Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Jul 1.
Published in final edited form as: Clin Neuropsychol. 2021 Jul 27;36(5):1069–1093. doi: 10.1080/13854046.2021.1942554

Co-occurring attention-deficit/hyperactivity disorder and anxiety disorders differentially affect males and females with autism

Ericka L Wodka a,b, Julia Parish-Morris c,d, Robert D Annett e, Laura Carpenter f, Emily Dillon g, Jacob Michaelson h, So Hyun Kim i, Rebecca Landa a,b; SPARK Consortium#, Stephen Kanne i
PMCID: PMC9181886  NIHMSID: NIHMS1789420  PMID: 34315336

Abstract

Objective:

To examine overlap and divergence of symptomatology in Autism Spectrum Disorder (ASD) with and without co-occurring Attention/Deficit Hyperactivity Disorder (ADHD) and/ or Anxiety Disorder by age and sex.

Method:

Participants included 25,078 individuals registered in the SPARK cohort, age 6–18 years. SPARK participation includes online consent and registration, as well as parent-reported ASD, ADHD, and Anxiety Disorder diagnoses, developmental, medical, and intervention history, and standardized rating scales. Individuals with ASD, ASD + ADHD, ASD + Anxiety, or ASD + ADHD + Anxiety were compared on measures assessing social communication, restricted and repetitive behaviors (RRBs), and motor functioning, and differences between male and female profiles were examined.

Results:

Significant differences in symptom presentation between females/males, school-age/adolescent individuals, and by co-occurring conditions (ASD/ADHD/Anxiety) are apparent, and the impact of co-occurring conditions differed by age and sex. Most notably, school-age femaleswith ASD without co-occurring conditions present with significantly fewer concerns about social communication skills and have better motor skills, but have more prominent RRBs as compared to same-aged males with ASD alone; co-occurring conditions were associated with increased social communication problems and motor concerns, most consistently for school-age females.

Conclusions:

School-age females with ASD are at highest risk for underestimation of autism-related symptoms, including underestimation of symptoms beyond core ASD features (motor skills). Further, across ages, particular consideration should be given when probing for social communication symptoms, RRBs, and motor skills in females with ASD alone, as well as with co-occurring ADHD and/or Anxiety. For females with co-occurring symptoms and conditions, use of symptom-specific measures in lieu of omnibus measures should be considered.

Keywords: Autism, ADHD, anxiety, sex/gender, female, repetitive, motor, age

Introduction

Autism Spectrum Disorder (ASD) is a neurodevelopmental condition characterized by limited social communication skills as well as restricted and repetitive patterns of behavior. Symptoms first present in early toddlerhood and impact boys more frequently than girls, with an estimated male:female ratio of 3:1 (Loomes et al., 2017) to 4:1 (Feliciano et al., 2018; Baio, 2012). Several theories have been proposed to explain the differential incidence of autism in males, one of which is that females present with a different symptom profile than males, and are thus under-diagnosed or misdiagnosed (Baio, 2012). Autism was originally described in males, and the current conceptualization of autism is heavily influenced by male-stereotypical behaviors and interests (e.g., transportation, technology). However, recent research suggests that females with autism may present with more female-typical behaviors and interests (e.g., people, animals), which may contribute to under-detection and misdiagnosis compared to males (Lai & Szatmari, 2020). This is problematic for many reasons; one being that if females with ASD do not receive a diagnosis, they miss opportunities to receive appropriate intervention services to improve adaptive functioning and quality of life. Additionally, when females with ASD are misdiagnosed or not diagnosed at all, they cannot be represented in scientific research on autism, which has traditionally been dominated by male research participants. Reduced research participation by this subsample further limits our understanding of ASD symptoms in females, with serious downstream effects on clinical knowledge. Gaps in clinical knowledge about females with ASD hinder timely and accurate diagnosis and access to effective treatments for this critically understudied and underserved subgroup.

Existing research suggests that females with ASD present a unique behavioral phenotype compared to males, and this pattern changes across development. Females diagnosed at school age are more likely to have stronger cognitive development and fewer repetitive and restricted behaviors compared to males, while females diagnosed in adolescence may present with more significant limitations in adaptive and social functioning (Hull et al., 2020). This divergence from the male presentation could lead to under-recognition in both clinical practice and research. This is particularly important to consider when designing studies that aim to reduce the impact of heterogeneity on ASD research by defining specific subtypes within the autism spectrum, such as those in particular age ranges, of certain cognitive levels, or who demonstrate specific behavioral profiles. Narrowing and defining the profile of participants in individual research studies limits the potential subject pool, particularly for the already smaller population of females with ASD, impeding research aimed at understanding subsamples of females with autism. This appears to have impacted extant research on females with ASD and co-occurring psychiatric conditions, wherein males are overrepresented and less is known about females with ASD and co-occurring conditions.

Co-occurring conditions

Up to 70% of children with ASD have at least one co-occurring psychiatric condition (American Psychiatric Association, 2013; Matson & Nebel-Schwalm, 2007; Simonoff et al., 2008) and 40% have two or more co-occurring conditions (Simonoff et al., 2008). The impact of these co-occurring conditions on functioning is unclear; however, differences in presentation between individuals with ASD that have other co-occurring medical, behavioral, neurodevelopmental, and psychiatric conditions has been described. For instance, co-occurring ASD + Anxiety Disorder (ASD + Anx) is associated with sleep problems (Mazurek & Petroski, 2015) and self-injurious behaviors (Kerns et al., 2015), both of which are commonly reported in individuals with each diagnosis alone (ASD without co-occurring Anxiety and Anxiety without co-occurring ASD) (Chartrand et al., 2012; McMakin & Alfano, 2015; Meir et al., 2019). Individuals with ASD plus another commonly co-occurring condition (Attention Deficit/Hyperactivity Disorder: ADHD) have increased ASD symptoms and heightened risk of anxiety and mood disorders (White et al., 2009) with potentially earlier onset of anxiety symptoms (White et al., 2009). Those with co-occurring ASD + ADHD also present with atypical executive functioning and motor movement patterns (Berenguer et al., 2018). This is not surprising, given the well-described motor difficulties and executive dysfunction present in each disorder alone (Craig et al., 2016; Mahone & Denckla, 2017; Pitzianti et al., 2016; Wodka et al., 2009). While a handful of existing studies have attempted to parse the impact of common co-occurring conditions on ASD (in predominantly male samples), none have addressed whether co-occurring conditions impact behavioral presentation in females with ASD differently than in males. This gap in the literature increases the chances that females with ASD will be misdiagnosed (due to misattribution of symptoms to co-occurring conditions), potentially delaying crucial early treatments and supports.

Attempts to examine sex differences in ASD with co-occurring conditions have been hindered by insufficiently powered studies with especially small female samples. One study reported that being a female and having lower adaptive skills scores predicted higher inattention severity (Avni et al., 2018), but there are no studies that report on differences in the presentation of females with ASD and co-occurring conditions specifically as compared to males. Therefore, the primary objective of our study is to leverage SPARK, the largest national ASD cohort to date, to examine sex differences in the behavioral presentation of children with ASD and common co-occurring conditions: ADHD (31–95% co-occurrence with ASD (Goldstein & Schwebach, 2004; Leyfer et al., 2006; Lee & Ousley, 2006)) and Anxiety (11–84% co-occurrence with ASD (White et al., 2009)). We aim to examine behavioral features of ADHD and Anxiety that overlap with ASD, namely motor functioning and restricted and repetitive behaviors (RRBs) respectively, as well as differences in ASD symptomatology (social communication) that may emerge in the presence of co-occurring conditions. Further, given sex differences described in typical development in motor coordination (where girls outperform boys (Rivard et al., 2014)) and repetitive behaviors (where boys present with more rigid and routine-bound behavior, as well as restricted interests than girls (Larkin et al., 2017)), we plan to examine whether the presence of co-occurring conditions differentially impacts motor coordination and repetitive behaviors in males and females with ASD. We hypothesize that children with ASD with co-occurring ASD + ADHD will demonstrate more significant motor dysfunction and ASD symptoms than those with ASD alone, where the presence of co-occurring ADHD will be associated with more severe motor deficits and ASD symptoms compared to their intensity/severity in ASD alone. Second, we hypothesize that children with ASD + Anx will demonstrate more significant RRBs and ASD symptoms compared to those with ASD alone, where the presence of co-occurring Anxiety will be associated with more RRBs and ASD symptoms compared to their intensity/ severity in ASD alone. We will further explore these findings by sex, by comparing males and females with ASD with and without co-occurring ADHD and/or anxiety. Finally, we conduct a preliminary examination of age effects, contrasting males and females during school-age and adolescence. We leverage the use of a descriptive study design based upon parent-report in SPARK’s large cohort to provide a foundation for future research in these important but understudied areas.

Methods and materials

Recruitment was conducted for the larger SPARK cohort by Autism centers across the country in the SPARK consortium and via online efforts, thus providing a geographically diverse sample. For an overview of recruitment and eligibility for participation in SPARK, see Feliciano et al., 2018 (Feliciano et al., 2018).

