magnitude.A systematic review (Kirov & Brand, 2014) also noted these inconsistencies, illustrating a general pattern whereby sleep difficulties are reported via subjective report in young people with ADHD, but these findings are not confirmed by objective measurement of sleep (i.e., PSG, actigraphy).Some studies have also highlighted the influence of other mental health conditions on sleep in children with ADHD (Becker et al., 2018;Bergwerff et al., 2016), with less research considering adolescents with ADHD, specifically.Becker et al. (2019), who found differences between ADHD and non-ADHD groups predominately on subjective sleep parameters, also found that anxiety disorders, depressive disorders, and externalising disorders were consistently associated with sleep variables (e.g., self-reported sleep problems, parentreported sleep problems).However, similarly to the examination of ADHD and objective sleep problems, little evidence was found for a relationship between other conditions and actigraphy variables.In the Becker et al. (2019) study, the interactions between ADHD, other conditions, and sleep problems were not the focus of the research.A more fine-grained analysis of these relationships is required in order to elucidate the interplay between these factors.Furthermore, the study by Becker et al., (2019) had a main focus on sleep, which means that families of individuals with sleep problems may be drawn to these studies.
In summary, there is a relationship between sleep and mental health problems in adolescents (Hestetun et al., 2018), and many adolescents with ADHD also experience other mental health difficulties (Reale et al., 2017).Emerging evidence suggests that sleep problems in adolescence may be a risk factor in the aetiology of suicidal thoughts and behaviours (Blake & Allen, 2020); however, few studies have considered the role of other mental health difficulties in understanding the sleep problems of adolescents with ADHD.It is necessary to examine these associations using a comprehensive battery of sleep measures in adolescents with and without various mental health and neurodevelopmental conditions, including ADHD, in a broad sample that has not been recruited for a specific study focused on sleep difficulties.Therefore, this study aimed to: Brain Network (HBN) initiative (https://healthybrainnetwork.org/), which aims to create and share a biobank of data from individuals (aged 5-21 years) in the New York area (Alexander et al., 2017).The initiative collects a broad range of psychiatric, cognitive, behavioural, genetic, and lifestyle data, as well as magnetic resonance imaging (MRI), electroencephalogram (EEG) recordings, and actigraphy recordings (Alexander et al., 2017).Participants were recruited from the community, using advertisements that encouraged participation of families who had noticed mental health difficulties in their child.These advertisements were distributed to community members, educators, and local care providers, as well as through email lists and events.

| Actigraphy
Actigraphy was used to measure objective adolescent sleep parameters.Participants were provided with a wrist-mounted actigraph (ActiGraph wGT3X-BT) which was typically returned within 3-4 weeks.Participants were instructed to wear the device on their non-dominant hand and keep it on as much as possible.Raw data, recorded at 30 Hz, were processed using the R package 'GGIR', an open source and actively managed package which provides summaries of various sleep parameters (Migueles et al., 2019;van Hees et al., 2015van Hees et al., , 2018)).The method used to aggregate tri-axial acceleration into epoch level summary was the Euclidean Norm Minus One (ENMO).The sleep parameters included in the study were: sleep period, sleep duration, WASO, and sleep efficiency.Participants were excluded from the dataset if they had <3 nights of actigraphy data available (Littner et al., 2003), which excluded 81 participants (18% of original sample).For the remaining participants, a mean (SD, range) of 16 (8.81, 3-36)

| Sleep Disturbance Scale For Children (SDSC)
The SDSC is a parent-reported 27-item questionnaire that assesses sleep difficulties (Bruni et al., 1996), and has been widely validated for use with children and adolescents (e.g., Lewandowski et al., 2011).The scale provides a total score and six subscale scores.In the present study, the total and 'Disorders of Initiating and Maintaining Sleep' (DIMS) scores were used (total ω = 0.83, DIMS ω = 0.85).Each item was rated using a 5-point rating scale from 1 = never to 5 = always (daily), indicating the frequency of the sleep problem over the past 6 months, with higher scores indicating poorer sleep quality.

