Abstract
Attention-deficit/hyperactivity disorder is one of the most common comorbidities in individuals with autism spectrum disorder. However, the clinical implications of the co-occurrence of these two disorders are still poorly understood. Based on a preregistered protocol (PROSPERO CRD42020193880), this systematic review identified 34 articles, published between January 1, 2014, and September 1, 2020, on cognitive, adaptive/social, and behavioral manifestations in children and adolescents with a diagnosis of autism spectrum disorder and attention-deficit/hyperactivity disorder (ASD+). The majority of available studies found a tendency toward a significant poorer cognitive performance in individuals with ASD+ compared with those with autism spectrum disorder alone (ASD−). The analysis of social/adaptive processes suggested that ASD+ is associated with lower functioning in comparison with ASD−. Finally, individuals with ASD+ were more likely to develop emotional/behavioral difficulties, in particular externalizing problems. Although some studies did not find differences among the two groups, overall co-occurring ASD+ may constitute a distinctive phenotype with a greater likelihood of cognitive, adaptive dysfunction, and mental health symptoms compared with ASD−. These results may inform the setting up and implementation of care pathways for individuals with attention-deficit/hyperactivity disorder and autism spectrum disorder.
Lay abstract
This work aimed to review recent research on the characteristics of individuals who have both autism spectrum disorder and attention-deficit/hyperactivity disorder due to their high co-occurrence. Thirty-four studies were analyzed and main findings summarized in two content domains focusing on areas that could enhance our understanding of the cognitive and behavioral characteristics of individuals with autism spectrum disorder + attention-deficit/hyperactivity disorder (ASD+). Most of the results suggested that ASD+ is a co-occurring condition associated with more severe impairments in cognitive functioning, adaptive behavior, and increased likelihood to present more emotional/behavioral problems. These results will be helpful to provide improved care plans for individuals with both attention-deficit/hyperactivity disorder and autism spectrum disorder.
Keywords
Introduction
Attention-deficit/hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) are two common neurodevelopmental disorders. A recent meta-analysis of 41 studies in 27 countries reported a worldwide prevalence rate in ADHD of 5%–7% in children and adolescents (Polanczyk et al., 2015). In the case of ASD, the median prevalence, based on an epidemiological survey conducted in 37 countries, was 0.97% in primary school children (Fombonne et al., 2021). Although ASD and ADHD can share some characteristics, they are two distinct conditions in their core symptoms and other phenotypic manifestations, with different diagnostic criteria. ADHD core symptoms include inattention, impulsivity, and hyperactivity, whereas ASD is characterized by communication and social interaction difficulties as well as the presence of repetitive and stereotyped patterns of behavior, activities, and interests (American Psychiatric Association (APA), 2013).
Empirical evidence suggests that children with ASD generally experience greater deficits in planning and flexibility (Happe et al., 2006; Salcedo-Marin et al., 2013), whereas children with ADHD exhibit more severe problems with inhibitory control and sustained attention (Corbett et al., 2009; Sinzig et al., 2008). In addition, social cognition processes are more affected in children with ASD and they usually experience withdrawal and avoidance of social interactions. In contrast, children with ADHD have less generalized impairments in social cognition, although they present difficulties in social contexts due to deficits in self-regulation that lead to frequent maladaptive behaviors (Bora & Pantelis, 2016; Taurines et al., 2012). Despite these important differences, ASD and ADHD symptoms often present together and they have a tendency to co-occur in families and share etiological mechanisms such as genetic risk factors and neurocognitive features (Yerys, 2020).
The co-occurrence of ASD and ADHD was not formally recognized until the fifth revision of the Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5; APA, 2013). The co-occurrence between both disorders is reported between 20% and 70% (Brookman-Frazee et al., 2018; Joshi et al., 2017; Llanes et al., 2020; Lyall et al., 2017; Salazar et al., 2015). This broad range depends on the type of sample (clinical or community, age), the evaluation procedures, the type of informants, the diagnostic criteria utilized, and specific characteristics of the individuals such as their cognitive level. A recent meta-analysis showed an overall pooled prevalence estimates of 28% for ADHD in ASD population (Lai et al., 2019) that increased with age and ASD severity (Gordon-Lipkin et al., 2018). In addition to being fairly frequent, a growing body of literature suggests that ASD + ADHD (ASD+) comorbidity enhances the vulnerability and clinical complexity. Indeed, in the presence of ADHD symptoms, autism seems to be generally associated with more severe impairments in executive functioning (EF), theory of mind (ToM), adaptive functioning, greater psychosocial problems, and poorer quality of life (Antshel et al., 2013; Berenguer-Forner et al., 2015; Leitner, 2014; Taurines et al., 2012).
Considering that the studies carried out before DSM-5 were not based on ASD participants having a formal ADHD diagnosis but rather ADHD symptoms, an overview of the studies published after the DSM-5 was implemented allows us to address the profile of comorbid ASD and ADHD. Thus, the present study aimed to carry out a systematic review of recent (published in or after 2014) studies exploring the characteristics of children and adolescents with ASD+ in different domains that have relevance for improving the diagnosis and treatment. The research questions of this systematic review were as follows:
Based on previous literature published before the formal diagnosis of ASD + ADHD was allowed, our first hypothesis was that children with ASD+ will experience greater executive and social cognition deficits compared with those with ASD− (Antshel et al., 2013; Berenguer-Forner et al., 2015; Taurines et al., 2012). The second hypothesis was that individuals with ASD+ present poorer adaptive/social skills and more emotional/behavioral difficulties, requiring greater treatment needs, compared with those with ASD− (Gillham et al., 2000; Leitner, 2014).