We investigated a subset of the full SPARK cohort, defined by the following inclusion criteria: the participant must: 1) have a parent-reported ASD diagnosis, 2) be between ages 6–18 years, 3) gestational age >32 weeks, 4) have Social Communication Questionnaire (SCQ) (Rutter et al., 2007) score of ≥12, and 5) have valid completion of the measures included in analyses as determined via algorithm (e.g., few missing items). Inclusion criteria resulted in an initial sample of 25,078 children with ASD, including 5,397 (21.5%) females.

SPARK participation was completed online, relying upon parent-report for all measures, including report of prior diagnosis of ASD and presence or absence of co-occurring conditions (ADHD and Anxiety Disorder). Previous studies using large online registries (i.e., Interactive Autism Network [IAN]) suggest that parent-reported ASD diagnostic status is valid (Rutter et al., 2007; Daniels et al., 2012). To further ensure confidence in the validity of ASD diagnostic status, children were only included if their SCQ score was ≥12. This criterion resulted in the exclusion of 2,060 participants (Female N = 566, Male N = 1,494; 27% Female). In the context of under-diagnosis of females, it is worth acknowledging that this cutoff may exclude a subset of females appropriately diagnosed with ASD who do not meet this criterion, as the SCQ may reflect male bias in development and norming of the instrument; however, research suggests that the SCQ performs adequately for both boys and girls with ASD (Evans et al., 2019).

Demographic, medical, and developmental sample characteristics were captured by several questionnaires created specifically for SPARK, including the Individual Data Questionnaire, Basic Medical Screening, and Background History Questionnaire. The following variables were examined (variables that rely on retrospective report are underlined). From the Individual Data Questionnaire, the following variables were extracted for this study: Age at SPARK registration (which corresponds to the child’s current chronological age), sex (Male/Female), ASD diagnosis (yes/No), age at ASD diagnosis, race/ethnicity, and cognitive impairment (“Has child/dependent ever been diagnosed with intellectual disability or cognitive impairment [previously may have been referred to as mental retardation]; yes/No.”). From the Basic Medical Screening, the following variables were captured: ADHD diagnosis (parent report of professional diagnosis; yes/No), Anxiety Disorder diagnosis (parent report of professional diagnosis; yes/No), and gestational age in weeks. From the Background History Questionnaire-Child, the following variables were extracted: Household income and whether or not the child is currently taking prescribed medication (yes/No).

Dependent variable measures

Social communication questionnaire lifetime (SCQ) (Rutter et al., 2007)

The SCQ is a 40-item parent-report measure of child behaviors symptomatic of ASD, including social communication challenges and restricted/repetitive behaviors. Parents respond yes/no to questions concerning both current functioning (items1–19) as well as retrospective symptoms when the child was 4–5 years old (items 20–40). Total scores range from 0–39; higher scores reflect a greater number of reported symptoms. Reliability and validity have been established (Berument et al., 1999; Rutter et al., 2007). A total score cut-off was used for inclusion in the present study to increase confidence in the validity of parent-reported ASD diagnosis (minimum score of 12) and total scores were used to measure dimensional social communication challenges. Participants missing more than three items were excluded from the sample per administration guidelines.

Developmental coordination disorder questionnaire (DCDQ) (Wilson et al., 2009)

The DCDQ is a 15-item measure wherein parents are asked to rate their child’s motor skills in three domains: fine motor, coordination, and motor control. Items are rated on a five-point Likert scale ranging from (1) “Not at all like your child” to (5) “Extremely like your child.” Seven items are reversed scored. Higher scores indicate better motor skills; cut-offs by age are offered such that scores 15–46 (ages 5– 7 years), 15–55 (ages 8– 9 years) and 15–57 (ages 10– 15 years) suggest/indicate Developmental Coordination Disorder. Reliability and validity for this measure have been established in a clinical context (Wilson et al., 2009). The DCDQ Total Score was used as a dependent variable in the present study.

Repetitive behavior Scale- Revised (RBS-R) (Lam & Aman, 2007)

The RBS-R is a 43-item questionnaire asking parents to rate the extent to which their child exhibits behaviors (within the past month) in the following domains: stereotyped behavior, self-injurious behavior, compulsive behavior, ritualistic behavior, sameness behavior, and restricted behavior. Items are rated on a four-point Likert scale ranging from (0) “behavior does not occur” to (3) “behavior occurs and is a severe problem.” Higher scores indicate more severe problems. Reliability and validity for this measure have been established in a clinical context (Lam & Aman, 2007). The RBS-R Total Score was used as a dependent variable in the present study.

Participant samples

All participants had a parent-reported diagnosis of ASD (N = 25,078; Female N = 5,397; Male N = 19,681; Table 1) and met the following inclusion criteria: chronological age between 6 and 17.99 years, gestational age greater than 32 weeks, valid (less than 3 missing items) SCQ scores ≥12, valid data on the Background History form, valid chronological age data, valid individual data, valid data on SPARK’s basic medical measure, and valid scores on the DCDQ and RBS-R (Table 2).

Table 1.

Demographic characteristics of 25,078 participants by sex; see Table 2 for additional characteristics by diagnostic subgroup.

Female Male Difference
Race – N (%) N = 5,397 N = 19,681 X2 = 55.42, p<.0001
 African American 172 (3.2%) 811 (4.1%)
 Asian 61 (1.1%) 270 (1.4%)
 Hawaiian/Pacific islander 25 (0.5%) 20 (0.1%)
 Multiracial 413 (7.7%) 1,647 (8.4%)
 Native American 37 (0.7%) 77 (0.4%)
 Other 150 (2.8%) 604 (3.1%)
 White 3,464 (64.2%) 12,354 (62.8%)
 Missing 1,075 (19.9%) 3,898 (19.8%)
Ethnicity – N (%)
 Hispanic 831 (15.4%) 2,614 (13.3%) X2 = 15.82, p<.0001
 Non-Hispanic 4,566 (84.6%) 17,067 (86.7%)
SES/Family Income – N (%)
 ≤ $20 K/year 471 (8.7%) 1597 (8.1%) X2 = 6.77, p=.15
 > $20 K/year – ≤ $65 K/year 1,382 (25.3%) 5,250 (26.7%)
 > 65 K/year – ≤ $100 K/year 874 (16.2%) 3,029 (15.4%)
 > $100 K/year 1,095 (18.4%) 3,543 (18.0%)
 Missing 1,675 (31.0%) 6,262 (31.8%)
Open in a new tab
Table 2.

Characteristics by sex for the overall sample and by diagnostic subgroup.

Overall ASD Only ASD + ADHD ASD + Anx ASD + ADHD + Anx Subgroup Difference
Sex distribution N = 25,078 N = 10,804 N = 7,707 N = 2,045 N = 4,522
 Female – N (%) 5,397 (21.5%) 2,536
(23.5%)		 1,268 (16.5%) 595 (29.1%) 998 (22.1%) X2 = 211.87*
 Male – N (%) 19,681 (78.5%) 8,268
(76.5%)		 6,439 (83.5%) 1450 (70.9%) 3,524 (77.9%)
 Sex ratio vs. ASD Only Reference More males* More females* ns
Current age in years
 Female – Mean (SD) 10.40 (3.24) 9.70 (3.18) 10.21 (2.99) 11.64 (3.32) 11.69 (3.06) ASD Only< ASD + ADHD < ASD + Anx = ASD + Anx + ADHD
 Male – Mean (SD) 10.48 (3.26) 10.04 (3.36) 10.40 (3.12) 11.42 (3.26) 11.30 (3.04)
 Sex difference ns Males older* Males older* ns Females older*
Age at diagnosis
 Female – Mean (SD) 5.45 (3.31) 4.28 (2.55) 5.81 (3.12) 6.75 (4.00) 7.21 (3.62) ASD Only< ASD + ADHD < ASD + Anx< ASD + Anx + ADHD
 Male – Mean (SD) 5.06 (2.94) 4.19 (2.47) 5.42 (2.93) 5.47 (3.11) 6.26 (3.27)
 Missing – N (%) 30 (.1%) 10 (.09%) 13 (.17%) 2 (.10%) 5 (.11%)
 Sex difference Females older* ns Females older* Females older* Females older*
Intellectual disability – yes
 Female – N (%) 1,240 (23.0%) 565 (22.3%) 325 (25.8%) 116 (19.6%) 234 (23.4%) Females: X2 = 10.23* Males: ns
 Male – N (%) 4,121 (20.1%) 1,760 (21.3%) 1,351 (21.0%) 306 (21.1%) 704 (20.0%)
 Missing – N (%)
 Sex difference More females* ns More females* ns More females*
Prescribed medication - yes
 Female – N (%) 2,112 (39.1 %) 513 (20.2%) 673 (53.1%) 288 (48.4%) 638 (63.9%) ASD + ADHD + Anx > ASD + ADHD > ASD + Anx > ASD Only
 Male – N (%) 8,430 (42.8%) 1,829 (22.1%) 3,577 (55.6%) 706 (48.7%) 2318 (65.8%)
 Missing – N (%)
 Sex difference More males* More males* ns ns ns
Gestational age 33–36 wks
 Female – N (%) 514 (9.5%) 217 (8.6%) 131 (10.3%) 61 (10.3%) 105 (10.5%) X2 = 43.40*
 Male – N (%) 1,886 (9.6%) 686 (8.3%) 619 (9.6%) 168 (11.6%) 413 (11.7%) ASD < ASD + ADHD < ASD + Anx < ASD + ADHD + Anx
 Sex difference ns ns ns ns ns
Open in a new tab

Notes:

*

p=.05 or less.