| Youth Self-Report (YSR)
The YSR, which is the self-reported version of the Child Behaviour Checklist (CBCL; Achenbach & Rescorla, 2001) was used to assess self-reported sleep difficulties.For the assessment of sleep, two questions from the measure were used: 'I sleep less than most kids during day and/or night', and 'I have trouble sleeping'.These items were scored using a 3-point response scale (0 = Not True [as far as you know], 1 = Somewhat or Sometimes True, 2 = Very True or Often True).Although not a comprehensive sleep measure, previous research has found some concurrence between these items and other sleep measures (Gregory et al., 2011).

| Procedure
The broader study, including data collection, was approved by the Chesapeake Institutional Review Board.The present study was During their first visit, participants were provided with an actigraph device.Participants were instructed to wear this during both waking and sleeping hours, and brought the device back at each visit to be recharged.

| Statistical analysis
All statistical analyses were conducted in the R environment (R Core Team, 2020), using the RStudio Integrated Development Environment.Linear mixed-methods modelling was utilised to examine differences in actigraphy sleep parameters between participants with and without ADHD and other mental health conditions, using the 'lme4' package (Bates et al., 2015).Analyses were conducted in both unadjusted and adjusted (i.e., controlling for variables such as age and sex) formats.Multiple regression was utilised to examine the relationships between ADHD, other mental health conditions, and subjective sleep problems.Again, analyses were conducted in both unadjusted and adjusted (i.e., controlling for variables such as age and sex) formats.

| Descriptive characteristics
See Table 1 for a description of the sample.The mean (SD) age was 13.21 (2.37) years.ADHD was the most common mental health condition in the overall sample (62.5%), followed by anxiety disorders (27.9%), and autism (17.2%).Rates of other mental conditions were similar in ADHD versus non-ADHD groups, with the exception of autism spectrum disorder, which was detected at nearly double the rate in the ADHD group (ADHD: 21%, non-ADHD: 10.7%).The mean (SD) sleep period (i.e., time from sleep onset to sleep offset) in the overall sample was 7.57 (2.41) h, while sleep duration (sleep period minus night awakenings) was 6.13 (2.08) h.

| Agreement between sleep measures
Correlations between actigraphy, self-and parent-report sleep variables are detailed in Figure 2. In general, self-and parent-report sleep difficulties were weakly correlated (rs = 0.19-0.25),while objective variables were very weakly correlated with subjective variables (rs = 0.10-0.15).

| Associations between ADHD and sleep variables
Linear mixed-methods modelling and multiple regression were utilised to examine whether sleep variables differed depending on an T A B L E 1 Descriptive characteristics of key variables  were associated with WASO (β = 0.07, p = 0.024), and autism was associated with both sleep period (β = 0.07, p = 0.040) and sleep duration (β = 0.08, p = 0.041).

| Group comparisons between various combinations of conditions
Group comparisons were carried out to understand group differences in sleep difficulties (Table 4; Figures 3 and 4).There were no group differences for any of the actigraphy variables, or the self-reported question 'I sleep less than most kids'.A significant difference was found