Materials and methods
The protocol of this systematic review was preregistered in PROSPERO (CRD42020193880), and the methods were developed following the Preferred Reporting Items for Systematic review and Meta-analysis (PRISMA) guidelines (Moher et al., 2015).
Inclusion and exclusion criteria
The complexity and significance of ASD+ have sufficient power to explore research questions within this subgroup regarding to ASD−. The primary aim was to identify the developmental profile of ASD+. Thus, we identified publications based on the following inclusion criteria: (1) empirical quantitative studies (cross-sectional, case-control, cohort study) written in English, with at least one group of participants with a comorbid diagnosis of ASD + ADHD (primary diagnosis of ASD) and another group with ASD−, according to the International Classification of Diseases, Ninth Revision (ICD-9), International Classification of Diseases, Tenth Revision (ICD-10), Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV), Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR), or DSM-5 criteria; (2) studies with data on ASD and ADHD symptoms, cognitive functioning, adaptive and social skills, comorbidity, or academic competence of individuals with co-occurring ASD + ADHD; (3) mean age of participants ⩽ 18 years; (4) studies published between January 1, 2014 (after the implementation of the DSM-5), and September 1, 2020, expanding the information from other recent reviews that focused only on the neuropsychological profile of individuals with ASD+ (Berenguer, Rosello, & Leader, 2018; Craig et al., 2016). Articles were excluded if they (1) did not include quantitative data and a formal peer-review process (i.e. editorials, extended abstracts, doctoral dissertations, symposium papers, research abstracts, book chapters, or proceedings, systematic reviews, and survey papers); (2) focused on molecular genetics, neuroimaging/neurophysiological data, medical conditions, and interventions (pharmacological and psychological)—these topics were beyond the scope of the present systematic review and are covered in other publications (Antshel et al., 2016; Antshel & Russo, 2019; Yerys, 2020); and (3) were case series with fewer than 10 participants.
Search strategy
PubMed, PsycINFO, and the Educational Resources Information Center (ERIC) were searched for peer-reviewed articles published in English between January 1, 2014, and September 1, 2020.
The following combination of search terms was used in the systematic searches: (“autis*” or “autistic disorder” or “pervasive developmental disorder” or “autism spectrum disorder” or “Asperger syndrome” or “high functioning autism” or “ASD”), combined with (“attention deficit*” or “ADHD” or “attention-deficit/hyperactivity disorder” or “hyperkinetic disorder” or “hyperactiv*” or “inattentive” or “impulsiv*”) and (“child*” or “adolescen*” or “teen*”). Variations in the syntax of the search terms and Medical Subject Headings (or equivalents) were used according to the specific database.
Two authors independently searched and reviewed all the references via the platform Covidence. Potential disagreement between authors during the selection process was resolved in regular meetings that were held to carefully discuss titles/abstracts in the first phase and full texts in the second phase to ensure agreement.
Study quality assessment
The quality of the reporting of included studies was assessed with a standardized and validated set of criteria based on the “STROBE Reporting Guidelines for writing and reading observational studies in epidemiology” (von Elm et al., 2014). The evaluation included 19 items related to the introduction (three items; justification of the study, well-described aims, and clear hypothesis/expectations), methods (nine items; for example, replicable protocol, number of participants, inclusion–exclusion criteria, detailed description of methods of measurement), results (two items; appropriate analyses plan and clear results presentation), and discussion sections (four items; for example, results discussed according the literature, limitations shown), and other information (one item about ethics). For each criterion, two of the authors assigned scores of 2, 1, or 0, depending on whether it was completely fulfilled, partially fulfilled, or not fulfilled at all. A second quality tool was added as an additional tool to assess the methodology of the studies. The checklist from the Centre for Evidence Based Management (CEBMA, 2014) includes 12 items (e.g. accuracy of the research question, methods sufficient to address it, recruitment description and potential bias, use of power calculation, validity of measures, statistical analyses, identification of confounding variables, and generalizability of results) and each item can be scored 0 or 1, with total higher scores indicating better quality. Two authors rated the studies and resolved disagreements through discussion until a consensus was achieved. Tables in Online Appendix A and B report the appraisal of the reporting of the studies and their methodological quality, respectively.
Results
Overall, 3604 potentially relevant references were initially identified (1851 in PubMed, 1213 in PsycINFO, and 540 in ERIC). Of these, 1455 were excluded due to duplication. Of the remaining 2149 references, titles/abstracts were screened by two authors according to the inclusion and exclusion criteria and 2073 were discarded as deemed not pertinent (i.e. studies on interventions, etiology, comorbidity with medical conditions, adult population, prevalence). Then, 76 full texts were screened by two authors and, after consensus was reached, a total of 32 articles were selected as pertinent. Finally, with the addition of two other studies located through references from other publications, a total of 34 studies were included in this systematic review (Figure 1).