Fixed factors

Diagnostic groups were defined as ASD alone (no ADHD or Anxiety), ASD + ADHD, ASD + Anx, ASD + ADHD + Anx (tri-occurring). Sex was coded as female = 0, male = 1 based on parent report. Given research showing developmental- and cohort-based differences in ASD phenotype, participants were grouped into younger/school age (6–11.99 years) and older/adolescent (12–17.99 years) age bins to facilitate the interpretation of age-related effects.

Dependent variables

All 25,078 participants had SCQ scores (Total Score), 15,285 had DCDQ scores (Total Score; Female N = 3,329; Male N = 11,956), and 17,430 had RBS-R scores (Total Score; Female N = 3,772; Male N = 13,658).

Statistical approach

Primary analyses were conducted using generalized linear models (GLM) from the ‘stats’ package in R (R Core Team, 2020). Overall scores for each primary dependent variable of interest (SCQ, DCDQ, RBS-R) were modeled as a function of the interaction between diagnostic category (ASD alone, ASD + ADHD, ASD + Anx, ASD + ADHD + Anx), sex (female, male), and age (younger: 6-<12 years; older: 12-<18 years). Given the complexity of the model and the sample size, Tukey-corrected estimated marginal means (EMM) were compared using the ‘em means’ package in R (Lenth, 2020). Although results were corrected for all possible pairwise comparisons to be conservative, the results section focuses primarily on patterns in the ASD alone sample compared to ASD plus co-occurring ADHD and/or Anxiety by sex and age. Effect sizes were calculated using raw means and pooled standard deviations, reported using Cohen’s d, and interpreted as small (d = 0.2), medium (d = 0.5), and large (d = 0.8) based on benchmarks suggested by Cohen (Cohen, 1988).

Preliminary analyses

To carefully examine the impact of important additional demographic and medical factors on our findings, we completed all analyses with a smaller and more conservative subsample that met all criteria for the larger sample, had complete data for all three outcome variables, and had complete data indicating the following: gestational age greater than 37 weeks, no cognitive impairment, valid annual household income data, and race data. Primary analyses presented below are from the larger sample as the pattern of findings remained largely consistent in the smaller, more conservative subsample.

Results

Demographic variables

Demographic information by sex is reported in Table 1, and by sex/diagnostic subgroup in Table 2. Examined across age bins (i.e., in the overall sample), females with ASD alone (9.70 years) were slightly younger than males with ASD alone (10.04 years, t=−4.65, p < 0.001, Cohen’s d=.19). Current chronological age differed by diagnostic subgroup, with slightly older participants in the co-occurring conditions subgroups (both males and females). Females with tri-occurring conditions (ASD + ADHD + Anx) were the oldest (11.69 years), and were significantly older than males with tri-occurring conditions (11.30 years; t = 3.363, p=.02, Cohen’s d=.22). Across the group as a whole, age at initial ASD diagnosis differed by sex, with females diagnosed later than males (females: 5.45 years, males: 5.06 years; t=−8.48, p<.001, Cohen’s d=.22). Children with co-occurring conditions were diagnosed later than children with ASD alone (Figure 1). Females with ASD + ADHD + Anx were the latest to receive the ASD diagnosis (7.21 years), which is significantly later than males with the same tri-occurring conditions (6.26 years, t = 9.11, p<.001, Cohen’s d=.51). Sex distribution differed by diagnostic subgroup (Chi-squared = 211.87, p<.0001). Compared to ASD alone (23% female, 77% male), females were over-represented in the ASD + Anx group (29%) and males were under-represented (71% male; Chi-squared comparing the sex distribution in the ASD sample to the sex distribution in the ASD + Anx sample = 29.191, p<.0001; Figure 2). In contrast, females were under-represented and males were over-represented in the ASD + ADHD group compared to the ASD alone group (ASD + ADHD: 16% female, 84% male; Chi-squared = 135.35, p<.0001). Sex ratios in the tri-occurring group (ASD + Anx + ADHD) did not differ from ASD alone (p=ns). Overall, females were slightly more likely than males to have cognitive impairment (females: 23%, males: 21%, Chi-squared = 10.67, p=.001). The distribution of cognitive impairment differed by diagnostic subgroup in females, such that females with ASD + Anx had the lowest rate of cognitive impairment (19.5%) and ASD + ADHD had the highest rate of cognitive impairment (25.8%; Chi-squared = 10.23, p=.02). The distribution of cognitive impairment across diagnostic subgroups did not differ for males (p=ns). Parent-reported medication intervention was greater in subgroups with co-occurring conditions than in ASD alone (ASD: 22%, ASD + ADHD: 55%, ASD + Anx: 49%, ASD + ADHD + Anx: 65%; Chi-squared = 3428, p<.0001), and overall greater in males (43%) than females (39%; Chi-squared = 23.65, p<.0001; Table 2). Lastly, both females (10.5%) and males (11.7%) with tri-occurring conditions were more likely to have a history of prematurity, when compared to those with ASD alone (females: 8.3%, males: 8.6%; Chi-squared=X2=43.40, p<.01; Table 2).

Figure 1.

Figure 1.

Open in a new tab

Mean age at autism spectrum disorder (ASD)diagnosis (in years) by sex, age, and diagnostic category. Not significant: nsASD: Autism Spectrum Disorder; ASD_ADHD: ASD with co-occurring attention deficit/hyperactivity disorder (ADHD); ASD_Anx: ASD with co-occurring anxiety disorder; ASD_Anx_ADHD: ASD with co-occurring ADHD and anxiety disorder. Total N = 25,078; Female N = 5,397; Male N = 19,681. Please refer to results section to appreciate significant findings.

Figure 2.

Figure 2.

Open in a new tab

Social Communication Questionnaire (SCQ) total scores by sex, age bin, and diagnostic category, represented as estimated marginal means. Higher scores indicate more social communication challenges. ASD: Autism Spectrum Disorder; ASD_ADHD: ASD with co-occurring attention deficit/hyperactivity disorder (ADHD); ASD_Anx: ASD with co-occurring anxiety disorder; ASD_ Anx_ADHD: ASD with co-occurring ADHD and anxiety disorder. Total N = 25,078; Female N = 5,397; Male N = 19,681. Please refer to results section to appreciate significant findings.

Social communication questionnaire

Omnibus GLM revealed a significant interactive effect of diagnostic group, sex, and age groups on SCQ scores (Figure 2). To illuminate the directionality of the interaction, Tukey-corrected comparisons of estimated marginal means were examined. Below, results are organized for readability.

SCQ scores in females

Older females with ASD alone had higher SCQ scores (i.e., more autism symptomology) than younger females with ASD alone (z=−10.845, p<.0001, Cohen’s d=.91), but females’ SCQ scores did not differ by age group for any other diagnostic category (all p=ns). Young females’ SCQ scores increased significantly in the presence of co-occurring conditions (ASD alone vs. ASD + ADHD, z=−5.525, p<.0001, Cohen’s d=.44; ASD vs. ASD + Anx, z=−3.815, p = 0.013, Cohen’s d=.46; ASD alone vs. ASD + ADHD + Anx, z=−5.936, p<.0001, Cohen’s d=.58). In contrast, older females’ SCQ scores decreased in the presence of Anxiety or ADHD (ASD alone vs. ASD + ADHD, z = 3.773, p=.0152, Cohen’s d=.45; ASD alone vs. ASD + Anx, z = 5.859, p<.0001, Cohen’s d=.76). However, the SCQ scores of the ASD only group did not differ from SCQ scores of adolescents with tri-occurring conditions (ASD alone vs. ASD + ADHD + Anx, z = 2.818, p=ns).

SCQ scores in males

Consistent with the pattern found in females, older males with ASD alone had higher SCQ scores than younger males with ASD alone (z=−11.04, p<.001, Cohen’s d=.51). Males’ SCQ scores did not differ by age bin for any other diagnostic subgroup. In young males, SCQ scores did not differ in ASD alone vs. ASD in the presence of co-occurring conditions (all p=ns). Older males with ASD alone had higher SCQ scores than older males with co-occurring conditions (ASD alone vs. ASD + ADHD, z = 9.388, p<.0001, Cohen’s d=.53; ASD alone vs. ASD + ADHD + Anx, z = 8.613, p<.0001, Cohen’s d=.53) except Anxiety (ASD alone vs. ASD + Anx, z = 2.54, p=ns).

