Executive Functioning in Everyday Life in Youth with Attention-Deficit/Hyperactivity Disorder and Autism Spectrum Disorders: A Two-Year Longitudinal Study

ABSTRACT Problems with executive function (EF) are considered a hallmark of Attention-Deficit/Hyperactivity Disorder (ADHD) and Autism Spectrum Disorder (ASD). However, little is known about the developmental pathway of everyday EF in these two groups, and whether the two groups follow a similar or different developmental course. In this longitudinal study, children and adolescents with ADHD (n = 84, Mage = 11.6, SD = 2.0), ASD (n = 38, Mage = 12.0, SD = 2.3), and typically developing children (TDC; n = 50, Mage = 11.6, SD = 2.0) were clinically diagnosed and assessed with parent-ratings of everyday EF at baseline and at two-year follow-up (97% retention). Results showed that both individuals with ADHD and individuals with ASD displayed elevated levels of everyday EF problems relative to the TDC at baseline (Hedges g = 1.99 to 2.99). Over the two-year period, everyday EF improved in individuals with ADHD relative to the TDC, whereas individuals with ASD displayed no improvement relative to the TDC. At two-year follow-up, individuals with ADHD and individuals with ASD continued to display elevated levels of everyday EF problems relative to the TDC (Hedges g = 1.61 to 3.19).

Converging empirical evidence indicates that children and adolescents with ADHD or ASD display elevated levels of EF problems in everyday life compared to typically developing children (TDC) (Gioia, Isquith, Kenworthy, & Barton, 2002;Hovik et al., 2017;Kouklari, Tsermentseli, & Monks, 2018;Lawson et al., 2015;Mahone et al., 2002;Rosenthal et al., 2013;Semrud-Clikeman, Walkowiak, Wilkinson, & Butcher, 2010;Sergeant, Geurts, & Oosterlaan, 2002;Skogli et al., 2013).However, the developmental trajectories of everyday EF in children and adolescents with ADHD or ASD remain unclear due to the scarcity of longitudinal studies.One two-year follow-up study of 39 children with ASD without intellectual disability, ages 7-14, reported significant impaired parent rated BRIEF scores relative to TDC, and no significant improvement across time relative to 34 TDC (i.e.EF problems persist across time) (Vogan et al., 2018).Another cross-sectional study of 185 children and adolescents with ASD, ages 5 to 18 years, reported a general aggravation of everyday EF problems with increasing ages based on BRIEF parent ratings (Rosenthal et al., 2013).In particular, BRIEF subscales forming the Metacognition Index were reported to show increased problems with age in the study by Rosenthal and colleagues (2013).A comparable decline in metacognitive skills with increasing ages was reported in a cross-sectional study of 57 children ages 7-15 years with ASD, based on BRIEF teacher ratings (Kouklari et al., 2018).

