Media Use, Sleep Quality, and ADHD Symptoms in a Community Sample and a Sample of ADHD Patients Aged 8 to 18 Years

Article
Introduction
ADHD is one of the most common neurobehavioral disor- ders presented to pediatric mental health professionals, affecting one of 20 children and adolescents worldwide (Polanczyk, de Lima, Horta, Biederman, & Rohde, 2007). In addition, the syndrome often occurs with other psychiat- ric conditions and attendant symptoms. According to parent reports, as many as 73% of children with ADHD suffer from sleep problems (Sung, Hiscock, Sciberras, & Efron, 2008). Frequently reported sleep complaints include diffi- culties initiating and maintaining sleep with increased night awakening (Cortese, Faraone, Konofal, & Lecendreux, 2009). The resulting fragmentation of the physiological sleep architecture reduces the sleep recovery process and, thus, may lead to excessive daytime sleepiness in people with ADHD (Cortese et al., 2009; Yoon, Jain, & Shapiro, 2012). Overall, patients with ADHD rate their sleep quality as worse when compared with healthy peers (Owens et al., 2009). As previous research mainly focused on parental assessment of sleep, data obtained from children and ado- lescents with ADHD are quite rare. Such studies, however,
suggest that self- and parent reports correlate weakly, with higher correspondence between self-report assessment of sleep and actigraphy data (Owens, Maxim, Nobile, McGuinn, & Msall, 2000; Owens et al., 2009). Thus, taking only the parent’s perception into consideration will lead to a biased and limited insight into the sleep’s role in ADHD.
At the present time, causal and maintaining factors of sleep problems in ADHD are of great interest, because ade- quate sleep and regular sleep-wake schedules are essential for general health and well-being. One potential underlying pathophysiology explaining why disrupted sleep is highly comorbid with childhood ADHD is suggested by various associations of ADHD with circadian rhythm disturbances (Coogan, Baird, Popa-Wagner, & Thome, 2016). Light,
802014 JADXXX10.1177/1087054718802014Journal of Attention DisordersThoma et al. research-article2018
1University of Freiburg, Germany 2University of Cologne, Germany
Corresponding Author: Christoph Klein, Department of Child and Adolescent Psychiatry, Faculty of Medicine, University of Freiburg, Hauptstraße 8, D-79104 Freiburg, Germany. Email: [email protected]
Media Use, Sleep Quality, and ADHD Symptoms in a Community Sample and a Sample of ADHD Patients Aged 8 to 18 Years
Vanessa K. Thoma1, Yoanna Schulz-Zhecheva1, Christoph Oser1, Christian Fleischhaker1, Monica Biscaldi1, and Christoph Klein1,2
Abstract Objective: Relationships between sleep, screen-based media, and ADHD symptomatology were investigated using a case- and community-based approach. Method: N = 357 healthy and N = 61 children with ADHD (12.72 ± 2.83 years) completed a sleep and media questionnaire. To measure ADHD symptomatology, parents filled out the Strengths and Weaknesses of ADHD symptoms and Normal behavior (SWAN) scale. Two samples were formed: a matched (N = 61 patients and N = 61 controls) and a community sample (N = 357 healthy participants and N = 20 patients). Results: Compared with controls, participants with ADHD reported delayed sleep onset and more screen time on school days. Adolescent patients showed more behavior promoting delayed sleep phase. In the community sample, media time, sleep deviation, and circadian rhythm were correlated with ADHD symptomatology. Furthermore, media time, sleep-wake behavior, and sleep deviation were predictive of ADHD symptomatology (variance explained = 4%-15%). Conclusion: Longer media time and inadequate sleep-wake behavior increase the risk of ADHD-like symptoms. However, research using objective assessments is needed to disentangle this distinct association and to provide possible directions for intervention. (J. of Att. Dis. 2020; 24(4) 576-589)
Keywords ADHD, sleep, (screen-based) media, chronotype, symptoms
especially short-wavelength light at around 460 nm, is the dominant human “zeitgeber” generating coherent, recurring biological rhythms at different levels of organization (Czeisler et al., 1989; Duffy & Czeisler, 2009; Hastings, O’Neill, & Maywood, 2007).