PRISMA flowchart of the search strategy.
In the 34 studies selected, two main content domains and several subdomains were identified: (1) cognitive phenotype, including the subdomains of EF, sensory processing, and social cognition; and (2) behavioral phenotype, including adaptive/social skills and emotional/behavioral difficulties. Of the 34 included studies, 15 focused on cognitive phenotype (Biscaldi et al., 2016; Dajani et al., 2016; Dellapiazza et al., 2021; Gargaro et al., 2018; Kado et al., 2020; Lundervold et al., 2016; Oerlemans et al., 2014; Salunkhe et al., 2021; Sanz-Cervera et al., 2017; Seernani et al., 2021; Tye et al., 2016; Unterrainer et al., 2016; Van der Meer et al., 2016; Waddington et al., 2018a, 2018b), 15 on behavioral phenotype (Ashwood et al., 2015; Avni et al., 2018; Carta et al., 2020; Chen et al., 2015; Chiang et al., 2018; Magnúsdóttir et al., 2016; McClain et al., 2017; McFayden et al., 2020; Ng et al., 2021; Rao & Landa, 2014; Salley et al., 2015; Scandurra et al., 2019; So et al., 2017; Turygin et al., 2015; Yamawaki et al., 2020), and four on cognitive and behavioral phenotypes (Berenguer et al., 2018b; Colombi & Ghaziuddin, 2017; Craig et al., 2015; Ng et al., 2019).
Tables 1 and 2 report the participants characteristics, measures used, and main findings of each study included in the systematic review in relation to cognitive processes (EF and social cognition) and behavioral functioning (adaptive/social skills and emotional/behavioral difficulties), respectively. Tables in Online Appendix A and B report the appraisal of the reporting of the studies and their methodological quality, respectively. The main findings across studies are summarized in the following subsections, and in the cases where the information was available, exploratory forest plots figure are presented.
Sample, measures, and findings of studies analyzing cognitive processes in children/adolescents with ASD+ADHD.
AP= Affective Prosody Task; ASD+ = ASD+ADHD; BRIEF = Behavior Rating Inventory of Executive Function; CBCL= Child Behavior Check List; CCT= Children’s Category Test; CFT20-R= Culture Fair Intelligence Test 20; CPT= Continuous Performance Test; CRS-R= Conners’ Rating Scale Revised; CV= coefficient of variability reaction time; D-KEFS= Delis Kaplan Executive Function; GARS= Gilliam Autism Rating Scale; IFE = Identification of Facial Emotion; ISV= Intra-subject variability; PACS = Parent Account of Childhood Symptoms; R. Mind in the Eyes = Reading the Mind in the Eyes; RT= Reaction time; SCQ = Social Communication Questionnaire; SRS= Social Responsiveness Scale; TD= typical development; TMT= Trail Making Test; TOL= Tower of London; WMI= Working Memory Index.
Sample, measures, and main findings of studies analyzing behavioral functioning of children/adolescents with ASD+ADHD.
ADDES= Attention Deficit Disorder Evaluation Scale; ASD+ = ASD+ADHD; BASC-2= Behavioral Assessment System for Children, 2nd Edition; BSID-III= Bayley Scales of Infant and Toddler Development Third Edition; CBCL= Child Behavior Check List; CPT-2 = Continuous Performance Test II; CRS-R= Conners’ Rating Scales Revised; D-EFS= Delis Kaplan Executive Function; Dx= diagnosis; ID= Intellectual Disability; K-SADS-PL= Schedule for Affective Disorders and Schizophrenia for School-Age Children–Present and Lifetime Version; m (months) M-CHAT= Modified Checklist for Autism in Toddler; SDQ= Strengths and Difficulties Questionnaire; SCT= Sluggish Cognitive Tempo; SNAP-IV= Swanson, Nolan and Pelham, version IV; SRS= Social Responsiveness Scale; PACS = Parent Account of Childhood Symptoms; VABS= Vineland Adaptive Behavior Scales;
Cognitive phenotype of co-occurring ASD + ADHD
Nineteen studies focused on the analysis of cognitive, sensory, and social cognition processing: EF (Berenguer, Roselló, Colomer, et al., 2018; Colombi & Ghaziuddin, 2017; Dajani et al., 2016; Gargaro et al., 2018; Kado et al., 2020; Ng et al., 2019; Salunkhe et al., 2021; Unterrainer et al., 2016; Van der Meer et al., 2016), variability in reaction time (RT; Biscaldi et al., 2016; Lundervold et al., 2016; Salunkhe et al., 2021; Seernani et al., 2021; Tye et al., 2016), ToM and emotion recognition (ER; Colombi & Ghaziuddin, 2017; Oerlemans et al., 2014; Waddington et al., 2018a, 2018b), and sensorial processing (SP; Dellapiazza et al., 2021; Sanz-Cervera et al., 2017). The majority had a cross-sectional design, including three clinical groups (ASD−, ASD+, and ADHD-only (ADHD−)). In total, the median number of participants was 160, aged between 5 and 15, and 79% were males. Participants’ cognitive levels were reported in all studies, with mean global intelligence quotient (IQ) scores ranging from 78 to 120. Fourteen studies were conducted in Europe, three in North America, one in Japan, and another one in Australia.