Knowledge Gaps
Several studies have examined the developmental trajectories of performance-based EF in children and adolescents with ADHD or ASD, but longitudinal studies examining the developmental trajectories of everyday EF are warranted (Hill, Berthoz, & Frith, 2004;Velikonja, Fett, & Velthorst, 2019).To the best of our knowledge, no studies investigating the developmental trajectories of everyday EF assessed with BRIEF or other behavioural ratings of everyday EF in children and adolescents with ADHD have been conducted.
As environmental demands typically increase from childhood into adolescence, it has been reported that functional impairment in individuals with neurodevelopmental disorders may aggravate with age (Turgay et al., 2012).Consequently, as BRIEF has been reported to correlate significantly with impairment level (Kouklari et al., 2018;Pugliese et al., 2015Pugliese et al., , 2016)), one possibility is that every day EF problems may increase with age in line with increasing environmental demands.At the same time, however, empirical evidence indicates that everyday EF is more closely related to ADHD symptoms and general psychopathology symptoms than to performance-based EF (McAuley, Chen, Goos, Schachar, & Crosbie, 2010).Consequently, everyday EF may develop separately from performance-based EF.In this context, it is worth noting that previous research has shown that ADHD symptoms and co-occurring psychopathology are declining in both individuals with ADHD and individuals with ASD from childhood and into young adulthood (Andersen et al., 2015a(Andersen et al., , 2017;;Orm et al., 2021;Skogli, Andersen, Hovik, & Oie, 2017;Stringer et al., 2020).Knowledge about the developmental course of everyday EF relative to performance-based EF may shed light on whether phenotypic EF (e.g.everyday EF) or endophenotypic EF (e.g.performance-based EF) follow separate or common developmental pathways.Based on the current literature, at least three different hypotheses about the development of everyday EF in individuals with ADHD and individuals with ASD from childhood to adolescence can be put forward: (1) based on studies of performance-based EF and the findings of Vogan et al. (2018), one would expect everyday EF to show stability over time in the ASD group (i.e.no improvement compared to TDC), (2) based on the cross-sectional studies of Rosenthal et al. (2013) and Kouklari et al. (2018), one would expect everyday EF, in particular the metacognitive aspect, to worsen over time compared to TDC in the ASD group, and (3) based on the close relationship between everyday EF and ADHD/psychopathology, and the decline over time in these symptom domains (e.g.Andersen et al., 2017;Skogli et al., 2017), one would expect everyday EF to improve over time in the ADHD group compared with TDC.Clearly, longitudinal studies are needed to see which of these three hypotheses can be empirically supported for each group.
Further, longitudinal studies comparing the developmental trajectories of everyday EF in individuals with ADHD with individuals with ASD from childhood to adolescence are needed to understand the unique and shared behavioural trajectories of these two disorders.Similar developmental trajectories may indicate that communalities in endophenotypes (e.g.performance-based EF) contribute to the developmental trajectories of everyday EF across both groups, whereas differential trajectories may indicate that distinct factors in the two disorders play an important role.Finally, enhanced knowledge regarding the developmental trajectories of EF behaviour problems in youth with ADHD and ASD may be of relevance for clinicians and will help parents and educators anticipate developmental challenges and plan support and interventions accordingly.

Study Aims
The main aim of this study is to examine and compare the developmental trajectories of parent-reported everyday EF problems among children and adolescents with ADHD, children and adolescents with ASD, and TDC across a two-year period.Our first research question was whether we could replicate previous findings showing elevated levels of everyday EF problems in the ADHD and the ASD group compared to TDC.We hypothesised that the ADHD and ASD group would display elevated levels of EF problems on all BRIEF index scores compared to TDC at baseline and at two-year follow-up.Our second research question was whether the ADHD and the ASD group would display stability, worsening, or improvement over the two-year period.Based on the current literature, it was not possible to predict a clear hypothesis.Our third research question was whether the ADHD and the ASD group would display differential everyday EF within the two time points and over time.Based on the scarcity of previous studies comparing everyday EF in these two groups, it was not possible to predict a clear hypothesis.