Higher rates for circadian-related abnormalities in chil- dren and adolescents suffering from ADHD exist for chronic sleep-onset insomnia (Corkum, Moldofsky, Hogg- Johnson, Humphries, & Tannock, 1999; Cortese et al., 2009) accompanied by circadian phase delay with later dim light melatonin onset (DLMO; Van der Heijden, Smits, Van Someren, & Gunning, 2005), a stronger evening circadian tendency (Benk Durmu, Rodopman Arman, & Ayaz, 2017; Gruber et al., 2012), and the association of a poly- morphism (rs1801260) of the circadian locomotor output cycles kaput (CLOCK) gene (Xu et al., 2010), which was reported to be related to eveningness typology and delayed sleep onset in patients with ADHD (Mishima, Tozawa, Satoh, Saitoh, & Mishima, 2005). These findings were confirmed by a more recent study, revealing a dose- response relationship between disrupted sleep patterns including sleep-onset latency, sleep deficiency, short sleep duration, delayed sleep phase disorder (DSPD), and ADHD-like behavior, with higher odds for inattention than for hyperactive/impulsive symptoms (Hysing, Lundervold, Posserud, & Sivertsen, 2016). Similarly, Hennig, Krkovic, and Lincoln (2017) identified the evening chronotype as a predictor for inattention.
Given the important influence of bright light on the cir- cadian system and daytime functioning, the impact of the exposure to light from self-luminous displays on biological clock functions has become a matter of great concern for researchers and clinicians. The light emitted by screen- based devices is enriched for short wavelengths in the blue range of the spectrum and, therefore, may interact with the biological clock (Touitou, Touitou, & Reinberg, 2016). Meanwhile, new media forms including screen-based elec- tronic devices enjoy great popularity among children and adolescents. In recent years, young people’s media use has increased dramatically to an average daily screen time of more than 4.5 hr in tweens (defined as 8- to 12-year-olds) and more than 6.5 hr in teens (defined as 13- to 18-year- olds; Rideout, 2015).
Indeed, a number of studies have examined the relation- ship of media consumption and sleep in young people, and have demonstrated that media use, especially in the late evening shortly before going to bed and overnight, increases the risk of sleep complaints, shorter sleep duration, irregu- lar sleep patterns, night awakening, and daytime sleepiness (Adam, Snell, & Pendry, 2007; Arora, Broglia, Thomas, & Taheri, 2014; Carter, Rees, Hale, Bhattacharjee, & Paradkar, 2016; Cespedes et al., 2014; Gamble et al., 2014; Hale & Guan, 2015; Hysing et al., 2015; Kubiszewski, Fontaine, Rusch, & Hazouard, 2014; Li et al., 2007; Van den Bulck,
2004). As confirmed in a study by Garrison, Liekweg, and Christakis (2011), evening media use had a greater impact on sleep than daytime use, showing that each additional hour of near-bedtime media use was associated with a sig- nificant increase in sleep problems in preschool children. Interestingly, Bruni et al. (2015) found that evening chrono- types have more technological devices in their bedrooms, engage in more technology-related activities after 9 p.m., and have later turning off devices time at night compared with early chronotypes.
Considering the parallel increase in both, diagnosis of ADHD and young people’s media use, there is a growing body of evidence suggesting a potential link between these factors. A meta-analysis from Nikkelen, Valkenburg, Huizinga, and Bushman (2014) revealed a significant posi- tive correlation between television viewing/video gaming and composite ADHD-like behavior, attention problems, and impulsivity. Findings from longitudinal studies identi- fied television viewing and video game playing in early childhood as potential contributors to greater subsequent attention problems (Gentile, Swing, Lim, & Khoo, 2012; Landhuis, Poulton, Welch, & Hancox, 2007; Swing, Gentile, Anderson, & Walsh, 2010).
