Video Game Use in Boys With Autism

Video Game Use in Boys With Autism SpectrumDisorder, ADHD, or Typical Development WHAT’S KNOWN ON THIS SUBJECT: Children with autism spectrum disorder (ASD) and those with ADHD are at risk for problematic video game use. However, group differences in media use or in the factors associated with problematic video game use have not been studied. WHAT THIS STUDY ADDS: Boys with ASD and ADHD demonstrated greater problematic video game use than did boys with typical development. Inattention was uniquely associated with problematic use for both groups, and role-playing game genre was associated with problematic use among the ASD group only. abstract OBJECTIVES: The study objectives were to examine video game use in boys with autism spectrum disorder (ASD) compared with those with ADHD or typical development (TD) and to examine how specific symp- toms and game features relate to problematic video game use across groups. METHODS: Participants included parents of boys (aged 8–18) with ASD (n = 56), ADHD (n = 44), or TD (n = 41). Questionnaires assessed daily hours of video game use, in-room video game access, video game genres, problematic video game use, ASD symptoms, and ADHD symptoms. RESULTS: Boys with ASD spent more time than did boys with TD playing video games (2.1 vs 1.2 h/d). Both the ASD and ADHD groups had greater in-room video game access and greater problematic video game use than the TD group. Multivariate models showed that inattentive symptoms predicted problematic game use for both the ASD and ADHD groups; and preferences for role-playing games predicted problematic game use in the ASD group only. CONCLUSIONS: Boys with ASD spend much more time playing video games than do boys with TD, and boys with ASD and ADHD are at greater risk for problematic video game use than are boys with TD. Inattentive symptoms, in particular, were strongly associated with problematic video game use for both groups, and role-playing game preferences may be an additional risk factor for problematic video game use among children with ASD. These findings suggest a need for longitudinal research to better understand predictors and outcomes of video game use in children with ASD and ADHD. Pediatrics 2013;132:260–266 AUTHORS: Micah O. Mazurek, PhD,a,b and Christopher R. Engelhardt, PhDc Departments of aHealth Psychology, and cPsychological Sciences, University of Missouri, Columbia, Missouri; and bThompson Center for Autism and Neurodevelopmental Disorders, Columbia, Missouri KEY WORDS video game, video game addiction, autism, autism spectrum disorder, attention-deficit/hyperactivity disorder, ADHD ABBREVIATIONS ANOVA—analysis of variance ADHD—attention-deficit/hyperactivity disorder ASD—autism spectrum disorder PVGT—Problem Video Game Playing Test SCQ—Social Communication Questionnaire TD—typical development VADPRS—Vanderbilt Attention Deficit/Hyperactivity Disorder Parent Rating Scale Dr Mazurek conceptualized and designed the study, coordinated and supervised data collection, carried out the initial analyses, and drafted the initial manuscript; Dr Engelhardt contributed to the analysis plan and reviewed and revised the manuscript; and both authors approved the final manuscript as submitted. www.pediatrics.org/cgi/doi/10.1542/peds.2012-3956 doi:10.1542/peds.2012-3956 Accepted for publication May 2, 2013 Address correspondence to: Micah Mazurek, PhD, Department of Health Psychology, University of Missouri, Thompson Center for Autism and Neurodevelopmental Disorders, 205 Portland St, Columbia, MO 65211. E-mail: mazurekm@missouri.edu PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275). Copyright © 2013 by the American Academy of Pediatrics FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose. FUNDING: All phases of this study were supported by a grant from the University of Missouri Research Board. 260 MAZUREK and ENGELHARDT Childrenwith autismspectrumdisorder (ASD)and thosewithADHDareat risk for preoccupation with video games. Stud- ies have shown that children with ASD spend substantial amounts of time playing video games,1–3 have difficulty disengaging from them,4 and show higher levels of problematic (addictive) video game use than do children with typical development.5 Similarly, children with ADHD exhibit greater problematic video game use than do children with TD,6 and ADHD symptoms are associated with greater time spent on video games7,8 and with greater problematic game use.8 Han et al9 also found a re- duction in problematic video game use among children with ADHD after an 8-week trial ofmethylphenidate. This line of research suggests that the behav- ioral and neurobiological mechanisms of ADHD may be closely related to problematic video game use, as dis- cussed in 2 recent reviews on the re- lationship between ADHD symptoms and video game addiction.10,11 Although video game addiction is not a formal di- agnosis, studies have found that it is