Procedure and Participants
Detailed recruitment strategy and diagnostic assessment are described in detail in separate publications (Andersen, Hovik, Skogli, Egeland, & Oie, 2013;Skogli et al., 2013).Demographic characteristics are presented in Table 1.This study included children from 8 to 17 years of age, recruited as consecutive referrals from seven outpatient Child and Adolescent Mental Health Centres in Innlandet Hospital Trust, Norway for assessment of ADHD and ASD.
The TDC group was recruited from local schools in Innlandet County, and they were enrolled in regular school classes at normal grade levels.At baseline, all participants underwent a comprehensive assessment according to common clinical practice.Diagnostic assessment at baseline was based on the Kiddie-Schedule for Affective Disorders and Schizophrenia/Present and lifetime version (K-SADS; Kaufman et al., 1997).K-SADS were conducted separately for children/adolescents and parents.The diagnostic evaluation with K-SADS was supplemented with information from the ADHD Rating Scale IV (DuPaul, Power, Anastoupolous, & Reid, 1998), the Autism Spectrum Screening Questionnaire (Ehlers, Gillberg, & Wing, 1999) and the Child Behaviour Checklist/6-18 (Achenbach & Rescorla, 2001), all filled out by the parents.Normative data from the Autism Spectrum Screening Questionnaire (Ehlers et al., 1999), from the ADHD Rating Scale IV manual (DuPaul et al., 1998), and T-scores above 65 on the syndrome and DSM-oriented scales in the Child Behaviour Checklist/6-18 (Achenbach & Rescorla, 2001) were applied to assess clinical significance.Additional information about school functioning, both academic and socially, which is mandatory on referral, was  Psychiatric Association, 2000).Exclusion criteria for all participants included prematurity (< 36 weeks), IQ below 70, or any disease affecting the central nervous system.The TDC were screened for mental disorders with the K-SADS in separate interviews for children/ adolescents and parents.Participants in the TDC group could not have been treated for a mental disorder, have a psychiatric diagnosis, have had a head injury with loss of consciousness or known dyslexia.At baseline 84 children and adolescents with ADHD (M = 11.6 years, 38 females), 38 with ASD (M = 12.0 years, 6 females), and 50 TDC (M = 11.6 years, 18 females), were included in the study.Eight of the children and adolescents with ASD also met DSM-IV criteria for ADHD at baseline.In the ASD group, 31 were diagnosed with Asperger's syndrome and seven with pervasive developmental disorder -not otherwise specified.The three groups (ADHD, ASD, TDC) did not differ significantly with regard to age at baseline but displayed an unequal gender distribution (p < .05) at baseline.The Wechsler Abbreviated Scale of Intelligence (WASI) (Wechsler, 1999) was administered to estimate full-scale IQ (FSIQ) in all participants.The groups differed significantly with regard to FSIQ (F = 5.81, p = .004),and Bonferroni post-hoc analysis showed that the ADHD group scored below the TDC.On average, mothers of children in the TDC group had 1.9 years more education than mothers of children with ADHD, and 1.8 years more education than mothers of children with ASD.Follow-up assessment was conducted approximately 24 months following baseline assessment.One participant from the ASD group and four from the ADHD group declined participation at follow-up.Unfortunately, we do not have information relating to their reasons for declining follow-up participation.The three groups differed significantly with regard to FSIQ at follow-up (F = 9.01, p < .001).Bonferroni post-hoc analysis showed lower FSIQ scores in both clinical groups compared to TDC at follow-up.After inclusion at baseline, participants with ADHD and ASD received standard psychological and/or medical treatment.Forty-four of the participants with ADHD were receiving stimulant medication at follow-up.Five of the participants with ASD and comorbid ADHD were receiving stimulant medication at follow-up.

Measures
In the current study, parents were asked to complete the written form of the BRIEF Norwegian parent rating version.The BRIEF includes 86 items, each of which is rated on a three-point scale with the values: 1 = Never, 2 = Sometimes and 3 = Often.These 86 items are composed into eight clinical scales (Inhibition, Shift, Emotional Control, Initiate, Working Memory, Plan/Organise, Organisation of Materials and Monitor).These 8 clinical scales form three broad classifications of executive functioning -Behavioural Regulation Index (BRI; comprising the inhibition, shift, and emotional control subscales), Metacognition Index (MI; comprising the initiate, working memory, plan/organise, organisation of materials, and monitor subscales), as well as an overall Global Executive Composite (GEC; comprising all subscales) score (Gioia et al., 2000).BRIEF scores are converted into standardised T-scores (M = 50, SD = 10).T-scores ≥ 65 (+1.5 SD) are defined as EF problems in the clinical range (Gioia et al., 2000).Fallmyr and Egeland (2011) reported high internal consistency (Cronbach's α = .76− .92) on the Norwegian parent rating version of the BRIEF.These values are at the same level as Cronbach's α reported in the BRIEF manual (.80 − .98)(Gioia et al., 2000).In the current study, the main focus will be on the clinical scales, T-scores are based on the original norms.Elevated BRIEF T-scores indicate a higher degree of impairment.