Although there is still no comprehensive information about media use in ADHD, the scant studies undertaken so far indicate that individuals with ADHD are more attracted to electronic media and engage in more screen time than typically developed peers (Acevedo-Polakovich, Lorch, & Milich, 2007; Bolic Baric, Hellberg, Kjellberg, & Hemmingsson, 2018). Examining the association of the availability of a bedroom TV with total screen time among 6- to 17-year-olds with ADHD revealed that they spend 150 min per weekday watching TV, watching videos, or playing video games (Lo, Waring, Pagoto, & Lemon, 2015). As high exposure to screen-based media may exacerbate behavioral problems and attention difficulties (Johnson, Cohen, Kasen, & Brook, 2007), these results emphasize the necessity of understanding and limiting screen-based media use in youth affected by this disorder.
As previously mentioned, insufficient sleep duration and disturbed sleep are considered as risk factors for behavioral problems among children. In turn, individual sleeping behavior is highly influenced by sleep hygiene, including various practices and habits that are intended to promote a good sleep quality (Mindell, Meltzer, Carskadon, & Chervin, 2009). Especially prebedtime screen-based media use is a common activity that is not in line with principles of good sleep health. It is, therefore, of particular interest to study how these variables interrelate, as to date, surpris- ingly, little data exist focusing on the association between prebedtime media exposure and behavioral problems.
One of the few studies related to the broader topic was provided by Arns, van der Heijden, Arnold, and Kenemans (2013), who examined the relationship of solar intensity
578 Journal of Attention Disorders 24(4)
with geographical variation in ADHD rates. Their results showed lower prevalence in areas with higher solar inten- sity. An apparent preventive effect of high sun intensity on ADHD was assumed, explaining 34% to 57% of the vari- ance in ADHD prevalence across 49 U.S. states and nine non-U.S. countries. According to Arns and colleagues, this protective effect is due to the strong phase-advancing impact of natural bright light during the morning, improv- ing circadian clock disturbances associated with ADHD. Use of electronic media shortly before bedtime might delay circadian rhythms even further through DLMO suppres- sion, which, in turn, results in delayed sleep onset and short- ened sleep duration (Arns et al., 2013). Furthermore, Arns and coworkers assume that at least a subgroup of children with ADHD are more vulnerable to sleep difficulties mani- fested by symptom exaggeration. As a conclusion, this sub- group would benefit from reduced evening use of modern media emitting light in the blue spectrum and increased daytime light exposure (especially in the morning) through counteracting delayed circadian phase (Arns et al., 2013).
Taken together, the current state of knowledge suggests that there may be a correlation between (prebedtime) media use, sleep difficulties, and manifestation of ADHD- related symptoms. However, there is still limited research, with the majority of studies examining solely media use or sleep and not assessing the association of these two factors with behavior and ADHD symptoms. The reported find- ings prompted us to hypothesize that more use of elec- tronic media through self-luminous displays will be associated with impaired sleep quality and increased behavioral problems. Furthermore, considering the high rates of sleep complaints and disturbed circadian rhythms in individuals with ADHD, their high exposure to screen- based media, the importance of light in regulating sleep/ wake cycle, and the vital necessity of sleep for optimal daytime functioning, we presume stronger relationships between time spent with screen-based media, sleep prob- lems, and ADHD-like behavior in patients than in nonre- ferred children and adolescents. However, due to the strong impact of photic stimulation on neurophysiological processes and the dramatic increase in media consumption among young people, an association is also anticipated in healthy individuals. For this reason, a case-based as well as a community-based approach was conducted, allowing both categorical and dimensional conceptualization of ADHD symptomatology.
Therefore, we aimed to (a) investigate differences between ADHD patients and healthy controls regarding sleep behavior, use of screen-based media, and parameters of circadian rhythm; (b) analyze the relationship between sleep, screen-based media use, circadian rhythm, and ADHD-related behavior in a community sample; and (c) test the contribution of sleep habits and screen-based media use to the manifestation of symptoms of ADHD in healthy
children and adolescents. Our cross-sectional study will, thus, assess the relationships between the relevant con- structs, media consumption, sleep, and ADHD, but not test causal pathways toward ADHD, which would require a lon- gitudinal approach.