EF
The retrieved studies explored the possible cognitive phenotypes to explain the main ADHD symptoms (Barkley, 2005) and the restrictive interests and repetitive behaviors present in ASD (Lopez et al., 2005): working memory (WM), attention, response inhibition, planning, and cognitive flexibility.
Enquiry of WM in children with ASD+ reported mixed findings. Performance on the WISC-IV (Wechsler Intelligence Scale for Children–Fourth Edition) working memory index (WMI) was found significantly poorer in children with ASD+ compared with children with ASD− (Colombi & Ghaziuddin, 2017), while other studies (Ng et al., 2019; Salunkhe et al., 2021) did not find any significant WMI differences between both groups. Several factors may have influenced the results, including possible medications effects (Ng et al., 2019) or computerized tasks (Salunkhe et al., 2021).
In terms of attentional processes, the ASD+ group on the continuous performance test (CPT) was found to be less accurate with more omission errors and less consistent on visual search than the pure ASD− group (Lundervold et al., 2016). Likewise, parent ratings highlighted more severe attentional problems in children with ASD+ and ADHD compared with those with ASD− (p = 0.06, Ng et al., 2019). Another finding (Gargaro et al., 2018) suggested a particular attention-processing profile of the ASD+ and ADHD− groups that was characterized by significantly slower RT in attention switching, whereas children with ASD− did not exhibit these types of impairments (p < 0.01).
Data on organizing and planning showed that ASD− and ASD+ had a similar performance (Colombi & Ghaziuddin, 2017). However, the developmental profile between the two conditions may be slightly different (Unterrainer et al., 2016). In global performance on the tower of London task (TOL), no significant differences were found between ASD− and ASD+ groups, although children with ASD+ were slightly more impaired and less accurate at younger ages on the computerized TOL, while at older ages their results were similar to typical development (TD). Thus, the expected greater difficulty for the addition of both diagnoses in ASD+ was not present, at least in older children, suggesting delayed development.
Cognitive flexibility was analyzed with a special focus on the scores of the second step of Wisconsin card sorting test (KWCST; Kado et al., 2020), in which all the cards are sorted again to assess the affectivity of experience in the first step and the instructions given to the examinee. Older ASD+ and ASD− individuals showed significant improvements in the second step of the KWCST test compared with the younger clinical subjects. The older ASD+ children obtained significantly more categories of achievement (d = 0.68, confidence interval (CI) = [−1.06, −0.31]) and less perseverative errors (d = 0.58, CI = [−0.52, 1.72]) than the ASD− children in the first step. On the contrary, the younger ASD+ showed on the second step a worse performance than ASD− regarding the number of responses (d = 0.78, CI = [−2.98, 1.30]). These findings suggested that younger individuals with ASD+ struggle to sustain attention and older individuals have flexibility problems and impulsivity symptoms of ADHD, likely to protect them from the behavioral rigidity.
The intrasubject variability (ISV) was analyzed in several studies. In a visual search task, the ASD− group exhibited significantly better search performance (slower mean fixation duration, d = 0.34, CI = [−12.69, 11.99]), whereas the ASD+ group showed increased ISV (more entry time to cue, d = 0.90, CI = [−0.97, −0.26]). Similarly (Biscaldi et al., 2016), on a broad test battery, RT was slower and ISV more increased in the ASD+ than in the ASD− group, although without reaching significant level. More important, the analysis of the profile between groups supported that the increase in ISV was essentially motivated by ADHD comorbidity.
In a subsequent study that included a group of children with ADHD− (Salunkhe et al., 2021), individuals with ASD+ or ADHD−, but not ASD−, presented high ISV levels across different tasks and temporal processing deficits (comparison of ISV between the two groups with ASD, d = 1.02). Even more, ASD+ in comparison with ASD− individuals showed worse variability indexes in slow tasks (mean RT, d = 1.02, CI = [−60.06, 58.01]) and fast incentive tasks (mean RT, d = 0.39, CI = [−49.37, 48.58]; Tye et al., 2016). While RT is a potential marker of ASD+ co-occurrence in slow and less rewarding conditions, it had no discrimination power in faster and more rewarding conditions.
Finally, two studies reported data from ecological scales of EF in daily life. The first of them (Dajani et al., 2016) conducted a latent profile analysis with a combination of EF indicators. In the ASD+ group, 92% of children were classified as having impaired EF, compared with 47% of children with ASD− and 63% of children with ADHD−. The second study (Berenguer et al., 2018b) showed that the groups with ASD+ and ADHD− experienced significantly more deficits in WM, planning, and monitoring than the group with ASD−. In addition, the ASD+ group showed significantly worse levels on inhibition and shift than the ASD− group (see Figure 2).
Forest plot summarizing the results of the studies on executive functions.