Data Analyses
Data analyses were conducted using the statistical package SPSS for Windows, version 26.0 (IBM, SPSS, Inc., Chicago, IL).We set the global alpha value at .05 and used Bonferroni corrections to control for chance findings by reducing the alpha value proportionately to the number of comparisons being made for each analysis.T-scores were used when comparing BRIEF parent ratings across groups.Demographic characteristics were investigated using the Chi-square test for independence (nominal variables) and analysis of variance (ANOVA) (continuous variables) followed up by Bonferroni post-hoc tests for group comparisons when adequate.Linear Mixed Models (LMM) were conducted for each dependent variable to estimate the effect of group on everyday EF across time.BRIEF T-scores were used as dependent variables.The estimation was based on restricted maximum likelihood, and random intercepts.The parameters of main interest were the fixed effect interaction terms 'time x group', contrasting the changes in the groups over time.We assessed three group comparisons separately (ASD -TDC, ADHD -TDC, ASD -ADHD).Assessment of model fit was done by Akaike information criteria investigation.Additionally, we ran two separate LMM analyses where we added baseline FSIQ and mother's education as covariates.Mother's education level at baseline, was used as an indicator of socioeconomic status.To examine group differences at follow-up (i.e.BRIEF index scores), we used multivariate analysis of variance (MANOVA) with the three BRIEF index scores BRI, MI and GEC as dependent variables.A significant group effect (Wilks Lambda) was followed-up with ANOVAs for each dependent variable and Bonferroni posthoc tests.

Trajectories of Parent Rated EF
The BRI, MI and GEC results across group and time are displayed in Table 2, while the results from the LMM are displayed in Table 3.A visual depiction of each group's test results across time is shown for the BRI, MI and GEC tests separately in Figure 1, Figure 2 and Figure 3, respectively.The two clinical groups performed significantly (p = <.001) higher on everyday EF problems on all BRIEF indexes than the TDC group at baseline (ADHD: BRI Hedges' g = 1.99,MI Hedges' g = 2.85, GEC Hedges' g = 2.63; ASD: BRI Hedges' g = 2.87, MI Hedges' g = 2.83, GEC Hedges' g = 2.99).The baseline differences between the two clinical groups were non-significant.Both the ADHD and ASD group displayed a significant decrease (improvement) in BRI scores across time (negative slope, main effect).Only the ADHD group displayed decreased MI scores relative to the TDC group (negative slope, interaction group x time) indicating greater improvement across time relative to the TDC group.The ADHD group also showed a significant decrease (improvement) in GEC scores relative to the TDC group, not shown in the ASD group (negative slope, interaction group x time).Effect size estimates indicate a different trajectory between the ASD and ADHD groups for BRIEF index scores MI and GEC across time.For the ASD group, the change in BRIEF index scores MI and

Group Differences at Follow-Up
A MANOVA with BRIEF index scores BRI, MI and GEC at follow-up as dependent variables showed a significant effect of group (F (2, 160) = 21.467,p < .001,η p 2 = .329).Separate ANOVAs for each dependent variable showed a significant group effect in all three index scores (see Table 2).For BRI, post-hoc tests showed that the ADHD (p < .001,Hedges' g = 1.61) and the ASD (p < .001,Hedges' g = 3.07) group displayed significantly more problems compared with the TDC group.For MI, post-hoc tests showed that the ADHD (p < .001,Hedges' g = 2.42) and the ASD (p < .001,Hedges' g = 2.64) group displayed significantly more problems compared with the TDC group.For GEC, post-hoc tests showed that the ADHD (p < .001,Hedges' g = 2.28) and the ASD (p < .001,Hedges' g = 3.19) group displayed significantly more everyday EF problems compared with the TDC group.There were no significant differences between the ADHD and the ASD group in any of the index scores.