Method
Ethical approval was obtained from the ethics committee of the Albert Ludwigs University of Freiburg, Germany. All participants and their parents signed an informed consent before enrolment in the study. Data were collected from June 2014 to June 2016.
Participants
This study included N = 357 children and adolescents from the general population and N = 61 children and adolescents with ADHD, both aged 8 to 18 years. Healthy participants were recruited through local schools and after-school pro- grams, and through announcements in regional publications and social media. Snowball sampling was also used, whereby the initial participants acted as informants and rec- ommended further eligible participants for the study. Of the participants with ADHD, 53 (87%) were outpatients of the Department of Child and Adolescent Psychiatry at the University Medical Centre Freiburg, where diagnosis of ADHD is based on criteria defined by the International Classification of Diseases-10 (ICD-10; World Health Organization, 1992). The remaining eight (13%) patients had received their diagnosis from office-based child psy- chiatrists. Psychostimulants were taken by 47 (77%) patients, but as this study was part of a larger study includ- ing a computer-based test battery on cognitive performance, they were free of medication at least 24 hr before the study day. All participants should have an IQ score 70 as assessed with the Culture Fair Intelligence Test 20–Revision (CFT 20-R; Weiß, 2006). Exclusion criteria for the healthy participants were (a) any neurological or psychiatric disor- ders and (b) pronounced developmental delay. Exclusion criteria for patients with ADHD were (a) diagnosis of an autism spectrum disorder (ASD) or (b) pronounced devel- opmental delay. Of the whole study population, two sam- ples were formed to test the hypotheses. The matched sample (N1 = 122) was composed of N = 61 patients with ADHD and N = 61 healthy controls. Gender, age, and IQ score are shown in Table 1 for each sample, respectively. In the matched sample, there were no significant differences in gender distribution, mean age, or IQ (all ps > .20). The community sample (N2 = 377) included all 357 healthy par- ticipants plus, according to the prevalence of ADHD of 5.29% (Polanczyk et al., 2007), 20 randomly selected patients from the ADHD sample. Of those, 14 (70%) patients received stimulant medication.
Thoma et al. 579
Measures
Sleep habits questionnaire. A self-report questionnaire based on the School Sleep Habits Survey (SSHS) developed by Wolfson and Carskadon (1998) was used to assess partici- pants’ usual sleep-wake behavior over the past 2 weeks. The SSHS has shown good validity and high correlation with other self-report measures (Wolfson et al., 2003). Two age-specific versions for children (8-11 years) and adoles- cents (12-18 years) were handed out. Information on bed- time, time to fall asleep, and total sleep time were obtained for school and weekend nights separately. Deviation from individual sleep requirement was calculated as deviation of total sleep time from self-assessed optimal sleep duration, both on school and weekend nights. The number of night awakenings was categorized (never/once/2 or 3 times/more than 3 times/I have no idea), and participants who ticked I have no idea were excluded from the data analysis. Day- time sleepiness (no problem at all/a little problem/more than a little problem/a big problem/a very big problem) was also assessed. In addition, the adolescents’ (12-18 years) version of the questionnaire contained a morningness/eve- ningness scale, which is composed of 10 multiple-choice items assessing the circadian preference for certain activi- ties (derived from Smith, Reilly, & Midkiff, 1989). Total score for determining one’s chronotype ranges from 10 (evening type) to 43 (morning type). Díaz-Morales, de León, and Sorroche (2007) suggest cutoffs at the 20th and 80th percentiles (evening type = 10-20, intermediate type = 21-27, morning type = 28-43). Cronbach’s alpha of the morningness/eveningness scale is .78 (Acebo & Carskadon, 2002). Furthermore, adolescents completed a delay scale, asking about the frequency of six typical behaviors that are likely to be related to delayed sleep phase: arrived late to class because you overslept, stayed up until at least 3 a.m., stayed up all night, slept in past noon, needed more than one reminder to get up in the morning, had an extremely hard time falling asleep. Frequency of each behavior was rated on a 5-point scale from 0 (never) to 4 (every day/night)
and were summed to obtain a total score. Cronbach’s alpha of the Delay scale is .70 (Acebo & Carskadon, 2002).