characterized by the same features as other behavioral addictions, including salience, tolerance, withdrawal, relapse, mood modification, and conflict.12–14 Ad- ditionally, the problematic video gameuse construct can be reliably assessed,15,16 has good construct validity,17 and is associated with negative functional outcomes.18 The primary diagnostic features of ASD include impairment in social and com- munication skills and engagement in restricted/repetitive behaviors.19 These core symptomsmay be directly related to the development of problematic video game play patterns. Specifically, a ten- dency to develop encompassing pre- occupations or intense interests may lead to difficulty disengaging from video games. The audiovisual and structural features of video games, coupled with their relative lack of social demands, may also make them particularly reinforcing for children with ASD. However, symp- toms of inattention and hyperactivity are also common co-occurring problems for children with ASD,20–22 which may also relate to problematic game use. Given that children with ASD and ADHD have distinct diagnostic features, yet share common symptoms and are at risk for problematic video game use, a better understanding of the relations among these variables across diagnostic groups is important. Surprisingly, studies have not yet examined how these groups compare in video game play or whether the correlates of problematic game use are similar across groups. The current study investigated these issuesby (1) comparingvideogameuse in boys with ASD, ADHD, or TD and (2) examining the relative contribution of specific symptom types and game fea- tures in predicting problematic video game use. Due to the significant gender differences inASDandADHDprevalence and phenotypes23–26 and the increased risk for problematic game use among boys,12,18 the current study examined video game use among boys only. Thiswas thefirst study toexamine these issues, so competing hypotheses were tested. Specifically, if inattention and hyperactivity largely account for vari- ability in problematic video game use across groups, no differences in video game use would be expected when comparing ASD and ADHD groups, but significant differences would be expected (for both groups) compared with TD children. Alternatively, if prob- lematic video game use is more closely related to core ASD symptoms, differ- ences would be expected when com- paring ASD and ADHD groups. METHODS Participants The sample included parents of 56 boys with ASD, 44 boys with ADHD, and 41 boyswith TD, ranging in age from8 to 18 (mean 11.7 years, SD 2.5 years). See Table 1 for demographic information. Participants with ASD were recruited through an academic medical center specializing in ASD diagnosis and treatment and had been previously di- agnosed with an ASD, including autistic disorder (46.4%), Asperger’s disorder (25.0%), or pervasive developmental disorder not otherwise specified (PDD NOS) (28.6%), according to the center’s clinical care model. The standard di- agnostic battery generally includes evaluations by a physician and/or psychologist and the use of standard- ized tools, including the Autism Di- agnostic Observation Schedule27 and/or Autism Diagnostic Interview–Revised.28 Only 4 participants had an IQ of #70, and only 2 were reported by parents to have no current use of phrase speech. Participants with ADHD were recruited through a university-affiliated develop- mental and behavioral pediatrics clinic, and they all had been previously di- agnosed with ADHD. See Table 2 for a list of current medications. Participants in the TD group were recruited through the community with the use of fliers and word-of-mouth recruitment and had no previous di- agnosis of ASD, ADHD, or other de- velopmental disorder as reported by parents. TDgroupparticipants fellbelow clinical cutoffs on diagnostic screening measures (described later) and were not taking medications for behavioral or developmental concerns. Measures Video game use was assessed by using a questionnaire designed for the cur- rent study. Parents reported the num- ber of hours per day their child spent “playing video or computer games” during out-of-school hours (data were collected during months in which school was in session). Parents pro- vided separate estimates for “typical” weekday and weekend days. An average ARTICLE PEDIATRICS Volume 132, Number 2, August 2013 261 daily use variable was created by mul- tiplying the weekday value by 5 and the weekend value by 2, and dividing the sum of these values by 7, consistent with previous research.1,5 Parents were also asked: “Does your child have a video game system in his room?” Parents also listed their child’s 3 most commonly played video games during the past month. Game titles were coded into mutually exclusive genre catego- ries based on descriptions from the Entertainment Software Rating Board website and were cross-referenced with descriptions