Discussion
This study is the first to compare the development of everyday EF across ADHD, ASD, and TDC over a two-year period.The findings from this two-year follow-up study provide the field with two pieces of new knowledge.First, the results showed that whereas everyday EF improves in the ADHD group relative to TDC, everyday EF problems remain stable in the ASD group relative to TDC.Although there were no significant differences between the slopes of the ADHD and the ASD group, this finding could suggest slightly different developmental pathways of everyday EF between individuals with ADHD and individuals with ASD.Second, the results indicate that both the ADHD and the ASD groups displayed elevated levels of everyday EF problems across time.The results in this longitudinal study thus support previous findings from cross-sectional studies.
The improvement in everyday EF in the ADHD group contrasts with earlier findings of stable performance-based EF problems relative to TDC over time (Barkley & Murphy, 2010b;Biederman et al., 2009;Coghill, Hayward, Rhodes, Grimmer, & Matthews, 2013;Skogli et al., 2017).This suggests that everyday (phenotypic) EF and performance-based (endophenotypic) EF might follow different developmental pathways in the ADHD group.Thus, it could be that everyday EF problems are more closely related to symptom domains (ADHD core symptoms and co-occurring psychopathology) than to performance-based EF (e.g.McAuley et al., 2010).Our data provide evidence for some overlap between questionnaires posed on the BRIEF and the ADHD Rating Scale-IV (DuPaul et al., 1998), indicating such an association between parent ratings of everyday EF and ADHD core symptoms (see Skogli, Egeland, Andersen, Hovik, & Oie, 2014).Previous findings from this project have indicated a significant decline in ADHD core symptoms and co-occurring psychopathology symptoms in the ADHD group from baseline to follow-up (Skogli et al., 2017), whereas only co-occurring symptoms, but not ASD core symptoms declined in the ASD sample of this study (Andersen, Skogli, Hovik, Egeland, & Oie, 2015a).This could also help explain the differences between the ADHD and the ASD group in their comparison with TDC.If everyday EF problems are closely tied to the core symptoms of the disorder (ADHD and ASD), then a decline in ADHD symptoms would be expected to be accompanied by an improvement in everyday EF, whereas stable ASD symptoms would be expected to be accompanied by stable everyday EF problems, in line with our findings.
Whereas it seems like phenotypic and endophenotypic EF follow differential pathways in the ADHD group, this seems less clear for the ASD group, who showed stable everyday EF problems during the two-year period.This is in line with findings of stable performance-based EF problems in this sample, and a 'working memory arrest' previously reported by Andersen and colleagues (2015b).Working memory is commonly defined as the ability to actively maintain and manipulate information (Baddeley, 2003), and working memory impairments in children and adolescents with ASD have been linked to manipulation shortcomings (Travers, Klinger, & Klinger, 2011).The working memory arrest displayed in the ASD group might explain the stability in the Metacognition Index of everyday EF.As the manipulation part of working memory is a prerequisite for successful mental shifting (Senn, Espy, & Kaufmann, 2004), the working memory arrest displayed by the ASD group might therefore be a driving factor in the Metacognition Index difficulties, representing problems with mentally shifting (Cirino et al., 2018).Furthermore, working memory is closely related to other EF (Best & Miller, 2010) and may be responsible for some of the persistence in phenotypic difficulties observed in the ASD group.However, more research is needed to understand whether the development of everyday EF is more related to ASD core symptoms or performance-based EF in the ASD group.
The different developmental pathways between the ADHD and the ASD group relative to the TDC group could indicate that unique factors in the disorders (e.g.core symptoms, treatment) rather than communalities in endophenotypic EF impairment, is responsible for the development of everyday EF.For example, both the decline in ADHD symptoms and the improvement of everyday EF in the ADHD group could be caused by stimulant treatment received by 54% of the participants in this group.In comparison, only 13% of the participants in the ASD group received stimulant treatment.Further, no medical treatment is established with reliable effects on ASD core symptoms or have been incorporated into clinical guidelines (NICE, 2013).In addition to stimulant treatment, participants received standard treatment after diagnosis at baseline in line with regional and national guidelines for ADHD and ASD in Norway.Treatment following baseline assessment was often individually tailored and may have differed amongst participants.Thus, different treatments could explain differences in the developmental pathways of everyday EF between the two groups.
In a clinician and educator perspective, it is important to note that despite improvements in everyday EF in the ADHD group, both the ADHD and the ASD groups continue to display elevated levels of EF problems at two-year follow-up, relative to TDC.The differences were associated with very large effect sizes, indicating high clinical and educational relevance, whereas the improvement in the ADHD group was associated with small effect sizes of moderate clinical and educational relevance.Thus, these findings suggest that everyday EF is an important target of interventions and educational support in children and adolescents with both ADHD and ASD.

Implications for Clinical and Educational Practices
Understanding the development of everyday EF in children and adolescents with ADHD or ASD is important for clinical and educational practices.For clinicians, our results suggest that assessment of everyday EF should be routine when diagnosing and treating children with ADHD or ASD.This information will be important for caregivers and educators, and to target treatment and support.Over the years, several intervention programs to improve EF have been developed (e.g.Andersen, Klausen, & Skogli, 2019;Jensen et al., 2021;Kenworthy et al., 2014;Skogli, Andersen, & Isaksen, 2020), many of which can be implemented in educational settings.Furthermore, simple strategies to compensate for children's and adolescents' everyday EF problems should be implemented by educators and parents.Such strategies include providing visual support, helping them to break down more complex tasks into smaller elements and helping them organise their everyday life by providing plans, checklists, and scripts for school work, social situations, and other situations that the child may find challenging.