SP
Research comparing children with ASD+ with individuals with a single diagnosis showed that the co-occurring condition presented an overall atypical SP in auditory (d = 1.09, CI = [−0.29, 2.53]) and multisensorial processing (d = 0.70, CI = [−0.27, 1.63]) (Dellapiazza et al., 2021). The parents’ and teachers’ ratings comparison of SP, social participation, and praxis revealed interesting differences (Sanz-Cervera et al., 2017). In the home context, the ASD+ group obtained a higher level of body awareness dysfunction than the ASD− group (d = 0.9, CI = [−3.03, 0.37]), alongside greater social participation dysfunction (SOC) than the ADHD− group (d = 1.4, CI = [−3.62, 0.50]). In the classroom context, the ASD+ group obtained worse scores than the ADHD− group on SOC (d = 2.3, CI = [−4.18, −0.55]) and hearing (d = 1.52, CI = [−3.66, 0.62]).
Social cognition
Greater impairments in affective prosody and several elements of facial ER such as happiness recognition were found in the ASD+ group compared with children with ASD− (Oerlemans et al., 2014). Similarly, children with ASD+ showed less development of empathy, measured by “reading the mind in the eyes,” than children with ASD−, with differences between groups close to significance (Colombi & Ghaziuddin, 2017). A further investigation (Waddington et al., 2018b) neither found significant differences in ER between the ASD+, ASD−, and ADHD− groups, despite the ASD+ group presented the greatest discrepancy with the control group on speed of visual ER (p = 0.007; d = 0.39), speed of auditory ER (p < 0.001; d = 0.77), and accuracy of auditory ER (p = 0.05; d = 0.75). In subgroups of children identified according to visual and auditory ER skills, poor performance was linked to an increased likelihood for ASD and ADHD (Waddington et al., 2018a). Children and adolescents with ASD+, ASD−, and ADHD− were included in all classes, but the lowest performing class had the highest percentage of individuals scoring on ASD/ADHD measures (66.07%).
On the contrary, better and average-performing classes displayed fewer ADHD and ASD symptoms.
The application of ToM skills in everyday social contexts in different domains, such as early skills, social references and understanding basic emotions, meta-representations, and second-order inferences, has been assessed using parent ratings (Berenguer et al., 2018b). The comparison analyses showed a similar deficit in the application of ToM skills in both groups with ASD, whose results were worse than those obtained for the ADHD− group(see Figure 3).
Forest plot summarizing the results of the studies on social cognition.
Behavioral phenotype of co-occurring ASD + ADHD
Nineteen studies dealt with the behavioral functioning of children and adolescents with ASD+: adaptive/social characteristics (Ashwood et al., 2015; Avni et al., 2018; Chiang et al., 2018; Craig et al., 2015; Magnúsdóttir et al., 2016; McFayden et al., 2020; Ng et al., 2019, 2021; Rao & Landa, 2014; Salley et al., 2015; Scandurra et al., 2019; Turygin et al., 2015) and emotional/behavioral difficulties (Berenguer et al., 2018b; Carta et al., 2020; Chen et al., 2015; Craig et al., 2015; McClain et al., 2017; So et al., 2017; Yamawaki et al., 2020). Most of the studies had a cross-sectional design; 14 of them included three clinical groups and four studies had ASD− and ASD+ groups. In total, the median number of participants was 169, aged between 2 and 18 years and 81.5% of males in the samples. Participants’ cognitive levels were reported in 12 studies with mean IQ scores ranging from 72 to 111. Six studies were conducted in Europe, six in North America, one in Israel, two in Japan, and another two in Taiwan. Please see Table 2.
Adaptive/social characteristics
Although there is no universal definition of the concept of “adaptive,” the most widely held view refers to an individual’s ability to perform skills independently to meet personal needs and environmental demands (Price et al., 2018). One empirical study (Turygin et al., 2015) reported that in early childhood, the adaptive skills of young children with ASD− and ASD+ differed from those of the ADHD− in several domains: adaptive, motor, communication, and personal/social. However, there were no significant differences between the ASD− and ASD+ groups. In mid-childhood (Ashwood et al., 2015; Craig et al., 2015; Scandurra et al., 2019), the same tendency toward a worse adaptive general profile in the ASD− and ASD+ groups than in the ADHD− group was observed, but without statistical significance between both ASD groups.
Nevertheless, other data (Rao & Landa, 2014) revealed that ASD+ mean scores were significantly impaired on the socialization and daily living domains in comparison with ASD−. The ASD+ group also had greater impairments in social awareness (d = 0.76, CI = [−3.66, 2.19]), social communication (d = 0.77, CI = [−3.92, 2.50]), and social motivation (d = 0.70, CI = [−3.83, 2.41]). The co-occurrence ASD and ADHD also showed an impact on all domains of school adaptation, being associated with worse attitudes toward school work (d = 0.79, CI = [−0.93, −0.65]), more school interaction problems (d = 0.57, CI = [−0.67, −0.48]), and more school behavioral problems (d = 0.76, CI = [−0.85, −0.70]; Chiang et al., 2018).