from 2 popular gam- ing websites (Gamespot and IGN). Genre categories included (1) action/action- adventure, (2) adventure, (3) role-playing, (4) strategy, (5) puzzle/mini-game, (6) educational, (7) fighting, (8) first-person shooter, (9) music, (10) platform, (11) racing, (12) simulation, (13) sports simulation, and (14) fitness. Problematic video game use was assessed by using amodified version of the Problem Video Game Playing Test (PVGT).15 The original PVGT was de- veloped as a self-report measure of problematic video game use based on the behavioral addiction model13 but was modified in a previous study for use with children.5 This parent-report version includes 19 items rated on a 4-point scale ranging from 1 (Never) to 4 (Always); total PVGT scores represent a sum of all items. Cronbach’sa ranged from 0.90 to 0.94 across groups. Inattention and hyperactivity/impulsivity symptoms were assessed by using the Vanderbilt Attention Deficit/Hyperactivity DisorderParentRatingScale (VADPRS).29 Items are rated on a 4-point scale ranging from 0 (Never) to 3 (Very Often), and the Inattention (9 items), Hyperac- tivity (9 items), and ADHD Total scale scores were examined. The VADPRS has good internal consistency, validity, and reliability30 and has been used in ASD studies.31 Cronbach’s a ranged from 0.87 to 0.94 across groups. Current ASD symptoms were assessed by using the Social Communication Questionnaire–Current (SCQ).32 The SCQ is a parent-report questionnaire with strong reliability and validity.32–34 The Current form was used, focusing on behavior over the most recent 3-month period. Items are rated as “Yes” or “No” (total possible scores range from 0 to 39). Cronbach’s a ranged from 0.62 to 0.83 across groups. RESULTS Sample Characteristics Groups did not differ by age, race, or number of siblings. Group differences in parent marital status and household income were observed (see Table 1). A 1-way analysis of variance (ANOVA) revealed that SCQ scores differed sig- nificantly across groups, F(2, 138) = 45.5, P , .001. Tukey post-hoc com- parisons showed that the ASD group had higher SCQ scores (mean 13.7, SD 5.6) than both ADHD (mean 8.1, SD 5.4, P, .001) and TD groups (mean 4.4, SD 2.4, P , .001). The ADHD group also had higher scores than the TD group (P = .001). A 1-way ANOVA also showed group differences on the ADHD Total score: F(2, 138) = 60.2, P, .001. Tukey post-hoc comparisons showed that the ADHD group exhibited higher ADHD To- tal scores (mean 30.3, SD 12.1) than the TD group (mean 7.5, SD 5.9, P , .001). However, the ADHD and ASD (mean 28.0, SD 12.0) groups did not differ. Video Game Use We examined diagnostic group differ- ences in daily video game hours using the general linear model while controlling for household income (di- chotomousvariable, 1= income#$80 000, and 2 = income .$80 000) and parent marital status (dichotomous variable, 1 = married, and 2 = not married). The results revealed significant diagnostic group differences in daily video game hours (Table 3). Bonferroni post-hoc comparisons revealed that boys with ASD spent more time playing video games than did boys with TD (P = .01, TABLE 1 Sample Characteristics ASD (n = 56) ADHD (n = 44) TD (n = 41) P Age, y, mean (SD) 11.7 (2.6) 11.1 (2.4) 12.2 (2.4) .14 No. of siblings, mean (SD) 2.2 (1.6) 1.8 (1.2) 1.9 (1.3) .37 Race, % .23 White 80.4 86.4 92.7 Nonwhite 19.6 13.6 7.3 Household income, % ,.001 ,$10 000 15.1 7.0 2.4 $10 000–$20 000 7.5 18.6 0 $21 000–$30 000 11.3 25.6 0 $31 000–$40 000 7.5 16.3 0 $41 000–$80 000 32.1 20.9 17.1 .$80 000 26.4 11.6 80.5 Parent marital status, % .001 Married 55.4 61.4 90.2 Separated 1.8 2.3 0 Divorced 30.4 11.4 7.3 Never married 12.5 25.0 2.4 Group comparisons for continuous variables were conducted by using ANOVA; group comparisons for categorical variables were conducted by using x2 tests. TABLE 2 Psychotropic Medication Use Across Groups ASD (n = 56) ADHD (n = 44) TD (n = 41) Anticonvulsant 10.7% 4.5% 0% a-Agonist 23.2% 29.5% 0% Antipsychotic 28.6% 11.4% 0% Stimulant 33.9% 70.5% 0% SSRI 16.1% 6.8% 2.4% SNRI 3.6% 6.8% 0% SSRI, selective serotonin reuptake inhibitor; SNRI, seroto- nin-norepinephrine reuptake inhibitor. 262 MAZUREK and ENGELHARDT d = .80), whereas boys with ADHD did not differ significantly from either group. In-Room Video Game Access The x2 tests revealed significant group differences for in-room video game access (Table 4). Post-hoc bivariate x2 tests showed that a greater percent- age of both the ASD and ADHD groups had in-room video game systems compared with the TD group, (P = .001 and P = .002, respectively), whereas the ASD and ADHD groups did not differ. Video Game Genre Dichotomous variables (coded 0 = no, 1 = yes) were created to indicate whether each game genre was 1 of the top 3 most frequently played. To reduce thepossibility of Type I error, aBonferroni correction was applied by dividing the a level (a = .05) by the number of total comparisons (14), yielding an adjusted significance level (a = .004). Two-tailed Fisher exact tests comparing all 3 groups revealed significant differences in the shooter and sports genres (see Table 4). Post-hoc bivariate Fisher exact tests showed that a greater percent- age of the TD (P = .002) group preferred shooter games compared with the ASD group. Also, a greater percentage of the TD group preferred sports games compared with the ADHD group (P = .001). Problematic Video Game Use Diagnostic group differences in PVGT score were examined by using the general linear model, controlling for household income (dichotomous), pa- rental marital status (dichotomous), anddaily video gamehours (a correlate of problematic game use in previous research12 and r = 0.51, P, .001 in the current study). The results indicated diagnostic group differences in PVGT score (P = .001). Bonferroni post-hoc comparisons indicated that the ASD and ADHD groups had higher PVGT scores than the TD group (P = .001, d = 1.12, and P = .03, d = .98, respectively). However, the ASD and ADHD groups did not differ (see Table 3). Regarding video game genre prefer- ence, 1-way ANOVA revealed that boys who played role-playing games had higher PVGT scores (mean 45.0, SD 11.9) than those who did not (mean 35.6, SD 10.7): F(1, 142) = 11.2, P = 001. A fac- torial ANOVA showed that the di- agnostic group 3 role-playing game interaction was nonsignificant, F(2, 135) = 1.7, P = .18, suggesting a similar relationship between role-playing preferences and PVGT scores across groups. Associations Between Symptoms and Problematic Video Game Use Subsequent analyses were conducted separately by group to examine the relative contribution of particular symp- toms in predicting problematic video game use. ASD Group PVGT score was correlated with daily video game hours (r = 0.36, P = .008), inattention (r = 0.48, P , .001), and hyperactivity (r = 0.38, P = .004) but not with SCQ (r = 20.01, P = .97). To determine the relative contribution of these variables to PVGT score, a simulta- neous linear regression was conducted with PVGT score as the dependent variable and daily video game hours, role- playing genre, inattention, and hyper- activity as independent variables. The results indicated that daily video game hours (b = .29, t = 2.6, P = .01), role- playing genre (b = .25, t = 2.2, P = .03), and inattention (b = .43, t = 2.8, P = .01) remained significant predictors of PVGT, whereas hyperactivity was not a signifi- cant predictor (P = .54) when control- ling for the other variables in themodel. ADHD Group PVGT score was correlated with daily video game hours (r = 0.47, P = .002) and inattention (r = 0.37, P = .01) but not with hyperactivity (r = 0.20, P = .19). Linear regression results indicated that daily video game hours (b = .49, t = 3.7, P = .001) and inattention (b = .36, t = 2.1, P = .04) remained significant predictors of PVGT, whereas hyperac- tivity and role-playing genre (P = .97 and P = .70, respectively) were not. TD Group PVGT score was correlated with daily video game hours (r = 0.70, P , .001), inattention (r = 0.55, P , .001), and hyperactivity (r = 0.50, P = .001). Linear regression results showed that daily video game hours (b = .59, t = 5.9, P, .001) and hyperactivity (b = .32, t = 2.9, P = .006) remained significant pre- dictors of PVGT, whereas inattention and role-playing genre (P = .07 and P = .87, respectively) were not. DISCUSSION This is the first study to examine video game use among boys with ASD com- pared with those with ADHD or TD. Consistent with previous findings,5 boys with ASD spent significantly more time than did boys with TD playing video games, and this alone (2.1 h/d) TABLE 3 Group Differences in Video Game Use ASD (n = 56) ADHD (n = 44) TD (n = 41) P h2p Video game, h/d, mean (SD) 2.1 (1.3)a 1.7 (1.1) 1.2 (0.9)a .01 .070 PVGT, mean (SD) 41.2 (12.7)a 37.7 (9.2)b 29.6 (7.1)ab .001 .107 Group comparisons were conducted by using the general linear model, controlling for household income, parent marital status, and video game hours per day (for the PVGT model). Groups with matching superscripts within rows differed at the P , .05 level in Bonferroni post-hoc comparisons. Unadjusted means and standard deviations are reported. Partial h squared (h2p) is reported for effect sizes. ARTICLE PEDIATRICS Volume 132, Number 2, August 2013 263 exceeded the American Academy of Pediatrics’ recommendation for com- bined screen-based media exposure.35 The estimates of daily video game time for both ASD and TD groupswere highly consistent with those from another recent study of similarly aged children5 and extend those findings by including a comparison group of boys with ADHD. Estimates of video game hours per day in children with ADHD have not been previously reported, yet our findings were also consistent with a small study showing that children with ADHD and TD did not differ on broader measures of game play frequency o