Study Strengths and Limitations
A longitudinal design with a comparatively large sample of children with ADHD or ASD, and a high retention rate (97%), are among the strengths of this study.Furthermore, to our knowledge, no other study to date has followed up children with ADHD and children with ASD simultaneously in the same study.The inclusion of these two distinct, but similar, disorders in the same study constitutes a major strength as it allows for comparison of developmental pathways across these two disorders and TDC.However, a few limitations should also be mentioned.The sample of children with ADHD and ASD was restricted to clinic-referred children, and thus, represents those who are willing to seek help from Child and Adolescent Mental Health Centers.These children may not be representative of the general ADHD or ASD populations.The sample also comprised children from a relatively large age-span (8-17 years at baseline), but small standard deviations reflect that most participants were close to the mean age.Furthermore, as treatment following baseline assessment was individually tailored and may have differed amongst participants, this could potentially have had a differing impact on EF development in the participants.
the current manuscript critically.All consented to their names on the final manuscript.All authors read and approved the final manuscript

Figure 1 .
Figure 1.Behavioural Regulation Index mean T-scores by group and assessment time.Vertical bars denote 95% confidence intervals.Attention-Deficit/Hyperactivity Disorder.ASD: Autism Spectrum Disorder.TDC: Typically Developing Children.Elevated BRIEF T-scores indicate a higher degree of impairment, with T-scores of 65 and above considered to represent clinically significant areas of concern.
GEC, relative to the TDC, were not significant (MI Cohen's d = −0.16,GEC Cohen's d = −0.16),but for the ADHD group, there was a significant reduction (improvement) in BRIEF index scores, relative to the TDC: MI (Cohen's d = −0.45),and GEC (Cohen's d = −0.35)across time.

Figure 2 .
Figure 2. Metacognition Index mean T-scores by group and assessment time.Vertical bars denote 95% confidence intervals.Attention-Deficit/Hyperactivity Disorder.ASD: Autism Spectrum Disorder.TDC: Typically Developing Children.Elevated BRIEF T-scores indicate a higher degree of impairment, with T-scores of 65 and above considered to represent clinically significant areas of concern.

Figure 3 .
Figure 3. Global Executive Composite mean T-scores by group and assessment time.Vertical bars denote 95% confidence intervals.Attention-Deficit/Hyperactivity Disorder.ASD: Autism Spectrum Disorder.TDC: Typically Developing Children.Elevated BRIEF T-scores indicate a higher degree of impairment, with T-scores of 65 and above considered to represent clinically significant areas of concern.

Table 1 .
Demographic characteristics for ADHD, ASD and TDC: means and standard deviations, by group and assessment time.into the diagnostic evaluation.Diagnoses were considered positive if, based on a comprehensive evaluation of K-SADS, teacher information and rating scales, DSM-IV (American Psychiatric Association, 2000) criteria were met.If both ADHD and ASD DSM-IV criteria were met, the ASD diagnosis took precedence over the ADHD diagnosis (American ADHD: Attention-Deficit/Hyperactivity Disorder, ASD: Autism Spectrum Disorder, TDC: Typically Developing Children.FSIQ: full scale IQ, estimated from Wechsler Abbreviated Scale of Intelligence (WASI).Mother's education in years.Time since baseline in months.*p < .05 with Bonferroni correction for multiple comparison, 1 = ADHD, 2 = ASD, 3 = TDC.a)Full Scale IQ (WASI).Test scores from seven participants with ADHD and two with ASD were missing at follow-up.Note: WASI follow-up: ADHD = 72, ASD = 35, TDC = 50.incorporated

Table 2 .
Results of BRIEF T-scores BRI, MI and GEC: Means and Standard Deviations by group and assessment time.

Table 3 .
Fixed effects in a Linear Mixed model with outcomes on BRIEF, with follow-up over 2 years in ADHD, ASD and TDC groups.