Sex differences in SCQ scores

Tukey-corrected EMM comparisons showed that young females with ASD alone had fewer ASD symptoms than young males with ASD alone (z=−12.998, p<.0001, Cohen’s d=.69); other within-diagnosis sex differences in the younger group were not significant (all p=ns). Older females had similar symptoms to older males with the same diagnoses (p=ns), except older females with ASD + Anx had significantly lower SCQ scores than older males with ASD + Anx (z=−6.311, p<.0001, Cohen’s d=.83).

Developmental coordination disorder questionnaire

Omnibus GLM revealed a significant interactive effect of diagnostic group, sex, and age group on DCDQ scores (Figure 3).

Figure 3.

Figure 3.

Open in a new tab

Developmental Coordination Disorder Questionnaire (DCDQ) total scores by sex, age bin, and diagnostic category, represented as estimated marginal means. Higher scores indicate better motor skills. ASD: Autism Spectrum Disorder; ASD_ADHD: ASD with co-occurring attention deficit/hyperactivity disorder (ADHD); ASD_Anx: ASD with co-occurring anxiety disorder; ASD_ Anx_ADHD: ASD with co-occurring ADHD and anxiety disorder. Total N = 15,285; Female N = 3,329; Male N = 11,956. Please refer to results section to appreciate significant findings.

DCDQ scores in females

Younger females with ASD alone had better motor skills (higher DCDQ scores) than older females with ASD alone (z = 5.39, p<.0001, Cohen’s d=.56), but scores did not differ by age bin for any other diagnostic subgroup (all p=ns). Young females with co-occurring ADHD diagnoses had lower motor functioning (ASD alone vs. ASD + ADHD, z = 9.985, p<.0001, Cohen’s d=.89; ASD alone vs. ASD + ADHD + Anx, z = 10.857, p<.0001, Cohen’s d = 1.19) but not Anxiety (ASD alone vs. ASD + Anx, p=ns). This pattern was broadly similar in older females, but did not reach statistical significance (all p=ns).

DCDQ scores in males

In contrast to the age-related pattern found in females, younger males with ASD + Anx had slightly lower DCDQ scores (i.e., poorer motor skills) than older males with ASD + Anx (z=−3.69, p<.05, Cohen’s d=.45). DCDQ scores did not differ by age for males in the ASD only, ASD + ADHD, or ADHD + ADHD + Anx groups (all p=ns).Within each age group separately, the DCDQ scores of younger and older males with ASD alone did not differ significantly from same-aged males with co-occurring ADHD or Anxiety (all p=ns), but they did differ from tri-co-occurring males, such that older and younger males with ASD + ADHD + Anx had lower scores(worse motor abilities)as compared to same-aged males with ASD alone (young: ASD alone vs. ASD + ADHD + Anx, z = 6.96, p<.0001, Cohen’s d=.39; old: ASD alone vs. ASD + ADHD + Anx, z = 4.953, p<.001, Cohen’s d=.38).

Sex differences in DCDQ scores

Younger females with ASD alone or with ASD + Anx had higher DCDQ scores (better motor skills) than males with the same diagnoses (ASD alone: z = 12.553, p<.0001, Cohen’s d=.72; ASD + Anx: z = 4.665, p=.0003, Cohen’s d=.67). There were no significant sex differences in the DCDQ scores of school-aged children with ASD + ADHD + Anx, nor were there any significant within-diagnosis sex differences in the older groups of participants with ASD, ASD + ADHD, ASD + Anx, or ASD + ADHD + Anx (all p>.05).

Repetitive behaviors scale – revised

Omnibus GLM revealed a significant interactive effect of diagnostic group, sex, and age group on RBS-R scores (Figure 4).

Figure 4.

Figure 4.

Open in a new tab

Repetitive Behaviors Scale – Revised (RBS-R) Total Scores by sex, age bin, and diagnostic category, represented as estimated marginal means. Higher scores indicate more restricted/repetitive behaviors. ASD: Autism Spectrum Disorder; ASD_ADHD: ASD with co-occurring attention deficit/hyperactivity disorder (ADHD); ASD_Anx: ASD with co-occurring anxiety disorder; ASD_Anx_ADHD: ASD with co-occurring ADHD and anxiety disorder. Total N = 17,430; Female N = 3,772; Male N = 13,658. Please refer to results section to appreciate significant findings.

RBS-R scores in females

Younger females consistently demonstrated more RRBs than older females, across all diagnostic groups [ASD alone (z = 31.291, p<.0001, Cohen’s d = 2.27), ASD + ADHD (z = 11.24, p<.0001, Cohen’s d = 1.19), ASD + Anx (z = 10.86, p<.0001, Cohen’s d = 1.98), and ASD + ADHD + Anx (z = 15.13, p<.0001, Cohen’s d = 1.63)]. Within adolescence, older females with any co-occurring condition had significantly more RRBs than older females with ASD alone (ASD alone vs. ASD + ADHD, z=−7.893, p<.0001, Cohen’s d=.72; ASD alone vs. ASD + Anx, z=−7.908, p<.0001, Cohen’s d=.80; ASD alone vs. ASD + ADHD + Anx, z=−18.56, p<.0001, Cohen’s d = 1.69). In the school-aged group, young females with ASD alone had more RRB symptoms than young females with ASD + ADHD (z = 6.873, p<.0001, Cohen’s d=.46), but not ASD + Anx (z = 1.026, p=ns). Young females with tri-occurring conditions demonstrated significantly greater RRB symptoms than ASD alone (z=−10.824, p<.0001, Cohen’s d=.88).

RBS-R scores in males

As in females, younger males had more RRBs than older males across all diagnostic groups [ASD alone (z = 29.800, p<.0001, Cohen’s d = 1.11), ASD + ADHD (z = 29.351, p<.0001, Cohen’s d = 1.25), ASD + Anx (z = 15.592, p<.0001, Cohen’s d = 1.23), and ASD + ADHD + Anx (z = 23.126, p<.0001, Cohen’s d = 1.29)].In the school-aged group, young males with co-occurring conditions had more RRBs than young males with ASD alone (ASD alone vs. ASD + ADHD, z=−21.564, p<.0001, Cohen’s d=.71; ASD alone vs. ASD + Anx, z=−19.667, p<.0001, Cohen’s d = 1.20; ASD alone vs. ASD + ADHD + Anx, z=−48.005, p<.0001, Cohen’s d = 2.05), with the same pattern evident in older males (ASD alone vs. ASD + ADHD, z=−13.473, p<.0001, Cohen’s d=.60; ASD alone vs. ASD + Anx, z=−15.106, p<.0001, Cohen’s d=.96; ASD alone vs. ASD + ADHD + Anx, z=−38.281, p<.0001, Cohen’s d = 1.85).

Sex differences in RBS-R scores

Tukey-corrected EMM comparisons showed that young females with ASD alone showed more RRBs than young males with ASD alone (z = 33.122, p<.0001, Cohen’s d = 1.39), but the RBS-R scores of older females with ASD alone did not differ from the RBS-R scores of older males with ASD alone (p=ns). Both younger and older females with ASD + ADHD showed more RRBs than males (younger sample: z = 3.992, p=.0066, Cohen’s d=.24; older sample: z = 4.159, p=.0033, Cohen’s d=.34). There was no significant sex difference in RRBs for younger or older participants with ASD + Anx (all p=ns). Finally, younger females with ASD + ADHD + Anx showed a difference in RRBs from younger males (more symptoms: z = 3.569, p=.0312, Cohen’s d=.31), but older females and males with tri-occurring conditions did not have significantly different RRBs (z=.04, p=ns).

Discussion

The goal of this study was to describe differences in the behavioral presentation of females and males with ASD and common co-occurring conditions by age (school age and adolescence), a topic that has been largely understudied due to limitations associated with sample sizes of females included in research. By leveraging SPARK, we were able to study over 5,000 females with ASD, and further subgroup these females by age bin (school age and adolescence) and co-occurring conditions (ASD alone, ASD + ADHD, ASD + Anx, and ASD + ADHD + Anx), thereby reducing within-subsample heterogeneity while specifying the profile of females with ASD (and contrasting that profile with their male counterparts). A sample of this nature does not yet exist in the literature, and thus reflects an important next step toward understanding sex differences in ASD. Deeper understanding of these sex differences will promote earlier and more accurate diagnosis for females with ASD, and ultimately, will be valuable for developing interventions and supports that are tailored to the needs of females on the autism spectrum.