Media use questionnaire. Data about participants’ screen- based media use habits were collected using an adapted ver- sion of the screen time-based sedentary behavior questionnaire from the HELENA study (Rey-López et al., 2012). As demonstrated by the developers, reliability was found to be good to excellent. Two age-specific versions for children (8-11 years) and adolescents (12-18 years) were administered. To assess participants daily media use, they had to report their habitual time spent with screen-based media (television viewing/video games/computer/mobile phone) during both school and weekend days by ticking the fitting category (no time/up to 30 min/30 min to 1 hr/1 to 2 hr/2 to 3 hr/3 to 4 hr/4 to 5 hr/more than 5 hr). Similarly, information on daily media use in the 2 hr before going to bed was obtained: Children aged 8 to 11 years were asked how often (never, once a month, once a week, 2-3 times a week, 4-6 times a week, every evening) and how long (no time, 5-15 min, 15-30 min, 30 min to 1 hr, more than 1 hr) they usually use electronic devices before bedtime. Adoles- cents aged 12 to 18 years were asked how many time they spend with screen-based media in the last 2 hr before bed- time (television viewing/video games/computer/mobile phone) with nine categories of response (no time/up to 15 min/15-30 min/30-45 min/45-60 min/1 to 1 hr 15 min/1 hr 15 min to 1 hr 30 min/1 hr 30 min to 1 hr 45 min/1 hr 45 min to 2 hr). If this sum exceeds the threshold of 2 hr, it was capped at the peak value of 2 hr. Furthermore, the presence of screen-based devices in the bedroom (television/gaming console/computer) was assessed through yes/no questions.
Strengths and Weaknesses of ADHD symptoms and Normal behavior (SWAN) scale. Participants’ ADHD-like behavior was measured by using the German version of the SWAN scale (Schulz-Zhecheva et al., 2019; Swanson et al., 2006). Based on the diagnostic criteria for ADHD listed in Diag- nostic and Statistical Manual of Mental Disorders (4th ed.;
Table 1. Study Samples: Participant Characteristics Gender, Age, and IQ Score.
Matched sample (N1 = 122)
Community sample (N2 = 377) ADHD (N = 61) Controls (N = 61)
Gender (m/f)
12.93 (±2.13) 8.96-17.80
12.97 (±2.22) 8.46-18.13
12.72 (±2.83) 8.09-18.96
94.64 (±13.35) 69a-124
97.51 (±11.22) 75-125
108.52 (±14.63) 69a-155
2.44 (±0.68) 0.89-4.72
3.84 (±0.95) 1.78-5.89
3.71 (±0.90) 1.39-6.00
Notes. m = male; f = female; SWAN-TOT = SWAN total scale; SWAN = Strengths and Weaknesses of ADHD symptoms and Normal behavior. aPrior IQ test score was 86.
580 Journal of Attention Disorders 24(4)
DSM-IV; American Psychiatric Association, 1994) and Diagnostic and Statistical Manual of Mental Disorders (5th ed., DSM-5; American Psychiatric Association, 2013), the SWAN scale contains 18 items, with Items 1 to 9 corre- sponding to the Attention Deficit subscale and items 10 to 18 representing the Hyperactivity/Impulsivity subscale. In contrast to other behavior rating scales, the items of the SWAN scale are reworded using a neutral to strength-ori- ented formulation. Parents were asked to quantify each item regarding the severity of ADHD symptoms manifested in their child compared with other children of the same age by using a 7-point Likert-type scale ranging from 0 (far below average) to 6 (far above average). Due to the balanced scoring system, the SWAN scale assesses symptoms of ADHD in a dimensional manner. Dividing participants’ total scores by the number of all items produces the score of the SWAN total scale (SWAN-TOT) and ranges from 0 to 6, with low scores indicating higher level of ADHD symptoms and symptom load. The German version of this scale has recently been proven to show high reliability with an over- all Cronbach’s alpha…