Furthermore, ADHD symptoms showed a different association with social impairments in children with ASD+ versus children with ADHD: Ratings of hyperactivity/impulsivity were the main explanatory factor of the Social Responsiveness Scale in the ADHD group, whereas inattentiveness was the strongest explanatory factor in the comorbid group (Ng et al., 2021). ASD+, but not ASD− or ADHD−, could predict global social impairment. This effect may be increased by the presence of sluggish cognitive tempo, measured with the sluggish cognitive tempo subscale (SCT) derived from four items on the child behavior checklist (CBCL) (McFayden et al., 2020). However, some results suggested the possibility that the negative effect of ADHD on ASD children in the communication domain and general score of adaptive functioning is limited to high functioning boys (Magnúsdóttir et al., 2016).
The influence of the severity of ADHD symptoms on adaptive skills has been explored across several clinical conditions (Avni et al., 2018). In the comparison among ASD−, ASD+, ASD + anxiety, and ASD + ADHD + anxiety, the ASD− group experienced higher adaptive functioning than the other groups. Specifically, ASD+ participants showed significant impairments in socialization compared with ASD−. Moreover, the presence of two co-occurring conditions, (ASD + ADHD + anxiety) was associated with more deficits, affecting the socialization and daily living domains (d = 0.89, CI = [−0.86, 2.65] and d = 0.62, CI = [−1.19, 2.44], respectively) (see Figure 4).
Forest plot summarizing the results of the studies on adaptive/social skills.
Emotional/behavioral difficulties
The assessment of mental health difficulties in young people with ASD−, ASD+, and ADHD− in a large sample reported that participants with ASD+ had a greater chance of experiencing bipolar disorder (3.4% vs. 1.2% vs. 4.0% vs. 0.1%, p < 0.001), depressive disorder (6.4% vs. 5.6% vs. 7.7% vs. 0.6%, p < 0.001), anxiety disorder (17.4% vs. 18.3% vs. 7.3% vs. 0.5%, p < 0.001), disruptive behavior disorder (8.2% vs. 6.7% vs. 2.1% vs. 0.1%, p < 0.001), and tic disorder (7.1% vs. 5.5% vs. 3.6% vs. 0.8%, p < 0.001) compared with ASD−, ADHD−, and TD (Chen et al., 2015). However, no significant differences were found regarding the prevalence and severity of internet addiction in adolescents with ASD+, ASD−, or ADHD− (So et al., 2017).
The ASD+ problems profile has been explored using the strengths and difficulties questionnaire (SDQ). The total problem score and hyperactivity symptoms were significantly higher in the ASD+ group than in ASD−, whereas peer relations were more affected in both groups with ASD than in the ADHD− group (d = 1.62, CI = [2.11, −1.13]; Berenguer et al., 2018b). The greater severity of problems in the comorbid condition was supported by studies analyzing an extensive profile of behaviors with the CBCL. Thus, externalization and other subscales such as conduct (d = 0.99, CI = [−2.37, 0.38]), aggressive (d = 0.89, CI = [−2.42, 0.64]), and thought problems (d = 0.94; CI = [−2.95, 1.07]) were more affected in the ASD+ than the ASD− group (Carta et al., 2020; Craig et al., 2015). Moreover, using the Conners rating scale, more impairments in oppositional defiant disorder (ODD) (d = 1.15, CI = [−4.06, 1.76]), cognitive problems (d = 1.05, CI = [−4.18, 2,08]), anxiety (d = 1.96, CI = [−3.52, 1.60]), perfectionism (d = 1.28, CI = [−3.93, 1.38]), social problems (d = 1.01, CI = [−4.74, 2.73]), and emotional lability (d = 1.18, CI = [−4.15, 1.79]) were reported for the ASD+ group than the ASD− (Craig et al., 2015).
The majority of studies highlighted more severe attention and hyperactivity/impulsivity problems in individuals with ASD+ compared with those with a single diagnosis. However, a recent study (McClain et al., 2017) reported unexpected results, suggesting that ASD+ exhibited similar levels of ADHD symptoms to children with ASD− or ADHD−, which is at odds with previous findings, as the authors themselves point out (see Figure 5).
Forest plot summarizing the results of the studies on emotional/behavioral difficulties.
Discussion
The DSM-5 provided official support for the diagnosis of co-occurring ASD + ADHD, prompting a progressive increase in research on the characteristics of individuals with the double diagnosis. The present systematic review summarizes the main findings from recent empirical literature. It included 34 studies focusing on areas that could enhance our understanding of the cognitive and behavioral characteristics of individuals with ASD+. However, the results can be considered as preliminary, considering the limited number of investigations carried out.
In relation to our first hypothesis, our findings extend those of previous nonsystematic reviews (Antshel et al., 2016; Antshel & Russo, 2019; Taurines et al., 2012; Yerys, 2020) showing that individuals with ASD+ display severe impairments in executive and social processing. EF and RT indexes continue to be the main focus of studies addressing cognitive domains, followed by studies about ToM and ER. However, research on SP is limited (two studies) despite that the hypo/hypersensitivity to visual, tactile, and auditory input is included within DSM-5 diagnostic criteria for ASD (APA, 2013). The presence of ADHD had a negative impact on inhibitory and attentional control as well as on increased ISV in RTs, providing evidence that the severity of the neuropsychological dysfunction is greater in the ASD+ and ADHD− groups than in the ASD− and TD groups. On the contrary, some inconsistencies in the results for WM processes do not allow us to firmly establish more pronounced impairments in ASD+, compared with one single diagnosis. The ability to mentally manipulate information for a time seems to be more impaired in children with ASD+ compared with children with ASD−. However, in two studies, no significant differences were found between the two groups of children with ASD, possibly due to the influence of factors related to the computerized presentation of the tasks or medication. Furthermore, studies on executive components of planning and flexibility with a developmental perspective pointed out that the results of older ASD+ children tended to match those of the control group. This finding suggests that over time, impulsivity that is characteristic of ADHD may counteract the inflexibility observed in ASD. In other words, the presence of ADHD symptoms in ASD could favor cognitive flexibility.