Demographic differences in ASD and co-occurring conditions

Given the large sample described here, our first analysis effort focused on assessing demographic differences by sex and diagnostic category. We found that both females and males with co-occurring conditions were diagnosed at an older age, and that participants with tri-occurring conditions were latest to be diagnosed. This effect was stronger in females than in males, with females with tri-occurring conditions diagnosed about 1 year later than males with tri-occurring conditions. Further, females with ASD and co-occurring conditions were diagnosed later than females with ASD alone, and this was particularly true for females with Anxiety, who were diagnosed almost 2 years later than females with ASD alone; notably, females were also significantly over-represented in the ASD + Anx sample (29% female in this subsample as compared to 22% female in the overall sample), suggesting that co-occurring Anxiety is a more common presentation in females with ASD than in males with ASD. Conversely, females were relatively under-represented in the ASD + ADHD subgroup (17% female compared to 22% female in the overall sample), which may reflect the compounding effects of greater male representation in ASD and ADHD diagnoses separately, resulting in even stronger male representation in the comorbid group. When examining level of functioning, the rate of co-occurring cognitive impairment was roughly 20% in the sample as a whole; however, co-occurring conditions only affected females with ASD and co-occurring Anxiety, who had a slightly lower rate of cognitive impairment (18%). Further, and not surprisingly, the presence of co-occurring conditions appears to increase the likelihood of being prescribed medication, and overall, males were more likely to be prescribed medication than females.

Taken together, these findings suggest that the presence of co-occurring conditions could delay both ASD diagnosis and access to resources (e.g., study participation), particularly for females. This may be due to the misattribution of ASD symptoms to co-occurring conditions, delaying ASD diagnosis. Overlapping symptoms of Anxiety (potentially related to repetitive behavior features being attributed to Anxiety rather than ASD, see further discussion below) could contribute to this phenomenon, particularly for females. Additionally, given the lower rate of cognitive impairment in females with ASD + Anx, these individuals may appear less functionally impaired (i.e., better able to rely upon their intact cognitive skills to manage or camouflage symptoms of ASD), thus making them more difficult to diagnose. Given that there are proportionally more females in the groups with ASD and co-occurring Anxiety than in other groups, the concerning implication is that females with this diagnostic combination may be diagnosed later, potentially postponing access to care.

Social communication in ASD and co-occurring conditions

Consistent with an ASD diagnosis, all of our diagnostic groups had clinically significant weaknesses in social and communication behaviors. Males had elevated levels of ASD symptom atology compared to females across diagnostic groups, although this difference only reached statistical significance for younger participants with ASD alone and adolescents with ASD + Anx. Younger females with ASD alone were reported to have the fewest symptoms compared to the other groups. Thus, the impact of co-occurring conditions on ASD-related social communication symptomatology appears to be most prominent in young females; specifically, for young females, the addition of co-occurring ADHD or Anxiety alone significantly increased reported symptomatology, and those with ASD + ADHD + Anx reported the most ASD-related social communication symptoms. This pattern was not seen in older females, or younger or older males.

The pattern demonstrated by younger females is consistent with the larger literature, suggesting that females with ASD may appear to have, in general, fewer or less prominent ASD symptoms than males (Lenth, 2020). Notably, only young females with ASD alone (without co-occurring ADHD and/or Anxiety) differed from their male counterparts on SCQ scores, suggesting that young females with ASD alone may show fewer social communication symptoms, and could be at greatest risk for remaining unidentified. Younger females with co-occurring conditions, as well as most older females were described as more similar to their male counterparts, thereby potentially reducing their risk of being misdiagnosed (the exception is older females with ASD + Anx, who also had significantly lower SCQ scores than their male counterparts). It may be that in younger females’, co-occurring symptoms prompt time lier referral for ASD evaluation (over ASD symptoms alone) or that for older females, ASD symptoms have grown to a point where they alone prompt evaluation. Either way, this suggests that females may require a greater number of autism symptoms or more evident behaviors in order to be considered problematic by parents.

Another notable finding is that concerns for social/communication skills decrease in older females who have co-occurring ADHD and/or Anxiety compared to older females with ASD alone. Although this pattern is also evident in older boys with ASD + ADHD and tri-occurring conditions (but not ASD + Anx), it is particularly salient in girls with ASD + Anx because they were diagnosed 2 years later than girls with ASD alone. Future research should explore whether girls’ (potentially unintentional) attempts to camouflage or mask their autism symptoms decreases (parent, clinician, or self) perceptions of social communication impairment while increasing anxiety and/or taxing executive functioning. Further, potential impact of clinician bias should be considered in future research, as clinicians may be prone to focus on higher frequency behavior (i.e., anxiety in females) versus lower frequency ASD symptomatology.

Motor functioning in ASD and co-occurring conditions

All of our diagnostic groups had clinically significant weaknesses in motor skills (DCDQ scores). As hypothesized, diagnostic group differences in motor functioning were driven primarily by co-occurring conditions, particularly ADHD status. This finding was most apparent for young females, wherein young females with ASD + ADHD or ASD + ADHD + Anx had weaker motor skills than young females ASD alone, while those with ASD + Anx did not differ in their motor skills from ASD alone. A similar, but statistically non-significant pattern was observed in older females. In males, motor skills were only significantly impacted by the addition of both ADHD andAnxiety. Importantly, younger females had better motor skills than younger males in ASD alone and ASD + Anx but not ASD + ADHD or ASD + ADHD + Anx, again highlighting the notable impact of co-occurring ADHD on motor functioning in younger females.

Younger females’ motor skills were most impacted by the presence of co-occurring conditions, and the presence of ADHD highly affected motor functioning in these young females. Consistent with the literature, young females with ASD alone had better motor skills than young males with ASD alone, but with the addition of ADHD, their motor skills fell below their male counterparts. This is a finding not yet reported in the literature, even when comparing females and males with Developmental Coordination Disorder (Rivard et al., 2014). This again reinforces the potential vulnerability of young females with ASD without co-occurring conditions to have their symptoms underestimated or overlooked, extending to symptoms outside of the core features of ASD (i.e., motor skills). Further, our findings suggest that for all children with ASD, the addition of co-occurring ADHD should warrant further and more careful examination of motor functioning, to inform treatment planning.

Repetitive behaviors in ASD and co-occurring conditions

As expected, all diagnostic groups were reported to have elevated concerns for RRBs. Interestingly, younger children had greater RRBs than older adolescents across both sexes, which may reflect DSM-related changes to the definition of ASD that placed greater emphasis on the restricted/repetitive behaviors domain. Within females, younger participants showed a strikingly different pattern than older females. Whereas co-occurring Anxiety or ADHD was not associated with increased RRBs in young females, these co-occurring disorders were associated with increased RRBs in older females. In both young and older samples, females with ASD + ADHD + Anx demonstrated the most RRBs. There are many potential explanations for this finding. For example, it may be that older females rely on different coping mechanisms than younger females (i.e., engage in more on RRBs for self-soothing in the context of increased symptoms related to co-occurring conditions). Alternatively, it may be that Anxiety symptoms in younger females manifest differently than in adolescent females with less overlap with RRBs (e.g., compulsions, self-injury).

One of our most compelling findings was that sex differences exist in the presentation of RRBs among females and males with ASD and co-occurring conditions. Specifically, young females with ASD alone were reported to have more RRBs than young males with ASD alone, with young females with ASD + ADHD + Anx reported to have the most significant RRBs. In contrast to these findings, the literature largely suggests that males display more RRBs than females, as was recently reported in a large, multisite study (Kaat et al., 2020)citing on findings from the Autism Diagnostic Observations Scale (ADOS) (Lord et al., 2012) and Autism Diagnostic Interview (ADI) (Lord et al., 1994). Notably, our study reports findings from the RBS-R, which may measure a wider breadth of RRBs and interests compared to the ADOS and ADI, potentially better capturing these symptoms in young females. Indeed, another recent report (Antezana et al., 2019) supports the utility of identifying sex differences via item analysis on the RBS-R. Specifically, items related to insistence on sameness, and compulsive, hoarding, and self-injurious behavior (e.g., pulling hair, scratching self), were elevated in females and contributed to the RBS-R items that best distinguished sex; the authors suggest that these particular RRBs align most with executive functioning problems and internalizing symptoms (Antezana et al., 2019).Our study uniquely captures increased RRBs, particularly in school-age females with multiple co-occurring conditions, a finding that has not been previously highlighted in the literature. Given that young females with ASD alone are reported to have fewer social/communication symptoms without notable internalizing or externalizing problems, their diagnosis may be driven by higher levels of reported/observed RRBs. Notably, our hypothesis that RRBs would be increased in those with co-occurring Anxiety was supported for younger and older males, and older females, but not younger females. However, RRBs were also found to increase in the presence of co-occurring ADHD for younger and older males, and older females. Taken together, our pattern of results suggests that for young females in particular, omnibus measures of ASD symptomatology may not be sufficient to classify RRBs or capture the overlap between RRBs and Anxiety and/ or hyperactivity/impulsivity, and a more detailed interrogation of these skills (potentially by interview or by standardized measures; e.g., RBS-R) should be considered. This topic requires further study (including RBS-R item analysis, which is an important future direction of our research program).