| DISCUSSION
This study examined the relationships between ADHD, other mental health conditions, and sleep problems in adolescents.To our knowledge, this is one of the first studies to examine the relationship between ADHD and comprehensively measured sleep problems while also considering the role of other mental health conditions.There was little evidence for an association between ADHD and objective sleep parameters; however, autism was associated with longer sleep period and duration, and externalising disorders were associated with greater WASO.ADHD was associated with parent-reported sleep problems, but not with self-  Associations between ADHD and self-reported sleep problems were not detected.This is consistent with Becker et al. (2019), who found that the associations detected between ADHD and parentreported sleep were not replicated when examining self-reported sleep.However, depression or dysthymia was associated with worse self-reported sleep.This is partially consistent with Becker et al.
(2019) who found self-reported sleep to be associated with depression, anxiety, and externalising conditions.However, the present results should be interpreted with caution, as we measured self-reported sleep using only two brief questions.Despite the brevity of the measure, the results are nonetheless consistent with previous results (Becker et al., 2019).The strongest and most consistent findings were detected when comparing adolescents who had at least one mental health condition (excluding ADHD) with adolescents who did not have any mental health condition.When comparing these groups, adolescents with other conditions reported that they slept worse than their peers with no conditions, a result which was replicated when examining parent-reported sleep.
The results showed evidence of an association between ADHD status and parent-reported total sleep problems.This is similar to Becker et al. (2019), who found evidence for an association between ADHD and parent-reported sleep.Sanabra et al. (2021) (Carskadon, 2011), parents might be less aware of their child's sleeping patterns.There is also the possibility of a 'negative halo effect' (Alacha & Lefler, 2021), whereby a parent could observe a challenging behaviour in the child, which then influences their ratings in other areas.It is possible that for children with other conditions, parents may be assuming that their sleep is more negatively impacted than it really is, but this needs to be explored further using more detailed measures of adolescentreported sleep.
In general, relationships were seen for subjective measures, but not for objective ones.Sleep problems have been consistently linked with a number of mental health conditions, ADHD being one of them.
However, the findings are inconsistent and variable, with no one pattern being identified.Moreau et al. (2014) found that ADHD itself was not associated with specific sleep problems, but when examining psychiatric diagnoses as a whole, associations were found.Similar to the results of the present study, this suggests that there may be a general relationship between mental health and neurodevelopmental difficulties and sleep problems, with no one condition being able to account for these problems.This may also account for the variability in past studies examining the relationship between ADHD and sleep, in that differential rates of other mental health conditions may be driving the inconsistent results.Another possibility is that adolescents with mental health difficulties require more sleep in order to maintain daily functioning, compared with adolescents who do not experience mental health conditions.Therefore, while objective differences in sleep parameters might not be detected when comparing adolescents with mental health conditions to those without, the subjective experience of the adolescent and their parents might illustrate a picture of insufficient sleep.
A surprising finding was the relationships between autism and actigraphy variables sleep period and sleep duration, with autism associated with longer sleep period and duration.This is inconsistent with previous studies finding that autism is associated with reduced sleep duration in children and adolescents (Goldman et al., 2012;Humphreys et al., 2014).This may indicate that the relationship between autism and sleep is highly complex; however, the result should be interpreted with caution, considering the relatively small magnitude of the effect.The effect may possibly be due to sampling bias, that is, parents who engage with the HBN study may also be engaged with healthcare professionals, and given that sleep disorders are known to be common in adolescents with autism (Cortese et al., 2020), and treatments have been found to be effective (Cortese et al., 2020) Limitations of the present study should also be noted.The crosssectional nature of the study precludes any casual inferences from being made.The sample, while sourced from the general population, was likely biased towards participants with higher levels of mental health issues, due to the sampling methods used (i.e., targeting families that had noticed mental health issues in their child).This limits our ability to generalise the present results to the general population.
Additionally, data on medication use were not analysed in the present study.While some data on medication use were available, they were only collected on the day that participants attended the clinic to complete testing.Therefore, their medication use on that day may not have been reflective of their medication use over the actigraphy collection period, so the decision was made not to include this variable in analyses.Previous research has shown that ADHD medication can affect sleep (Stein et al., 2012), although recent research has found that stimulant medication does not necessarily affect sleep in children and adolescents with ADHD (Sanabra et al., 2021).With regards to sleep measurement, sleep diaries were not collected, preventing an analysis of actigraphy variables such as sleep onset latency.An additional measurement limitation was the basic measurement of selfreported sleep.This may have limited our understanding of adolescents' perceptions of their own sleep.We would also like to point out that WASO estimates were high; however, our estimates were consistent with previous research in similar populations (e.g., Becker et al., 2019;Migueles et al., 2021).Lastly, mental health conditions This study examined the relationships between ADHD, other mental health conditions, and sleep problems in adolescents.No evidence was found for a relationship between ADHD and objectively measured sleep; however, a relationship was found between ADHD and parent-reported sleep difficulties.Additionally, parent-reported

1.
Describe objective (assessed via actigraphyincluding sleep period, sleep duration, sleep efficiency and WASO) and subjective (parent-and self-report) sleep parameters in adolescents across a range of mental health conditions, compared to adolescents with no mental health conditions, and to examine which conditions are most strongly associated with sleep variables, and 2. Examine differences in objective and subjective sleep parameters across four groups of adolescents: (i) ADHD and other conditions; (ii) ADHD and no other conditions; (iii) other mental health/ neurodevelopmental conditions but no ADHD; and (iv) no ADHD or other mental health/neurodevelopmental conditions. 2 | METHODS 2.1 | Participants Participants were 373 adolescents aged 10-19 years (mean [SD] = 13.21[2.37]).See Figure 1 for a representation of how the final sample size was reached.Data were accessed from the ongoing Healthy Inclusion criteria for initial data collection are as follows: aged 5-21 years, ability to provide consent (or assent when aged 5-17 years), and fluency in English.Participants were excluded from initial data collection if there were safety concerns present (e.g., danger to self or others), insufficient verbal abilities, intelligence quotient (IQ) <66, or neurological concerns that could interfere with MRI and EEG interpretation.Participants were included in the present study if they were aged between 10 and 19 years and had at least 3 nights of actigraphy data available, in line with recommendations made byLittner et al. (2003).

F
I G U R E 1 Healthy Brain Network recruitment flow chart 2.2 | Measures Data were collected both via self-and parent-report.For participants aged <11 years, a trained research assistant read and explained individual items and collected responses from participants.2.2.1 | Schedule For Affective Disorders And Schizophrenia -Children's version (K-SADS) Participants and their parents/guardians were administered the K-SADS by a trained member of the clinical team.For participants aged <11 years, clinicians determined whether the K-SADS would be administered to both participants and parent/guardian or solely parent/guardian based on verbal function and expected ability to tolerate the clinical interview.Selected modules (neurodevelopmental disorders, depressive disorders, anxiety disorders, stress and obsessivecompulsive disorders, and disruptive behaviour and impulse control disorders) from the K-SADS were used in reaching consensus diagnoses.Previous studies have established the validity and reliability of the measure (de la Peña et al., 2018; Nishiyama et al., 2020).Consensus diagnoses were reached based on Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5) criteria based on information from the parent and child along with historical records and other HBN assessments (such as the K-SADS).Where clinical information indicated the presence of autism spectrum disorder, additional assessments were conducted (Autism Diagnostic Interview -Revised [ADI-R]; Autism Diagnostic Observation Schedule, second edition [ADOS-2]).
nights of actigraphy data were collected.A sleep diary was not collected, although past research has found that collection of actigraphy data without a sleep diary is valid, despite preventing variables such as sleep onset delay from being derived (van Hees et al., 2018).
approved by the Deakin University Human Research Ethics Committee (2019-325).Participants or their legal guardians (if they were aged <18 years) completed a pre-screening interview.Participants then attended a HBN office in either Staten Island or Midtown Manhattan to participate in testing.Participants typically attended four testing sessions.Participants taking stimulant medication were asked to discontinue their medication during days of in-person testing.
between adolescents with ADHD and other conditions and adolescents with ADHD and no other conditions on the parent-reported SDSC DIMS subscale (β = À3.73,p = <0.001),such individuals with ADHD and other conditions had worse sleep than individuals with ADHD and no other conditions.A significant difference was found between adolescents without ADHD but with other conditions and adolescents without ADHD and no other conditions, on the selfreported sleep question 'I have trouble sleeping' (β = À2.14, p = 0.034), as well as on the SDSC total score (β = À2.38,p = 0.019) and SDSC DIMS subscale score (β = À2.76,p = 0.007), such that individuals with other conditions had worse parent-and self-reported sleep.A significant difference was found between adolescents with ADHD and other conditions and adolescents without ADHD but with other conditions on the SDSC total score, (β = À3.01,p = 0.003), such that individuals with ADHD and other conditions had worse sleep than those with other conditions only.No group differences were detected between adolescents with ADHD and no other conditions and adolescents without any conditions at all.
reported sleep problems.Anxiety, depression, and externalising disorders were also associated with parent-reported sleep.Lastly, depression was associated with self-reported sleep problems.When comparing groups based on other conditions, the presence of other mental health conditions (i.e., not ADHD) was associated with poorer sleep per parentreport, regardless of whether ADHD was present or not.When comparing groups with ADHD and other conditions with adolescents without ADHD and other conditions, the group with ADHD and other conditions had worse sleep on parent-reported sleep parameters.The absence of associations between ADHD status and objectively measured sleep is partially consistent with studies such asBecker et al. (2019), who found small differences between ADHD and non-ADHD groups only on objectively measured time in bed.Consistent with our results,Sanabra et al. (2021) did not find an association between objectively measured sleep and ADHD status in a sample of children and adolescents.A systematic review(Lunsford-Avery et al., 2016) concluded that there was little evidence for a relationship between objective sleep variables and ADHD status, although many of the studies were hampered by methodological limitations.The results of the present study indicate that in adolescents, there is limited evidence of an association between objectively measured sleep problems and ADHD.

(
both ADHD and other) were analysed categorically.Despite the utility of categorical classification, effects may be masked because individuals fall just below the cut-off for a particular disorder.Future studies should consider the use of dimensional measurement of symptoms.The present study builds on previous research suggesting that sleep and ADHD may not be as closely linked as previously thought, particularly when sleep is assessed using objective methods.The study also demonstrates the links between subjective measures of sleep and various internalising and externalising disorders, such as depression, anxiety, and conduct disorder.This has implications for clinical practice, in that clinicians should be aware of these relationships when assessing adolescents with ADHD, particularly if sleep problems are present.Sleep problems may indicate the presence of other disorders or may be contributing to the manifestation of other disorders.Additionally, when assessing individuals, multiple sources of information should be considered, given the differences between parent-and self-reported sleep problems, and the increased independence of adolescence meaning that parents might not be as aware of sleep issues.

Table 3
for self-and parent-reported sleep variables).First, only ADHD status, age, and sex were included in the models.Subsequently, other conditions were also included in the models (anxiety disorders, depressive disorders, externalising disorders, autism spectrum disorder).ADHD status was not associated with sleep parameters in either iteration of the model.Regarding actigraphy sleep variables, in Model 1, age was significantly associated with sleep period (β = À0.14, p < 0.001) and sleep duration (β = À0.13,p = 0.001).These associations held in Model 2, and additionally, conduct disorder and oppositional defiant disorder Group differences between ADHD, no ADHD, other conditions and no other condition groups T A B L E 4Note: Bold values statistically significant at p < 0.05.Abbreviations: ADHD, attention-deficit/hyperactivity disorder; CI, confidence interval; DIMS, Disorders of Initiating and Maintaining Sleep subscale; SDSC, Sleep Disturbance Scale For Children.
Comparisons of parent-reported sleep problem scores (Disorders of Initiating and Maintaining Sleep [DIMS] subscale between different mental health groups.*** p < 0.001, ** p < 0.01, * p < 0.05.Parent-reported sleep problems measured using the Sleep Disturbance Scale For Children.ADHD, attention-deficit/ hyperactivity disorder F I G U R E 3 Comparisons of total parent-reported sleep problem scores between different mental health groups.*** p < 0.001, ** p < 0.01, * p < 0.05.Parent-reported sleep problems measured using the Sleep Disturbance Scale For Children.ADHD, attentiondeficit/hyperactivity disorder independence from parents