Regarding sensory processing, the data reviewed suggest an overall atypical sensory processing and greater auditory impairments in individuals with ASD+, compared with the ASD− group. In fact, more than 60% of children with ADHD− exhibit atypical sensory processing (Mimouni-Bloch et al., 2018), particularly on scales of the sensory processing measure, with small to medium effect sizes (Pfeiffer et al., 2015). These deficits may contribute to the inaccurate behavioral and learning responses (Shimizu et al., 2014).
The results also suggested the greater overall deficit in social cognition in the co-occurring condition. Direct comparisons of the groups with ASD+, ASD−, and ADHD− on ER and ToM tasks showed that the former group was associated with more pronounced impairments in reading other people’s emotions, feelings, and affective prosody, as well as less development of empathy. The problems with RT speed, inhibition, and attention that are often present in ADHD may partly explain the deficits in social processing tasks (Oerlemans et al., 2014). EF abilities guide social competence (Pellicano, 2013) and ToM (Pellicano, 2010) in ASD−, so a similar relationship could occur in the case of ASD+.
Regarding our second hypothesis, we expected that individuals with ASD+ would present poorer adaptive/social functioning and more emotional/behavioral difficulties as reported in reviews prior to the publication of the DSM-5 (Gillham et al., 2000; Leitner, 2014). Overall, the ASD+ condition seemed to exhibit many adaptive and social problems in the communication, daily living skills, and socialization domains that are often present since early childhood. Although the results of the studies on this area are not totally consistent, the general trend showed more severe impairments in individuals with ASD+ in comparison with those with a single diagnosis of ASD or ADHD. Furthermore, co-occurring ASD+, but not an ASD diagnosis alone, predicted greater adaptive and social impairments, with inattention being the most important predictor of social responsiveness.
The ASD+ group tended to present more externalizing problems than the group with ASD− and often similar levels of emotional/behavioral difficulties to those individuals with ADHD−. The findings are in line with a cohort study of ASD− participants (Mansour et al., 2017) where ASD severity did not contribute to emotional and behavioral impairments and, by contrast, greater ADHD severity was significantly related to anxiety, depression, somatic complaints, and social problems. The majority of the seven studies that addressed the co-occurrence with other mental health difficulties relied the assessment on parent-/teacher-reported questionnaires such as SDQ or CBCL while rarely used clinical standardized interviews. Another study that used ICD-9 (Chen et al., 2015), identified ASD+ increased the likelihood for developing psychiatric conditions such as bipolar disorder or depressive disorder compared with ASD− or ADHD−. Despite this, the reviewed literature does not inform potential factors that predict the presence, continuity, and change of frequent mental health disorders in this co-occurring condition.
Summarizing, our data underline that individuals with ASD+ deserve specific attention. The cognitive and behavioral needs that here have been identified could inform possibilities of treatment. Specific psychosocial interventions for the co-occurring condition have not been implemented and further investigation is needed to improve the intervention. In a comprehensive review, Davis and Kollins (2012) mention that there are similarities across approaches to treat both disorders. Initially, the co-occurring presentation could benefit from combinations of strategies that have proven effective for ASD and for ADHD, such as cognitive-behavioral therapies, social skills training, or executive function training. However, it should be taken into account that these techniques may need to be modified in the context of ASD because they may have less insight and self-awareness and lower communication abilities. In addition, although the tolerability and efficacy/effectiveness of ADHD pharmacotherapy in individuals with ASD+ are lower compared with those with ADHD−, there is support for its use in the treatment of inattentive and hyperactive/impulsive symptoms in individuals with ASD+ (Cortese et al., 2012; Rodrigues et al., 2020).
Limitations and future perspectives
There are several limitations of the current systematic review that must be considered. First, studies on molecular genetics and neuroimaging/neurophysiological data were not included in this study. Although the co-occurrence of ASD and ADHD seems to be supported, at least partly, by shared genetic contribution (Ghirardi et al., 2018) and neurocognitive networks (Bethlehem et al., 2017), addressing these topics was beyond the scope of this work. Second, we excluded articles that were not written in English Third, the exclusion of gray literature might also be considered a limitation. However, despite its inclusion provides a broader view of the available evidence, there are concerns regarding the quality of the data reported in such literature (Mahood et al., 2014). Reference must also be made to limitations found in the research design of the reviewed studies. In spite of the time elapsed since the introduction of the DSM-5, the majority of studies used various diagnostic classification systems. In contrast, only a small percentage of studies applied the most updated criteria from the DSM-5 to diagnose ASD and ADHD. Sample sizes were generally small and homogeneous, including mainly male participants. Nevertheless, the gender proportion was representative of the prevalence of males of the condition. The studies were not age-matched that limits the generalizability of our findings and they frequently included participants with a level of intellectual functioning ranging from low average to above average, so the conclusions cannot be extrapolated to children or adolescents with ASD+ and intellectual disability.
Based on the quality assessment carried out, in general terms most studies included in the present systematic review meet to a large extent scientific quality indicators; specifically 24 articles meet between 89%–95% and 75%–92% of the established criteria of the STROBE and CEBMA. The methodology such as the study setting, the recruitment, or the sample size was the section that generally obtained the lowest scores in comparison with the introduction, the results, or the discussion. Of the seven articles assessed with the lower scores in methodology, three focused on adaptive domain, two on mental health difficulties, and two on EF. Future research could use larger samples, include groups of girls and individuals with intellectual disability, and differentiate between ADHD presentations (combined, inattentive, and impulsive/hyperactive) that have not received enough attention. Another limitation has to do with the developmental stage addressed in the studies. The focus was on mid-childhood and preadolescence. However, identifying and providing information about the specific comorbid group, with an emphasis on early childhood and the preschool period, will help to develop early intervention strategies.
In addition, some studies suggest differentiated developmental trajectories for the ASD− versus the ASD+ condition, although these findings rely on cross-sectional designs. There is a lack of studies that analyze the course of ASD+ over time. In the future, prospective designs will make it possible to examine how clinical diagnoses of ASD− and ASD+ differ in cognitive and behavior domains throughout the life span and to identify mediators involved in different trajectories. The study of educational needs is another challenge for the research in next years. Despite its importance in sociopersonal adjustment, the topic has not been considered until now.
Finally, the results of genetic, neuropsychological, and neuroimaging studies suggest possible pathophysiological links between ASD and ADHD affecting key frontostriatal and frontoparietal circuits that are important for performing EF and complex cognitive functions as attentional process (Rommelse et al., 2011). It is possible that cognitive and behavioral phenotypes of individuals with ASD+ are linked to brain abnormalities of ASD and ADHD (Christakou et al., 2013) and the pathophysiology of ASD+ also is related to somatosensory deficits and delayed maturation of the left postcentral gyrus (Mizuno et al., 2019). In this sense, there is an evident need in the upcoming years for a better understanding of the etiology of this co-occurring condition and determine the shared and distinctive genetic and neurobiological basis as well as the complex interactions with environmental factors such as poverty, sociocultural deprivation, and poor parenting.
Supplemental Material
sj-docx-1-aut-10.1177_13623613211065545 – Supplemental material for Cognitive, social, and behavioral manifestations of the co-occurrence of autism spectrum disorder and attention-deficit/hyperactivity disorder: A systematic review
Supplemental material, sj-docx-1-aut-10.1177_13623613211065545 for Cognitive, social, and behavioral manifestations of the co-occurrence of autism spectrum disorder and attention-deficit/hyperactivity disorder: A systematic review by Rocio Rosello, Jose Martinez-Raga, Alvaro Mira, Juan Carlos Pastor, Marco Solmi and Samuele Cortese in Autism
Supplemental Material
sj-docx-2-aut-10.1177_13623613211065545 – Supplemental material for Cognitive, social, and behavioral manifestations of the co-occurrence of autism spectrum disorder and attention-deficit/hyperactivity disorder: A systematic review
Supplemental material, sj-docx-2-aut-10.1177_13623613211065545 for Cognitive, social, and behavioral manifestations of the co-occurrence of autism spectrum disorder and attention-deficit/hyperactivity disorder: A systematic review by Rocio Rosello, Jose Martinez-Raga, Alvaro Mira, Juan Carlos Pastor, Marco Solmi and Samuele Cortese in Autism
Supplemental Material
sj-docx-3-aut-10.1177_13623613211065545 – Supplemental material for Cognitive, social, and behavioral manifestations of the co-occurrence of autism spectrum disorder and attention-deficit/hyperactivity disorder: A systematic review
Supplemental material, sj-docx-3-aut-10.1177_13623613211065545 for Cognitive, social, and behavioral manifestations of the co-occurrence of autism spectrum disorder and attention-deficit/hyperactivity disorder: A systematic review by Rocio Rosello, Jose Martinez-Raga, Alvaro Mira, Juan Carlos Pastor, Marco Solmi and Samuele Cortese in Autism
Supplemental Material
sj-docx-4-aut-10.1177_13623613211065545 – Supplemental material for Cognitive, social, and behavioral manifestations of the co-occurrence of autism spectrum disorder and attention-deficit/hyperactivity disorder: A systematic review
Supplemental material, sj-docx-4-aut-10.1177_13623613211065545 for Cognitive, social, and behavioral manifestations of the co-occurrence of autism spectrum disorder and attention-deficit/hyperactivity disorder: A systematic review by Rocio Rosello, Jose Martinez-Raga, Alvaro Mira, Juan Carlos Pastor, Marco Solmi and Samuele Cortese in Autism
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: RR is recipient of an Alicia Koplowitz Foundation Fellowship.
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