Limitations

The sheer size of the SPARK cohort is a considerable strength of this study that allowed us to create groups defined by sex, age, and co-occurring conditions. However, it also offered highly powered statistical analyses, where statistical significance may not equate to clinical significance. In addition to strict correction for all possible multiple comparisons, we therefore attempted to carefully interpret and describe only patterns in our findings that could hold clinical relevance; we also report and consider effect sizes in our interpretation. The SPARK sample is also limited by reliance on parent-report for all information, including diagnosis (both of ASD and the presence or absence of co-occurring conditions) and dependent variables (which are therefore correlated by nature of parent report). This limits the questions that can be asked of the data, including our ability to explore the impact of ID or global developmental delay on our findings. Further, though other symptoms of interest overlap between ASD, ADHD, and Anxiety (e.g., executive functioning), data in other domains are not collected by the SPARK registry. Prior research has supported the validity of online parent report of ASD diagnosis, but the reliability of parent-reported co-occurring conditions has not been specifically examined. As such, we limited the scope of this project to be descriptive in nature, relying upon the breadth and size of the sample to protect against potential systematic bias in parental report. Positively, except for some developmental questions and SCQ scores, our outcomes ask parents to report on current functioning and thus do not rely on retrospective report. SPARK is making efforts to better address limitations associated with reliance on parent report by collecting clinical IQ data. SPARK intends to report on any differences in parent reported and measured IQ in the SPARK cohort to support future research projects. SPARK also supports projects that aim to re-contact participants to gather information on other areas of concern for individuals with ASD (e.g., executive functioning, sensory processing).

The present study is limited by its cross-sectional design; thus, potential cohort effects should be considered. The literature indicates that co-occurring conditions, including Anxiety, are often diagnosed in later childhood and adolescence (Lord et al., 1994), and thus, children who enter SPARK at older ages are more likely to have had the opportunity to present with symptoms and to receive a diagnosis. Further, as some measures used in the present study rely upon retrospective reporting (e.g., SCQ, Developmental History), telescoping effects may have influenced our results, wherein parents of older children seem to recall symptoms as having been more severe than those reporting symptoms at a time more proximal to the present (Hus et al., 2011; Ozonoff et al., 2018). Lastly, as autism awareness has increased exponentially over the past decade and diagnostic criteria have changed over time, older children who received their ASD diagnosis historically earlier may have required more severe symptoms to receive a diagnosis than those receiving diagnosis in the past 5–10 years, and thus older children in our sample may present with more significant symptoms (Keyes et al., 2012). For these reasons, we considered age closely in our analysis, and examined developmental age bins individually (school age vs. adolescence); however, future studies should employ a longitudinal design to fully assess the impact of age on the presentation of ASD and ASD with co-occurring conditions, clinically validating all diagnoses, and using direct measurement and observation of outcomes.

Finally, we acknowledge that the SPARK registry under-represents minority groups. As such, our findings may not generalize to the greater and more diverse ASD population. We are pleased that SPARK is undertaking efforts to engage minority groups, including developing materials in Spanish and providing additional funding opportunities for clinical sites to target under-represented groups where health disparities exist (Tregnago & Cheak-Zamora, 2012). Future research exploring differential symptom presentation in a more diverse sample of males and females with ASD and co-occurring conditions is sorely needed.

Conclusions

Our study highlights the vulnerability of school-age females with ASD for potential mis- or under diagnosis, and extends our understanding of sex differences to symptoms outside of the core diagnostic features of ASD (i.e., motor skills). We further highlight the differential impact of co-occurring ADHD and/or Anxiety on symptom presentation in younger and older females, supporting the consideration of symptom specific measurement in females with potential co-occurring conditions.

Acknowledgements

We would like to acknowledge the families who participated in SPARK.

SPARK Consortium members: Leonard Abbeduto, Gabriella Aberbach, Shelley Aberle, Gupta Abha, John Acampado, Andy Ace, Kaitlyn Ahlers, Charles Albright, Michael Alessandri, Nicolas Alvarez, David Amaral, Alpha Amatya, Alicia Andrus, Claudine Anglo, Rob Annett, Eduardo Arzate, Irina Astrovskaya, Kelli Baalman, Melissa Baer, Gabriele Baraghoshi, Nicole Bardett, Sarah Barnes, yan Bartholomew, Asif Bashar, Heidi Bates, Katie Beard, Juana Becerra, Malia Beckwith, Landon Beeson, Josh Beeson, Brandi Bell, Monica Belli, Dawn Bentley, Natalie Berger, Anna Berman, Raphael Bernier, Elizabeth Berry-Kravis, Mary Berwanger, Shelby Birdwell, Elizabeth Blank, Stephanie Booker, Aniela Bordofsky, Erin Bower, Catherine Bradley, Stephanie Brewster, Elizabeth Brooks, Aliso Brown, Melissa Brown, Jennylyn Brown, Cate Buescher, Martin Butler, Eric Butter, WentengCaI, Norma Calderon, Kristen Callahan, AlexiesCamba, Claudia Campo-Soria, Paul Carbone, Laura Carpenter, Lindsey Cartner, MyriamCasseus, Lucas Casten, Ashley Chappo, Tia Chen, Wubin Chin, SharmistaChintalapalli, Daniel Cho, Dave Cho, yB Choi, Wendy Chung, Renee Clark, Cheryl Cohen, Kendra Coleman, CostanzaColumbi, Joaquin Comitre, Sarah Conyers, Lindsey Cooper, Leigh Coppola, Lisa Cordiero, Jeanette Cordova, Dahriana Correa, Hannah Cottrell, Michelle Coughlin, Eric Courchesne, Dan Coury, Joseph Cubeis, Sean Cunningham, Mary Currin, Michele Cutri, Sophia D’Ambrosi, Amy Daniels, Sabrina Davis, Nickelle Decius, Jennifer Delaporte, Brandy Dennis, Kate Dent, Gabrielle Dichter, Katharine Diehl, Chris Diggins, Emily Dillon, Erin Doyle, Andrea Drayton, Megan DuBois, Gabrielle Duhon, Megan Dunlevy, Rachel Earl, Catherine Edmonson, Sara Eldred, Barbara Enright, Craig Erickson, Amy Esler, Anne Fanta, Carrie Fassler, FarisFazal, Pamela Feliciano, Angela Fish, Kate Fitzgerald, Chris Fleisch, Eric Fombonne, Emily Fox, Sunday Francis, Margot Frayne, Sandra Friedman, Laura Fuller, Virginia Galbraith, Swami Ganesan, Jennifer Gerdts, Mohammad Ghaziuddin, HaidarGhina, David Giancarla, Erin Given, Jared Gong, Kelsey Gonring, Natalia Gonzalez, Antonio Gonzalez, Rachel Gordon, Catherine Greay, Tunisia Greene, Ellen Grimes, Luke Grosvenor, Amanda Gulsrud, Jaclyn Gunderson, Chris Gunter, Anibal Gutierrez, Frampton Gwynette, Melissa Hale, Lauren K. Hall, Jake Hall, Kira Hamer, Bing Han, Nathan Hanna, Antonio Harden, Eldric Harrell, Jill Harris, Nina Harris, Caitlin Hayes, TerynHeckers, Kathryn Heerwagen, Susan Hepburn, Lynette Herbert, Clara Herrera, Brittani Hilscher, Kathy Hirst, Theodore Ho, Dabney (Eugenia) Hofammann, Margaret Hojlo, Gregory Hooks, Dain Howes, Lark Huang-Storm, Samantha Hunter, Hanna Hutter, Teresa Ibanez, Dalia Istephanous, Suma Jacob, Andrea Jarratt, Stanley Jean, Anna Jelinek, William Jensen, Mya Jones, Mark Jones, Alissa Jorgenson, Jessyca Judge, Taylor Kalmus, Stephen Kanne, Hannah Kaplan, Lauren Kasperson, So Hyun Kim, Annes Kim, Cheryl Klaiman, Robin Kochel, MisiaKowanda, Melinda Koza, Sydney Kramer, Eva Kurtz-Nelson, Hoa Lam, Elena Lamarche, Erica Lampert, Rebecca Landa, Alex Lash, Jessica K Law, Noah Lawson, Holly Lechniak, CD Lehman,BruceLeight, Laurie Lesher, Deana Li, Robin Libove, Natasha Lillie, Danica Limon, Desi Limpoco, Nathan Lo, Brandon Lobisi, Marilyn Lopez, Daniella Lucio, Addie Luo, Audrey Lyon, Natalie Madi, Malcolm Mallardi, Lacy Malloch, AnupMankar, Patricia Manning, Julie Manoharan, Olena Marchenko, Richard Marini, Christa Martin, Gabriela Marzano, Sarah Mastel, Sheena Mathai, Clara Maxim, Caitlin McCarthy, Nicole Mccoy, Julie McGalliard, Anne-Marie McIntyre, Brooke McKenna, Alexander McKenzie, Megan McTaggart, Sophia Melnyk, Alexandra Miceli, Sarah Michaels, Jacob Michaelson, Anna Milliken, Amanda Moffitt Gunn, Sarah Mohiuddin, Jessie Montezuma, Amy Morales-Lara, Kelly Morgan, Hadley Morotti, Michael Morrier, Maria Munoz, Karla Murillo, Kailey Murray, Vincent Myers, Natalie Nagpal, Jason Neely, Katelyn Neely, Olivia Newman, Richard Nguyen, Victoria Nguyen, Amy Nicholson, Melanie Niederhauser, Megan Norris, Kaela O’Brien, Mitchell O’Meara, Molly O’Neil, Brian O’Roak, Eirene O’Connor, Edith Ocampo, Cesar Ochoa-Lubinoff, Jessica Orobio, Elizabeth (Libby) Orrick, Crissy Ortiz, Opal Ousley, MotunrayoOyeyemi, Samiza Palmer, Katrina Pama, Juhi Pandey, Katherine Pawlowski, Michah Pepper, Diamond Phillips, Karen Pierce,JosephPiven, Jose Polanco, Natalie Pott-Schmidt, Lisa Prock, Angela Rachubinski, Desiree Rambeck, Rishiraj Rana, Shelley Randall, VaikuntRanganathan, Ashley Raven, Madelyn Rayos, Kelli Real, Richard Remington, Anna Rhea, Catherine Rice, Harper Richardson, Stacy Riffle, Chris Rigby, Ben Right, Beverly Robertson, Erin Roby, Casey Roche, Nicki Rodriguez, Katherine Roeder, Jou Roger, Daniela Rojas, Cordelia Rosenberg, Jacob Rosewater, Katelyn (Neely) Rossow, Payton Runyan, Nicole Russo, Tara Rutter, Mustafa Sahin, Marina Sarris, Dustin Sarver, Madeline Savage, Jessica Scherr, Hayley Schools, Gregory Schoonover, Robert Schultz, Brady Schwind, Cheyanne Sebolt, Rebecca Shaffer, Swapnil Shah, Neelay Shah, Roman Shikov, MojeebShir, Amanda Shocklee, Clara Shrier, Lisa Shulman, Matt Siegel, Andrea Simon, Laura Simon, Kaitlyn Singer, Emily Singer, Vini Singh, Kaitlin Smith, Chris Smith, Ashlyn Smith, LeeAnne Green-Snyder, LathaSoorya, DikshaSrishyla, Danielle Stamps, Laura Stchur, Morgan Steele, Alexandra Stephens, Catherine Sullivan, Amy Swanson, Megan Sweeney, Anthony Sziklay, MairaTafolla, Nicole Takahashi, Amber Tallbull, Nicole Targalia, Cora Taylor, Sydney Terroso, Angela Tesng, Samantha Thompson, Jennifer Tjernagel, Jaimie Toroney, Laina Townsend, Katherine Tsai, Ivy Tso, Maria Valicenti-Mcdermott, Bonnie VanMetre, Candace VanWade, Dennis Vasquez Montes, Alison Vehorn, Mary Verdi, Brianna Vernoia, Natalia Volfovsky, Lakshmi Vrittamani, Jermel Wallace, Corrie Walston, Audrey Ward, Zachary Warren, William Curtis Weaver, Sabrina White, Casey White-Lehman, Fiona Winoto, Ericka Wodka, Jessica Wright, Sabrina Xiao, Simon Xu, WhaJames yang, Amy yang, Meredith yinger, Christopher Zaro, Hana Zaydens, Cindy Zha, Allyson Zick

SPARK Consortium Sites: Baylor College of Medicine/Texas Children’s Hospital; Boston Children’s Hospital; Center for Autism and the Developing Brain/New york-Presbyterian; Children’s Hospital of Philadelphia; Children’s Specialized Hospital; Cincinnati Children’s Hospital Medical Center; Emory university and Marcus Autism Center; Geisinger Autism and Developmental Medicine Institute; Kennedy Krieger Institute; Maine Medical Center Research Institute; Medical university of South Carolina; Nationwide Children’s Hospital; Oregon Health and Science university; The Rose F. Kennedy Children’s Evaluation and Rehabilitation Center at Montefiore; Rush university Medical Center; Seattle Children’s Hospital; Stanford university; university of California, Davis; university of California Los Angeles; university of California, San Diego; Southwest Autism Research and Resource Center; university of Colorado; university of Iowa; university of Miami; university of Michigan; university of Minnesota, Twin Cities; university of Mississippi; university of Missouri; university of North Carolina, Chapel Hill; university of utah; Vanderbilt university Medical Center; yale university

Disclosure statement

SPARK: Simons Foundation Powering Autism Research for Knowledge, Simons Foundation (SPARK #534041, ELW).

Funding

This work was supported by SFARI (SPARK #534041, ELW).

References

  1. American Psychiatric Association. (2013). Anxiety disorders. In Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: American Psychiatric Publishing. 10.1176/appi.books.9780890425596.dsm05 [DOI] [Google Scholar]
  2. Antezana L, Factor RS, Condy EE, Strege MV, Scarpa A, & Richey JA (2019). Gender differences in restricted and repetitive behaviors and interests in youth with autism. Autism Research: Official Journal of the International Society for Autism Research, 12(2), 274–283. Feb. 10.1002/aur.2049 Epub 2018 Dec 17. PMID: 30561911. [DOI] [PubMed] [Google Scholar]
  3. Avni E, Ben-Itzchak E, & Zachor DA (2018). The Presence of Comorbid ADHD and anxiety symptoms in autism spectrum disorder: Clinical presentation and predictors. Frontiers in Psychiatry, 9, 717. 10.3389/fpsyt.2018.00717 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Baio J (2012). Prevalence of autism spectrum disorders: Autism and developmental disabilities monitoring network, 14 sites, united States, 2008. In Autism Spectrum Disorders: Guidance, Research and Federal Activity. [Google Scholar]
  5. Berenguer C, Roselló B, Colomer C, Baixauli I, & Miranda A (2018). Children with autism and attention deficit hyperactivity disorder. Relationships between symptoms and executive function, theory of mind, and behavioral problems. Research in Developmental Disabilities, 83, 260–269. 10.1016/j.ridd.2018.10.001 [DOI] [PubMed] [Google Scholar]
  6. Berument SK, Rutter M, Lord C, Pickles A, & Bailey A (1999). Autism screening questionnaire: Diagnostic validity. British Journal of Psychiatry, 175(5), 444–451. 10.1192/bjp.175.5.444 [DOI] [PubMed] [Google Scholar]
  7. Chartrand H, Sareen J, Toews M, & Bolton JM (2012). Suicide attempts versus nonsuicidal self-injury among individuals with anxiety disorders in a nationally representative sample. Depression and Anxiety, 29(3), 172–179. 10.1002/da.20882 Epub 2011 Sep 21. PMID: 21948315. [DOI] [PubMed] [Google Scholar]
  8. Cohen J (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Erlbaum. [Google Scholar]
  9. Craig F, Margari F, Legrottaglie AR, Palumbi R, de Giambattista C, & Margari L (2016). A review of executive function deficits in autism spectrum disorder and attention-deficit/hyperactivity disorder. Neuropsychiatric Disease and Treatment, 12, 1191–1202. 10.2147/NDT.S104620 PMID: 27274255; PMCID: PMC4869784. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Daniels AM, Rosenberg RE, Anderson C, Law JK, Marvin AR, & Law PA (2012). Verification of parent-report of child autism spectrum disorder diagnosis to a web-based autism registry. Journal of Autism and Developmental Disorders, 42(2), 257–265. 10.1007/s10803-011-1236-7 [DOI] [PubMed] [Google Scholar]
  11. Evans SC, Boan AD, Bradley C, & Carpenter LA (2019). Sex/gender differences in screening for autism spectrum disorder: Implications for evidence-based assessment. Journal of Clinical Child and Adolescent Psychology: The Official Journal for the Society of Clinical Child and Adolescent Psychology, American Psychological Association, Division 53, 48(6), 840–854. 10.1080/15374416.2018.1437734 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Feliciano P, Daniels AM, Green Snyder L, Beaumont A, Camba A, Esler A, Gulsrud AG, Mason A, Gutierrez A, Nicholson A, Paolicelli AM, McKenzie AP, Rachubinski AL, Stephens AN, Simon AR, Stedman A, Shocklee AD, Swanson A, Finucane B, … Chung WK (2018). SPARK: A uS Cohort of 50,000 Families to Accelerate Autism Research. Neuron, 97(3), 488–493. 10.1016/j.neuron.2018.01.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Goldstein S, & Schwebach AJ (2004). The comorbidity of pervasive developmental disorder and attention deficit hyperactivity disorder: results of a retrospective chart review. Journal of Autism and Developmental Disorders, 34(3), 329–339. 10.1023/B:JADD.0000029554.46570.68 [DOI] [PubMed] [Google Scholar]
  14. Hull L, Petrides KV, & Mandy W (2020). The female autism phenotype and camouflaging: a narrative review. Review Journal of Autism and Developmental Disorders, 7(4), 306–317. 10.1007/s40489-020-00197-9 [DOI] [Google Scholar]
  15. Hus V, Taylor A, & Lord C (2011). Telescoping of caregiver report on the Autism Diagnostic Interview-Revised. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 52(7), 753–760. 10.1111/j.1469-7610.2011.02398.x Epub 2011 Mar 16. PMID: 21410473; PMCID: PMC3549439. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kaat AJ, Shui AM, Ghods SS, Farmer CA, Esler AN, Thurm A, Georgiades S, Kanne SM, Lord C, Kim y. S., & Bishop SL (2020). Sex differences in scores on standardized measures of autism symptoms: a multisite integrative data analysis. Journal of Child Psychology and Psychiatry, 62, 97–106. 10.1111/jcpp.13242 Epub ahead of print. PMID: 32314393. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Kerns CM, Kendall PC, Zickgraf H, Franklin ME, Miller J, & Herrington J (2015). Not to be overshadowed or overlooked: functional impairments associated with comorbid anxiety disorders in youth with ASD. Behavior Therapy, 46(1), 29–39. 10.1016/j.beth.2014.03.005 [DOI] [PubMed] [Google Scholar]
  18. Keyes KM, Susser E, Cheslack-Postava K, Fountain C, Liu K, & Bearman PS (2012). Cohort effects explain the increase in autism diagnosis among children born from 1992 to 2003 in California. International Journal of Epidemiology, 41(2), 495–503. 10.1093/ije/dyr193 Epub 2011 Dec 7. PMID: 22253308; PMCID: PMC3324454. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Lai MC, & Szatmari P (2020). Sex and gender impacts on the behavioural presentation and recognition of autism. Current Opinion in Psychiatry, 33(2), 117–123. 10.1097/yCO.0000000000000575 PMID: 31815760. [DOI] [PubMed] [Google Scholar]
  20. Lam KS, & Aman MG (2007). The Repetitive Behavior Scale-Revised: independent validation in individuals with autism spectrum disorders. Journal of Autism and Developmental Disorders, 37(5), 855–866. 10.1007/s10803-006-0213-z PMID: 17048092. [DOI] [PubMed] [Google Scholar]
  21. Larkin F, Meins E, Centifanti LCM, Fernyhough C, & Leekam SR (2017). How does restricted and repetitive behavior relate to language and cognition in typical development? Development and Psychopathology, 29(3), 863–874. 10.1017/S0954579416000535 Epub 2016 Jul 15. PMID: 27417193. [DOI] [PubMed] [Google Scholar]
  22. Lee DO, & Ousley O. y. (2006). Attention-deficit hyperactivity disorder symptoms in a clinic sample of children and adolescents with pervasive developmental disorders. Journal of Child and Adolescent Psychopharmacology, 16(6), 737–746. 10.1089/cap.2006.16.737 [DOI] [PubMed] [Google Scholar]
  23. Lenth R (2020). emmeans: Estimated Marginal Means, aka Least-Squares Means. R package version 1.4.6. https://CRAN.R-project.org/package=emmeans [Google Scholar]
  24. Leyfer OT, Folstein SE, Bacalman S, Davis NO, Dinh E, Morgan J, Tager-Flusberg H, & Lainhart JE (2006). Comorbid psychiatric disorders in children with autism: interview development and rates of disorders. Journal of Autism and Developmental Disorders, 36(7), 849–861. 10.1007/s10803-006-0123-0 [DOI] [PubMed] [Google Scholar]
  25. Loomes R, Hull L, Polmear W, & Mandy L (2017). What is the male-to-female ratio in autism spectrum disorder? a systematic review and meta-analysis. Journal of the American Academy of Child and Adolescent Psychiatry, 56(6), 466–474. 10.1016/j.jaac.2017.03.013 [DOI] [PubMed] [Google Scholar]
  26. Lord C, Rutter M, DiLavore P, Risi S, Gotham K, & Bishop S (2012). Autism Diagnostic Observation Schedule Second Edition (ADOS-2) Manual (Part 1): Modules (pp. 1–4). Western Psychological Services. [Google Scholar]
  27. Lord C, Rutter M, & Le Couteur A (1994). Autism Diagnostic Interview-revised: A revised version of a diagnostic interview for caregivers of individuals with possible pervasive developmental disorders. Journal of Autism and Developmental Disorders, 24(5), 659–685. 10.1007/BF02172145 [DOI] [PubMed] [Google Scholar]
  28. Mahone EM, & Denckla MB (2017). Attention-deficit/hyperactivity disorder: A historical neuropsychological perspective. Journal of the International Neuropsychological Society: JINS, 23(9–10), 916–929. 10.1017/S1355617717000807 PMID: 29198277; PMCID: PMC5724393. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Matson JL, & Nebel-Schwalm MS (2007). Comorbid psychopathology with autism spectrum disorder in children: an overview. Research in Developmental Disabilities, 28(4), 341. 10.1016/j.ridd.2005.12.004 [DOI] [PubMed] [Google Scholar]
  30. Mazurek MO, & Petroski GF (2015). Sleep problems in children with autism spectrum disorder: examining the contributions of sensory over-responsivity and anxiety. Sleep Medicine, 16(2), 270–279. 10.1016/j.sleep.2014.11.006 [DOI] [PubMed] [Google Scholar]
  31. McMakin DL, & Alfano CA (2015). Sleep and anxiety in late childhood and early adolescence. Current Opinion in Psychiatry, 28(6), 483–489. 10.1097/yCO.0000000000000204 PMID: 26382163; PMCID: PMC4670558. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Meir P, Alfano CA, Lau S, Hill RM, & Palmer CA (2019). Sleep patterns and anxiety in children interact to predict later suicidal ideation. Children’s Health Care: Journal of the Association for the Care of Children’s Health, 48(4), 372–393. 10.1080/02739615.2019.1630283 Epub 2019 Sep 25. PMID: 32675893; PMCID: PMC7365578. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Ozonoff S, Li D, Deprey L, Hanzel EP, & Iosif AM (2018). Reliability of parent recall of symptom onset and timing in autism spectrum disorder. Autism: The International Journal of Research and Practice, 22(7), 891–896. 10.1177/1362361317710798 Epub 2017 Sep 13. PMID: 28903580; PMCID: PMC5832544. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Pitzianti M, D’Agati E, Pontis M, Baratta A, Casarelli L, Spiridigliozzi S, Curatolo P, & Pasini A (2016). Comorbidity of ADHD and high-functioning autism: A pilot study on the utility of the overflow movements measure. Journal of Psychiatric Practice, 22(1), 22–30. 10.1097/PRA.0000000000000120 [DOI] [PubMed] [Google Scholar]
  35. R Core Team. (2020). R: A language and environment for statistical computing. Austria: R Foundation for Statistical Computing. https://www.R-project.org/. [Google Scholar]
  36. Rivard L, Missiuna C, McCauley D, & Cairney J (2014). Descriptive and factor analysis of the Developmental Coordination Disorder Questionnaire (DCDQ’07) in a population-based sample of children with and without Developmental Coordination Disorder. Child: care, Health and Development, 40(1), 42–49. 10.1111/j.1365-2214.2012.01425.x Epub 2012 Aug 29. PMID: 229282. [DOI] [PubMed] [Google Scholar]
  37. Rutter M, Bailey A, & Lord C (2007). The Social Communication Questionnaire (SCQ), second printing. Western Psychological Services. [Google Scholar]
  38. Simonoff E, Pickles A, Charman T, Chandler S, Loucas T, & Baird G (2008). Psychiatric disorders in children with autism spectrum disorders: prevalence, comorbidity, and associated factors in a population-derived sample. Journal of the American Academy of Child and Adolescent Psychiatry, 47(8), 921–992. 10.1097/CHI.0b013e318179964f [DOI] [PubMed] [Google Scholar]
  39. Tregnago MK, & Cheak-Zamora NC (2012). Systematic review of disparities in health care for individuals with autism spectrum disorders in the united States. Research in Autism Spectrum Disorders, 6(3), 1023–1031. 10.1016/j.rasd.2012.01.005 [DOI] [Google Scholar]
  40. White SW, Oswald D, Ollendick T, & Scahill L (2009). Anxiety in children and adolescents with autism spectrum disorders. Clinical Psychology Review, 29(3), 216–229. 10.1016/j.cpr.2009.01.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Wilson BN, Crawford SG, Green D, Roberts G, Aylott A, & Kaplan B (2009). Psychometric properties of the revised developmental coordination disorder questionnaire. Physical & Occupational Therapy in Pediatrics, 29(2), 182–202. 10.1080/01942630902784761 [DOI] [PubMed] [Google Scholar]
  42. Wodka EL, Simmonds DJ, Mahone EM, & Mostofsky SH (2009). Moderate variability in stimulus presentation improves motor response control. Journal of Clinical and Experimental Neuropsychology, 31(4), 483–488. 10.1080/13803390802272036 PMID: 18686112; PMCID: PMC2892264. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES