Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Oct 1.
Published in final edited form as: J Autism Dev Disord. 2023 Aug 16;54(10):3777–3791. doi: 10.1007/s10803-023-06104-5

Autism Prevalence and the Intersectionality of Assigned Sex at Birth, Race, and Ethnicity on Age of Diagnosis

Jessica E Goldblum 1, Tyler C McFayden 2, Stephanie Bristol 3, Orla C Putnam 3, Amanda Wylie 1, Clare Harrop 3,4
PMCID: PMC10869641  NIHMSID: NIHMS1952092  PMID: 37584770

Abstract

Purpose

An official autism diagnosis is required to access timely intervention and is associated with better long-term well-being and mental health. Certain demographic characteristics, such as being female or a racially or ethnically minoritized youth, have been associated with significant diagnostic lag. However, it remains unclear how assigned sex, race, and ethnicity interact with each other in predicting the prevalence and age of autism diagnosis.

Methods

To examine the interactions between assigned sex, race, and ethnicity, we used data from the National Survey of Children’s Health (NSCH; 2016 > 2021).

Results

One in 38 children had an autism diagnosis and 3.8 males were diagnosed per 1 female. Hierarchical linear regressions yielded diagnostic delays in some females, particularly those who were non-Hispanic white, Black, and Asian. Ethnic and racial minority children had significantly earlier diagnoses than white and non-Hispanic children when not accounting for sex.

Conclusion

This study demonstrates slight increases in reported autism prevalence, a diagnostic lag in some autistic females that was strongly associated with ethnicity, and earlier diagnoses in racial and ethnic minority youth, a finding that may be explained by factors associated with phenotypic differences. This study has important implications for the diagnosis of minority autistic youth, particularly females and females who are non-Hispanic, who may experience a greater propensity for diagnostic delays.

Keywords: Autism spectrum disorder, Autism prevalence, Sex ratio, Diagnosis, Demographic factors

Recent population estimates in the United States (US) indicate that 1 in 36 children have an autism diagnosis (Maenner et al., 2023). However, variations in prevalence and age of diagnosis have been reported based on demographic characteristics such as race and ethnicity (Wiggins et al., 2020), socioeconomic status (SES) (Durkin et al., 2010), and assigned sex at birth (hereafter, sex) (Li et al., 2022). The goal of this study was to explore the interactive effects of demographic variables, including sex, race, and ethnicity, on the prevalence and age of an autism diagnosis.

Although obtaining an autism diagnosis can be a highly complex and years-long process for many, some children can achieve a reliable and stable diagnosis as early as 18 months (Hyman et al., 2020). However, the median diagnostic age remains four years (Maenner et al., 2023) and research suggests that females1 (Harrop et al., 2021), those on Medicaid (Mandell et al., 2010), and minoritized racial/ethnic groups experience pertinent diagnostic delays (Mandell et al., 2002, 2007). The “white privilege problem” and male bias of autism have been highlighted in clinical diagnoses and research (Onaiwu, 2020; Strang et al., 2020), demonstrating the under, missed, and late diagnosis of females and people of color relative to non-Hispanic, white, and male children.

Research suggests that four males are diagnosed with autism per one female (Fombonne, 2009), although depending on methodology and sample, some estimates are closer to 3:1 (Kogan et al., 2018; Loomes et al., 2017). Females are typically diagnosed 12 to 24 months later than males, particularly when they have more advanced language abilities (Harrop et al., 2021). Additionally, females with fewer support needs are often missed in the diagnostic process entirely (Lockwood Estrin et al., 2021), which may be linked to a genetic female protective effect (Zhang et al., 2020). This contrasts with identified females who tend to have more co-occurring conditions and greater support needs (Volkmar et al., 1993). Autistic females and their caregivers show frustration in access to diagnosis (Milner et al., 2019) and though clinician and teacher knowledge of how autism presents in females is improving, a male bias remains (Whitlock et al., 2020).

Historically, racial and ethnic minority children are less likely to be identified with autism and are more likely to experiences diagnostic delays (Liptak et al., 2008; Mandell et al., 2002, 2007). Prevalence studies from the last decade, however, consistently find racial/ethnic minority children to have heightened prevalences and earlier age at diagnosis (Emerson et al., 2016; Martin et al., 2022). Wallis et al. (2023) examined primary care data from the 2011–2015 Children’s Hospital of Philadelphia Care Network and found a higher autism prevalence in Asian and Black children, and those with higher socioeconomic risk and who received primary care at urban sites. A study using National Survey of Children’s Health data found that Black and Hispanic/Latinx children were more likely to have earlier diagnoses than white, non-Hispanic children (Jo et al., 2015). Separate analyses of the same data extended this finding to Multiracial children, suggesting a potential shift in the impact of race and ethnicity on age of diagnosis (Emerson et al., 2016).

In earlier research, lower prevalence coupled with older age of diagnosis in racial/ethnic minority children may be linked to systemic-level factors such as clinician biases when delivering an autism diagnosis. For example, Black children are 2.4 times more likely to be misdiagnosed with conduct disorder and 5.1 times more likely to be misdiagnosed with adjustment disorder prior to receiving an autism diagnosis, compared to white autistic children who are more likely to be misdiagnosed with attention-deficit/hyperactivity disorder (ADHD) (Mandell et al., 2007). Similarly, older age at diagnosis and lower prevalence rates in racial/ethnic minority children may be explained by child-level factors. For example, when racial/ethnic minority children are diagnosed as autistic, they disproportionally fall into the “severe” end of the spectrum (Jo et al., 2015), suggesting that prevalence rates of minoritized diagnosed youth may only reflect those with higher support needs. Interestingly, parents of Black and Hispanic/Latinx children report fewer autism-specific concerns than parents of white and non-Hispanic children(Blacher et al., 2019; Donohue et al., 2019; Ratto et al., 2016) despite their children having greater clinician-reported severity ratings (Blacher et al., 2019). Therefore, discrepancies between clinician and caregiver concerns could be another indicator of clinician biases within the diagnostic process.

Although relationships between demographic factors and autism prevalence and age of diagnosis remain unclear, it is evident that minority children face significant disparities and engrained biases when navigating the diagnostic process. Hispanic/Latinx and Black caregivers report frustration and dissatisfaction in service access, including that practitioners spend less time with their children and that their concerns had not been heard (Magaña et al., 2012) with these trends persisting five years later (Magaña et al., 2015). Families of Black autistic children report consistent difficulties utilizing early intervention services due to issues such as work schedules and having to “aggressively advocate” for their child (Pearson & Meadan, 2018). Parents of Black children also report significant delays in the timing when first sharing their concerns with providers to an official autism diagnosis (Constantino et al., 2020). Taken together, these findings suggest a racially biased or culturally insensitive model of diagnosing autism, or at least significant discrepancies between clinician and parent experiences of the same child.

Diagnostic delays and misdiagnoses postpone the start of early intervention services, which are frequently associated with better adaptive behavioral, cognitive, educational, verbal, and symptomatic trajectories and the need for fewer ongoing supports (Anderson et al., 2014; Estes et al., 2015). Delays are especially problematic for females and racial/ethnic minority children who have a higher likelihood of experiencing poor mental health and well-being (Eilenberg et al., 2019; Solomon et al., 2012; South et al., 2020). Thus, mitigating diagnostic disparities is essential to improve long-term wellbeing and overall quality of life in these vulnerable groups.

Current Study

Funded by the Maternal and Child Health Bureau, the National Survey of Children’s Health (NSCH) provides data across multiple aspects of health at national levels for US children (Child and Adolescent Health Measurement Initiative, 2016). NSCH data have been used previously to study autism prevalence, diagnostic disparities, and service access by race, ethnicity, and income (Kogan et al., 2018; Kogan et al., 2018; Liptak et al., 2008). NSCH analyses of data collected in 2016 reported a 2.5% population prevalence of autism not differing by race or ethnicity, a sex ratio of 3.4:1 (male: female), and higher difficulty accessing healthcare services in autistic children compared to children with other disabilities (Kogan et al., 2018). Analyses of prior waves of NSCH data point to longstanding disparities in service access in Hispanic/Latinx, Black, and low SES children (Liptak et al., 2008) and varying autism prevalence by race, ethnicity, and SES (Kogan et al., 2018). To our knowledge, research using NSCH has yet to examine the intersectionality of race, ethnicity, and sex on autism prevalence and age of diagnosis. Moreover, no study to date has utilized more than two consecutive years of NSCH data to examine age of diagnosis.

Using 2016 to 2021 NSCH data, the goals of this study were twofold. First, we aimed to examine autism prevalence according to sex, race, and ethnicity. Second, we sought to characterize differences in age of diagnosis as a function of sex, race, and ethnicity. We predicted an autism prevalence of 2.27%, or 1 in 44 children and a 3.4:1 (male: female) sex ratio consistent with prevalence and sex ratio findings by Maenner et al. (2021) and sex ratio findings by Kogan et al. (2018). Second, we predicted that sex, race, and ethnicity would predict age of diagnosis, with racial/ethnic minority and female children diagnosed later than white, non-Hispanic, and male children. We predicted that multiply minoritized children (e.g., Black Hispanic females) would be diagnosed later than racial-, ethnic-, and sex-minority children alone.

Method

Data Source

This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines(Von Elm et al., 2007) and was approved for exemption by the (masked university) Institutional Review Board. Data for this study were the 2016–20212 NSCH. NSCH data is adjusted and weighted to represent population estimates of all national US resident children under age 18. Respondents were parents/guardians knowledgeable about their child’s health/healthcare needs. Respondents completed a screener survey about the number of children in the household and one child was randomly selected. For more information about survey weights, see https://www.childhealthdata.org/learn-about-the-nsch/FAQ.

Participants

There were N = 218,456 respondents from 2016 to 2021 with n = 5,952 autistic children (see Dependent Variables). Children were on average 10.84 years (SD = 4.40). The majority of the sample was white (76.26%) and had household incomes of < 100% of the federal poverty level (FPL) (31.50%). See Table 1 for participant characteristics.

Table 1.

Participant Characteristics

Mean (SD) Weighted Median
Diagnostic age in years (N = 5952) 5.09 (3.35) 4.00
Child’s age in years 10.84 (4.40) 11.00
N (%) Weighted %
ID No ID 4955 (83.67) 82.02%
ID 967 (16.33) 17.98%
Missing 30
Home English 5615 (94.66) 88.38%
Language Another Language 317 (5.34) 11.62%
Missing 20
Caregiver <High School 145 (2.45) 9.33%
Education High school/GED 922 (15.55) 23.64%
Some college 1650 (27.83) 25.51%
College+ 3212 (54.17) 41.52%
Missing 23
Caregiver No employment 155 (2.60) 2.42%
employment 1 + employed 5797 (97.40) 97.58%
Insurance Insurance gaps / No 276 (4.66) 5.62%
Coverage Insurance
Full coverage 5644 (95.34) 94.38%
Missing 32
Family Pov- < 100% of FPL 174 (15.12) 31.50%
erty Level 100–200% of FPL 215 (18.68) 20.12%
200–400% of FPL 339 (29.45) 23.84%
> 400% of FPL 423 (36.75) 24.54%
Missing 4801
Open in a new tab

Independent Variables

Independent variables were sex, ethnicity, and race (See Supplemental Table 1 for in-depth descriptions). Respondents selected male or female for sex and Hispanic/Latino Origin (hereafter, Hispanic/Latinx) or Not Hispanic/Latino Origin (here-after, non-Hispanic) for ethnicity. Respondents selected white alone, Black/African American alone (here-after, Black), American Indian/Alaska Native alone (hereafter, AIAN), Asian alone, Native Hawaiian/Other Pacific Islander alone (hereafter, NHOPI), Some Other Race Alone, or Two or More Races (hereafter, Multiracial) for race. American Indian/Alaska Native alone, NHOPI, and Some Other Race Alone were removed from age of diagnosis analyses due to too few cases. Therefore, age of diagnosis analyses included white, Black, Asian, and Multiracial children. See Table 2 for a breakdown of autistic children by sex, race, and ethnicity.

Table 2.

Breakdown of Autistic Children by Sex, Race, and Ethnicity

Autistic Children
Male Female
Non-Hispanic Hispanic Non-Hispanic Hispanic Total
White 3175 437 807 120 4539
Black 357 25 72 8 462
Asian 192 6 52 4 254
AIAN 27 12 11 4 54
NHOPI 7 17 3 4 31
Other 6 36 5 14 61
Multiracial 358 70 99 24 551
Ethnicity/Sex Total 4122 603 1049 178 5952
Sex Total 4725 1227 5952
Open in a new tab

Note. AIAN = American Indian/Alaska Native; NHOPI = Native Hawaiian and other Pacific Islander; Other = Some other race alone

To further characterize study children, we used the following NSCH variables as covariates (See Supplemental Table 2 for in-depth descriptions): age in years (centered), intellectual disability (ID), and primary household language (hereafter, home language), with possible answers of English, Spanish, or other. Modeled after SES covariates in Akobirshoev et al. (2020), SES characteristics for this study were: (1) insurance coverage, dichotomized as full coverage (full coverage for the past 12 months) or insurance gaps (no coverage and/or gaps in coverage in the past 12 months); (2) highest education level of adults in the home (hereafter, caregiver education), derived from the highest level of education of adults with possible answers of: <high school; high school or GED; some college or associate degree; or college degree>; (3) poverty status (hereafter, FPL), a variable derived from the family poverty ratio and six NSCH FPL implicates created via sequential regression using SAS-callable SUDAAN software version 9.4 (SAS Institute Inc. 2016), with possible answers of < 100% of FPL, 100–200% of FPL, 200–400% of the FPL, or > 400% of FPL); and (4) caregiver employment, dichotomized as yes (caregiver employment for at least 50 of the past 52 weeks) or no (caregiver employment for < 50 of the past 52 weeks).

Dependent Variables

Dependent variables were autism prevalence and age of diagnosis (See Supplemental Table 3). Respondents answered, “Has a doctor or other health care provider EVER told you that this child has…Autism or autism spectrum disorder (ASD)? Include diagnoses of Asperger’s Disorder or Pervasive Developmental Disorder (PDD)?” (K2Q35A) and, “If yes (to K2Q35A), does this child CURRENTLY have the condition?” (K2Q35B). Age of diagnosis (years) was obtained with numeric answers to, “How old was this child when a doctor or other health care provider FIRST told you that they had Autism, ASD, Asperger’s Disorder, or PDD?” (K2Q35A_1_YEARS).

Analytic Plan

Analyses were completed in SAS version 9.4 (SAS Institute Inc., 2016). Descriptive statistics were estimated for predictors and covariates. Autism prevalence was calculated as the number of children with a current autism diagnosis (“Yes” to K2Q35B) divided by the number with a reported lifetime diagnosis (“Yes” or “No” to K2Q35A; N = 101,851). We used 95% Wald confidence intervals (CI) to examine prevalence within the total sample and by race, sex, and ethnicity. Chi Square tests were used to examine differences in prevalence by sex and ethnicity and logistic regressions were used to examine differences by race. Except age of diagnosis, sample characteristics and prevalences are reported as weighted averages, medians, or percentages.

Hierarchical linear regressions were used to examine the variance in diagnostic age, controlling for covariates (Model 1 & 2) and including two-way (Model 3) and three-way (Model 4) interactions. Model 3 had all possible two-way interactions between sexes, races, and ethnicities, although the Asian*Hispanic interaction was not included due to too few cases (n = 10). We included the Black*Hispanic interaction exploratorily, although results should be interpreted with caution as only 27 children were Black and Hispanic. For Model 4, the Black*Hispanic*Female interaction (n = 5) and the Asian*Hispanic*Female interaction (n = 3) were omitted, so only the Multiracial*Hispanic*Female interaction was included exploratorily (n = 23), although results should be interpreted with caution. We report results according to the fourth/final model. Significant interactions (α = 0.05) were probed using simple slopes (Dawson, 2014). As FPL was imputed across six datasets, regression models were estimated in each dataset with estimates pooled across datasets with PROC MI ANALYZE. We report unstandardized beta estimates, standard errors, t values, adjusted R squares, and p-values per model. Post hoc exploratory Chi Square analyses were run to further understand intersectionality results.

Results

Sample Characteristics and Prevalence Rates

Of the N = 5,952 respondents who reported they had a child with an autism diagnosis, n = 1,227 (21.73%) were female. Children were majority white (64.47%) with smaller percentages of Black (15.55%) and Multiracial (4.68%) children (see Table 2). Children were 28.64% Hispanic/Latinx. The overall median age of autism diagnosis and median age within sex was 4.00 years (overall SD = 3.35; female SD = 3.76; male SD = 3.22). Non-Hispanic children had a median age of diagnosis of 4 years (SD = 3.39) and Hispanic children, 3 years (SD = 3.00). The median age of diagnosis by racial groups were: AIAN: 5 years (SD = 3.45); Multiracial: 4 years (SD = 3.27); white: 4 years (SD = 3.41); Asian: 3 (SD = 3.04); Black: 3 (SD = 2.78); NHOPI: 3 years (SD = 2.67); and Some Other Race Alone: 3 years (SD = 2.92).

Autism prevalence was 2.65% or 1 in 38 children. For males, prevalence was 3.84%, 95% CI [3.83, 3.84]3 and females, 1.11%, 95% CI [1.11, 1.12], resulting in a 3.8:1 (male: female) ratio. Autism prevalences by race were: Some Other Race Alone: 3.90%, 95% CI [3.87, 3.92]; NHOPI: 3.00%, 95% CI [2.97, 3.04]; Multiracial: 2.82%, 95% CI [2.80, 2.83]; Black: 2.80%, 95% CI [2.79, 2.81]; white: 2.41%, 95% CI [2.41, 2.42]; Asian: 1.65%, 95% CI [1.64, 1.66]; and AIAN: 1.50%, 95% CI [1.47, 1.52]. Prevalence was 2.84%, 95% CI [2.84, 2.85] in Hispanic/Latinx and 2.39%, 95% CI [2.38, 2.39] in non-Hispanic children. Prevalences significantly differed by ethnicity, X2(1) = 11570.86, p < .0001, (Hispanic/Latinx > non-Hispanic); sex, X2(1) = 549998.00, p < .0001 (Male > Female); and race: Some Other Race Alone > NHOPI > Multiracial = Black > white > Asian > AIAN (p < .0001).

Diagnostic Age

To determine how demographic variables predicted age of diagnosis, we ran one hierarchical regression with four levels, the first with covariates, the second adding focal predictors, the third, two-way interactions, and the fourth, a three-way interaction. Model 4 had n = 5427 observations and accounted for 25.3% of the variance in diagnostic age (Adj R2 = 0.25).

Significant main effects were observed for children who were female, β = 1.00, p < .001, Hispanic, β= −0.43, p < .001, Black, β= −0.33, p = .011 and Asian, β= −1.20, p < .001, older at the time of survey completion, β = 0.33, p < .001, and children who had ID, β= −0.97, p < .001, a non-English home language, β = 0.36, p = .007, caregivers who were unemployed, β= −1.00, p < .001, and insurance gaps, β = 1.54, p < .001 (Table 3). We translate these parameter estimates accordingly: Hispanic children were diagnosed 5.17 months earlier than white children. Black children were diagnosed nearly 4 months earlier than white children, and Asian children were diagnosed over 14 months earlier than white children. Females were diagnosed one year later than males. Children with ID were diagnosed nearly one year earlier than children without ID and children whose caregivers were unemployed were diagnosed a year earlier than children with at least one employed caregiver. Older children (at the time of survey completion) were diagnosed 4 months later; children with insurance gaps were diagnosed over 18 months later than children with full coverage; and children with a non-English home language, over 4 months later than children who spoke English.

Table 3.

Hierarchical Regression Models with Outcome Age of Autism Diagnosis

Model β SE t p R2
1 Constant 1.56 0.13 11.65 < 0.001 0.22
Older Age 0.33 0.00 36.70 < 0.001
Intellectual Disability −0.97 0.10 −9.48 < 0.001
Non-English Home Language −0.04 0.12 −0.34 0.734
Some College −0.09 0.11 −0.85 0.397
High School 0.11 0.11 1.01 0.311
Less than High School 0.01 0.17 0.07 0.943
Both Parents Unemployed −1.03 0.25 −4.02 < 0.001
No Insurance / Gaps in Coverage 1.48 0.17 8.79 < 0.001
200 < FPL < 400 −0.05 0.14 −0.34 0.737
100 < FPL < 200 −0.15 0.14 −1.11 0.269
FPL < 100 −0.14 0.16 −0.89 0.380
2 Constant 1.67 0.13 12.07 < 0.001 0.24
Older Age 0.33 0.00 36.57 < 0.001
Intellectual Disability −0.92 0.10 −9.06 < 0.001
Non-English Home Language 0.30 0.13 2.22 0.026
Some College −0.07 0.10 −0.71 0.478
High School 0.12 0.11 1.06 0.288
Less than High School −0.01 0.17 −0.11 0.910
Both Parents Unemployed −1.07 0.25 −4.23 < 0.001
No Insurance / Gaps in Coverage 1.59 0.16 9.48 < 0.001
200 < FPL < 400 0.00 0.14 0.01 0.990
100 < FPL < 200 −0.07 0.14 −0.56 0.577
FPL < 100 −0.00 0.15 −0.02 0.981
Female 0.66 0.09 7.12 < 0.001
Black −0.38 0.11 −3.38 < 0.001
Asian −0.96 0.22 −4.29 < 0.001
Multiracial −0.43 0.13 −3.20 0.001
Hispanic/Latinx −0.72 0.10 −7.21 < 0.001
3 Constant 1.62 0.14 11.55 < 0.001 0.25
Older Age 0.33 0.00 36.5 < 0.001
Intellectual Disability −0.94 0.10 −9.22 < 0.001
Non-English Home Language 0.36 0.13 2.60 0.009
Some College −0.07 0.10 −0.68 0.497
High School 0.10 0.11 0.93 0.354
Less than High School 0.03 0.17 0.21 0.837
Both Parents Unemployed −0.98 0.25 −3.87 < 0.001
No Insurance / Gaps in Coverage 1.54 0.16 9.20 < 0.001
200 < FPL < 400 0.01 0.14 0.07 0.941
100 < FPL < 200 −0.06 0.14 −0.48 0.635
FPL < 100 −0.03 0.15 −0.25 0.804
Female 0.90 0.13 6.95 < 0.001
Black −0.35 0.13 −2.71 0.006
Asian −1.23 0.26 −4.60 < 0.001
Multiracial −0.44 0.19 −2.30 0.021
Hispanic/Latinx −0.49 0.12 −4.08 < 0.001
Female * Black −0.20 0.28 −0.71 0.479
Female * Asian 0.83 0.47 1.77 0.077
Female * Multiracial 0.56 0.33 1.69 0.090
Hispanic/Latinx * Black 0.34 0.47 0.73 0.468
Hispanic/Latinx * Multiracial −0.22 0.27 −0.83 0.407
Female * Hispanic/Latinx −1.00 0.21 −4.74 < 0.001
4 Constant 1.57 0.14 11.13 < 0.001 0.25
Older Age 0.33 0.00 36.73 < 0.001
Intellectual Disability −0.97 0.10 −9.50 < 0.001
Non-English Home Language 0.36 0.13 2.67 0.007
Some College −0.05 0.10 −0.50 0.618
High School 0.11 0.11 1.04 0.299
Less than High School 0.06 0.17 0.36 0.718
Both Parents Unemployed −1.00 0.25 −3.94 < 0.001
No Insurance / Gaps in Coverage 1.54 0.16 9.22 < 0.001
200 < FPL < 400 0.02 0.14 0.15 0.885
100 < FPL < 200 −0.05 0.14 −0.38 0.707
FPL < 100 −0.04 0.15 −0.27 0.785
Female 1.00 0.13 7.58 < 0.001
Black −0.33 0.13 −2.53 0.011
Asian −1.20 0.26 −4.50 < 0.001
Multiracial −0.20 0.20 −1.00 0.317
Hispanic/Latinx −0.43 0.12 −3.52 < 0.001
Female * Black −0.29 0.28 −1.04 0.299
Female * Asian 0.77 0.47 1.63 0.103
Female * Multiracial −0.56 0.43 −1.28 0.199
Hispanic/Latinx * Black 0.30 0.46 0.65 0.515
Hispanic/Latinx * Multiracial −0.73 0.30 −2.45 0.014
Female * Hispanic/Latinx −1.29 0.22 −5.77 < 0.001
Female * Hispanic/Latinx * Multiracial 2.58 0.66 3.89 < 0.001
Open in a new tab

Note. Reference Values (RV): Home Language RV = English; Education RV = College or more; FPL RV = FPL > 400; Employment RV = One or more parents employed; Insurance RV = Full coverage; Sex RV = male; Race RV = White; Ethnicity RV=Non-Hispanic

Our final model yielded two significant two-way interactions: Hispanic*Multiracial, β= −0.73, p = .014, (Fig. 1) and Female*Hispanic, β= −1.29, p < .001 (Fig. 2). A significant three-way interaction of Female*Hispanic*Multiracial, β = 2.58, p < .001, qualified these two-way interactions, which was probed by subsetting our data by Multiracial children (Fig. 3) and white, Black and Asian (Non-Multi-racial) children (Fig. 4) and running simple slope analyses. Being female had disproportionate impacts on non-Multiracial children: females who were non-Hispanic and white, Black, or Asian were diagnosed on average at 4.40 years, β = 1.01, p < .001, (Fig. 4), significantly later than females who were Hispanic and white, Black, or Asian, who on average, were diagnosed at 2.66 years, β= −1.70, p < .001 (Fig. 4). In children who were Hispanic white, Black, or Asian, age of diagnosis did not significantly differ by sex, β= −0.26, p = .158: females were diagnosed at 2.66 years and males were diagnosed at 2.92 years.

Fig. 1.

Fig. 1

Open in a new tab

Interaction between Female * Hispanic/Latinx on age of diagnosis

Fig. 2.

Fig. 2

Open in a new tab

Interaction effect of Multiracial* Hispanic/Latinx ethnicity on age of diagnosis

Fig. 3.

Fig. 3

Open in a new tab

Effect of Hispanic/Latinx ethnicity on Sex on diagnostic age when subsetting data by Multiracial Children

Fig. 4.

Fig. 4

Open in a new tab

Effect of Hispanic/Latinx ethnicity on Sex on diagnostic age when subsetting data by Non-Multiracial Children

Children who were non-Hispanic, regardless of race or sex, had later diagnoses (3.19–4.45 years) than children who were Hispanic/Latinx (1.46–2.96 years), with the exception females who were Hispanic/Latinx and Multiracial (4.45 years), and this varied as a function of sex. Males who were non-Hispanic were diagnosed at 3.35 years if they were white, Black, or Asian, β = 3.35, p < .001 (Fig. 4), compared to males who were Hispanic/Latinx and white, Black, and Asian, diagnosed at 2.92 years, β= −0.43, p < .001. Males who were non-Hispanic and Multiracial were diagnosed at 3.64 years, β = 3.64, p < .001, (Fig. 3), in contrast to males who were Hispanic/Latinx and Multiracial, were diagnosed at 2.87 years, β= −0.76, p = .003.

Given that children who were Black, Asian, and Hispanic/Latinx had significantly earlier diagnoses, contrary to our hypotheses, we conducted post-hoc chi-square analyses of race and ethnicity for presence of ID. Black children were significantly more likely to have ID than children who were not Black, Χ2(1) = 15.50, p < .001, Asian children were more likely to have ID than children who were not Asian, Χ2(1) = 29.18, p < .001, and Hispanic/Latinx children were more likely to have ID than non-Hispanic children, Χ2(1) = 8.54, p < .001. Multiracial children were no more likely to have ID than non-Multiracial children, Χ2(1) = 2.97, p = .084.

Discussion

We examined the intersectionality of sex, race, and ethnicity and their associations with autism prevalence and age of diagnosis using the 2016 to 2021 NSCH. We found an autism prevalence of 2.65% with a population ratio of 1:38 children, comparable to recent CDC estimates (1:36, or 2.76%, Maenner et al., 2023) and in line with increases in autism prevalence in the last decade. Our sex ratio of 3.8:1 is consistent with studies finding a ratio of 4:1 (Maenner et al., 2023). Prevalences ranged from 1.6 (Asian) to 3.0 (NHOPI), with other groups falling in this range and rates differed by race and ethnicity. Children had a median age of diagnosis of 4 years of age and there were overall differences in age of diagnosis by race, sex, and ethnicity: Females were diagnosed later than males particularly if they were non-Hispanic and white, Black, or Asian. Black, Asian, and Hispanic/Latinx children were diagnosed earlier than children of other races and ethnicities. Given the increased incidence of ID found in Black, Asian, and Hispanic children, ID may play a nuanced role in more efficient diagnostic timing.

Our final hierarchical regression model explained 25.3% of the variance in age at autism diagnosis, which only contributed an additional 3% of variance explained from previous models. Although an R2 of 0.25 is comparable to the variance explained in similar linear models investigating age at diagnosis (e.g., 16% in Emerson et al., 2016), the majority of the variance in age of diagnosis is not explained by our model and should be explored in further research.

Our research supports previous findings (e.g., Constantino et al., 2020) that children with co-occurring ID have earlier age of diagnosis. Children were also diagnosed earlier if both of caregivers were unemployed. Although not predicted, this finding may be accounted for by several factors. It is plausible that unemployed caregivers have access to public assistance programs such as Medicaid and Children’s Health Insurance Program (CHIP), as caregivers with privately insured children have shown to be less likely to report that their health insurance covered their child’s autism-specific services than caregivers with publicly insured children (Zhang et al., 2022). Alternatively, children of unemployed caregivers may have earlier diagnoses since, by simply being home more, caregivers may have had more surveillance over their child’s development.

We found that older children at the time of survey completion had later average diagnostic ages as did children who had insurance gaps and children who spoke a non-English home language. These findings were expected: children in older birth cohorts tend to be diagnosed later than children from more recent birth cohorts (Daniels & Mandell, 2014), insurance status has continually shown to be predictive of age of diagnosis (DeGuzman et al., 2017), and English proficiency has shown to be associated with barriers experienced during diagnosis in Hispanic/Latinx families (Chavez et al., 2021). It should be noted that older children have also had more time to accrue a diagnosis, so we suggest interpreting our finding of later age of diagnosis in older children with caution. In addition, experiencing diagnostic delays due to a non-English home language does not connect with our finding that Hispanic/Latinx children had an earlier age of diagnosis than non-Hispanic children. Importantly, our findings of earlier age of diagnosis in Hispanic/Latinx children and later age of diagnosis in children with a home language other than English are main effects, representative of the pattern of findings when other characteristics are accounted for. When considering that n = 190 Hispanic/Latinx children spoke a non-English home language (59.94%), it is important to consider each individual’s intersectionality in informing age of diagnosis. We found that Hispanic/Latinx children too had significantly higher rates of ID which could also account for earlier diagnoses.

We found females were diagnosed 12 months later than males, consistent with previous research (Harrop et al., 2021). Further, our analyses highlight the role of intersectionality in the diagnostic timing of autistic females, such that being female coupled with other racial/ethnic identities differentially impacts diagnostic age. For non-Hispanic white, Black, and Asian children, being female delayed age of diagnosis by nearly two years (4.40 versus 2.67 years in Hispanic/Latinx white, Black, or Asian females). In addition, Multiracial children were diagnosed 9 months later if they were male and non-Hispanic, 14 months later if they were female and Hispanic/Latinx, and nearly 19 months later if they were female and non-Hispanic, suggesting that non-Hispanic white, Black, and/or Asian females are at particular risk of delays.

Our research sheds new light on Multiracial children, particularly females and non-Hispanic males, who, like other multiply minoritized groups (e.g., Black autistic females, see Diemer et al. (2022), are functionally non-existent in autism research. Moreover, our insignificant two-way interactions between females and race but significant two-way interactions between females and ethnicity, Multiracial race and ethnicity, and three-way interaction between female Multiracial children across ethnicities demonstrates the complex role of ethnicity on diagnostic age in autistic females. However, our three-way interaction results when subsetting by Multiracial children (n = 512 children) were limited by the small number of children (n = 24) who were Multiracial, Hispanic/Latinx, and female (Fig. 3) so this group requires further study. Although there were too few cases to analyze associations between ethnicity and diagnostic age in Black and Asian females, our findings suggest that ethnicity is an important factor to consider and evaluate when making conclusions about diagnostic discrepancies, particularly in females.

Collapsing across sex, racial/ethnic minority children were diagnosed earlier than white and non-Hispanic children. Black children were diagnosed almost 4 months earlier than children who were not Black, Hispanic/Latinx children were diagnosed 5 months earlier than non-Hispanic children, and Asian children were diagnosed over 14 months earlier than children who were not Asian. Interestingly, whereas Multiracial children were generally diagnosed later, Multiracial and Hispanic/Latinx children were diagnosed 18 months earlier (Fig. 2) and had the earliest age of diagnosis of any group (1.47 years). These results were contrary to our hypothesis that multiply minoritized youth would be diagnosed later and contradict research suggesting significant diagnostic delays in racial/ethnic minorities (Mandell et al., 2002, 2009). However, our results align with recent studies finding increased autism prevalence and earlier age of diagnosis in minority children (Emerson et al., 2016; Jo et al., 2015; Martin et al., 2022; Wallis et al., 2023), two of which used NSCH data (Emerson et al., 2016; Jo et al., 2015). Emerson et al. (2016) used 2011–2012 data and found that being Black, Hispanic/Latinx, or Multiracial significantly predicted an earlier age of diagnosis. Similarly, Jo et al. (2015) used 2009–2010 data and found that white 5–17-year-olds had a significantly higher proportion of delayed diagnoses than Black children and Hispanic/Latinx children with a non-English home language, although findings were specific to children with caregiver-reported severity ratings of mild or moderate but not severe.

Emerson et al. (2016) and Jo et al. (2015) note that their findings of increased autism prevalence and earlier age of diagnosis in minority children were unexpected, but each suggested important explanations to consider. Emerson et al. (2016) noted factors contributing to earlier diagnoses in minority children, including increased caregiver and pediatrician autism awareness within the last few decades, the 2006 American Academy of Pediatrics initiation of 18- and 24-month autism screenings guidelines for pediatricians, and that earlier autism diagnoses in minority children – who are more likely to be diagnosed with ID (Maenner et al., 2020) - may actually be reflective of ID diagnoses received prior to their autism diagnosis. Jo et al. (2015) posited that their findings were still reflective of diagnostic lag in that minority children with mild and moderate symptoms were less likely to be diagnosed and were thus not well-represented in the dataset.

Although these findings of historically marginalized groups receiving a diagnosis earlier than white and non-Hispanic groups are contrary to previous work, they are in alignment with other research using NSCH data and recent studies (Martin et al., 2022; Wallis et al., 2023). Although not evaluated here, one possible explanation for these findings is that racially minoritized individuals were more likely to have ID, and therefore identified earlier. In a recent analysis of autism prevalence in the New Jersey, Black and Hispanic children had a significantly higher likelihood of being diagnosed as autistic with ID and Black children and children living in more affluent areas had a significantly lower likelihood of being identified as autistic without ID (Shenouda et al., 2023). Our post-hoc, chi-square analyses demonstrated that autistic children who were Black, Hispanic/Latinx, and Asian were significantly more likely to also have ID, consistent with Shenouda et al’s findings. Future research should evaluate the role of trait severity, medical complexity, and ID more specifically as moderators between racial identity and age of diagnosis.

However, symptom severity and co-occurring conditions such as ID may not alone account for findings that racial/ethnic minority children have equal autism prevalence and earlier diagnoses. In one study examining data from thirteen Early Autism Evaluation (EAE) Hub networks aiming to reduce age at diagnosis in Indiana, USA, Martin et al. (2022) found, across all sites, an increased autism prevalence in children whose race was not white even after accounting for symptom severity and developmental impairment. Martin et al. (2022) suggest the EAE Hub system may demonstrate an equitable means of diagnosing autism in toddlers that may even prioritize services for minority youth. In another study, Penner et al. (2023) found that Black, Asian, Hispanic/Latinx, Middle Eastern, and Multiracial children were more likely than white children to receive an accurate pediatrician diagnosis of autism even after controlling for current age, autism symptom severity, and cognitive development. With these findings in mind and the rise in autism awareness, our findings may reflect that autistic minority youth are beginning to be diagnosed at equitable rates to children in majority groups. Another possibility is that racial/ethnic minority children without ID, who are likely diagnosed later, may not be equally represented in this dataset compared to children from majority cultures without ID. It is also possible that relying on racial categories alone may be oversimplifying more complex, interactive patterns, which when accounted for, better reflect the intersectionality of sex, race, and ethnicity.

Strengths and Limitations

Our study had several strengths. Many studies examining diagnostic disparities have used Medicaid-eligible samples (Mandell et al., 2002, 2007) which oversample for racial/ethnic minorities. A strength of our study was our nationally representative sample of 218,456 US children in years 2016 to 2021. Our large sample afforded us the ability to examine sex, race, and ethnicity on autism prevalence and diagnostic age within and between groups of children. We were unfortunately not afforded the ability to examine the intersectionality of select racial groups (Some Other Race Alone, NHOPI, or American Indian/Alaska Native children) due to limited sample size. Although simply excluding these participants is an inelegant approach to this data, future work could take a descriptive approach to evaluate intersectionality.

Our study has several limitations to consider. First, NSCH is a parent report survey. Though parent reports can provide in-depth information about child medical history, they also present concerns for potential discrepancies. Specifically, forward telescoping bias poses a potential risk “in which people report events as having occurred more recently (closer to the time of recollection) than they actually took place” (Ozonoff et al., 2018, p. 891). For example, Ozonoff et al. (2018) documented forward telescoping specifically related to parent reported age of milestone achievement and skill regression for their autistic child at approximately 2–3 years of age and again around age 6. This is significant as, depending on when the child was diagnosed relative to survey completion, the accuracy of the parent recall may vary.

Other limitations pertain to our regression models. First, we did not consider autism severity. Although NSCH has a three-level severity variable (K2Q35C, rated mild, moderate, or severe), we did not include K2Q35C in analyses due to subjective parent-report (Hus et al., 2011; Ozonoff et al., 2018) as opposed to more consistent severity measures (i.e., comparison scores, normed measures). Future studies with clinician report data could further delve into the role of trait severity in age of diagnosis. Secondly, when subsetting by Multiracial children to probe our three-way interaction (Fig. 3), our findings were limited by the small number of children who were Multiracial, Hispanic/Latinx, and female (n = 24). Future research should corroborate our findings in this specific group of children. Finally, since ID significantly predicted age of diagnosis, it is important to acknowledge that ID was caregiver reported. National surveillance studies of ID rates in autism have been consistently reported as 30–40% (Baio et al., 2018; Maenner et al., 2021, 2023) and our findings demonstrate a rate of only 18%, corroborating research showing that caregivers are likely to underreport child ID (Lee et al., 2023). It is also important to consider that racial/ethnic minority youth may still face delays in access to early intensive behavioral intervention (EIBI) despite earlier age of autism diagnosis in these groups. Autistic children with earlier access to EIBI may be more likely to circumvent delays in cognitive development: research has shown that autistic children with shorter lags from diagnosis to EIBI are more likely to have a general education placement (Dimian et al., 2021). Future research should explore access to and experiences with EIBI in autistic minority youth and the influence of ID.

Future Directions

Our sample, like the overall US population, was majority white, non-Hispanic. Although our analyses contained a considerable number of racial/ethnic minority autistic children and females, we did not include children who were Some Other Race Alone, NHOPI, or American Indian/Alaska Native or many children from multiply minoritized groups (e.g., Asian Hispanic females). Future research should use multi-year population-based studies and other national (Medicaid-eligible samples) and research-based datasets that oversample for racial/ethnic minorities and females to examine autism prevalence in understudied groups. Research should also examine prevalence and diagnostic disparities in autistic young adults with later diagnostic ages (17 years of age +) not represented in this sample. Finally, future research should investigate the interactive effect of ID on the relationship between sex, race, and ethnicity and age of diagnosis.

Conclusion

This study used the 2016 through 2021 NSCH to examine the role of race, sex, and ethnicity in autism prevalence and age of diagnosis. We found equal autism prevalence rates across racially and ethnically minoritized children and earlier age of diagnosis in children who had multiple minoritized identities. Our results highlight the importance of the intersection of sex, race, and ethnicity, as well as demographic variables, to answer these questions. This research has important implications for many autistic minority children. To optimize outcomes and reduce diagnostic delays, targeted work is needed to increase autism screening in non-Hispanic children and in children who may be overlooked during diagnoses (e.g., females, children without ID).

Supplementary Material

Supplemental Tables

Acknowledgements

We would like to thank the staff who collected the NSCH data and the families who took the time to participate. We would also like to thank Drs Wanqing Zhang and Peter Halpin for their discussion of the data and analysis. Data were presented at the International Society for Autism Research conference in 2021. This research is supported by a KL2 Career Development Award which was awarded to C.H. through the National Center for Advancing Translation Sciences Award (UL1TR001111; PI: Buse) and a training fellowship awarded to T.C.M. (2T32 HD040127–21).

Footnotes

Conflicts of interest The authors report no conflicts of interest.

1

We use World Health Organization definitions for sex and gender: “sex” is genetically and biologically determined whereas “gender” represents a socio-cultural construct (“WHO | Gender and Genetics,” 2010). We use “male” and “female” to refer to biological sex assigned at birth.

2

Although we examine five years, research using NSCH found that the COVID-19 pandemic did not significantly impact parent-reported autism prevalence from 2016 to 2021 (Wang et al., 2023).

3

To be consistent with APA formatting, we write decimals to the tenths place. Therefore, while our 95% confidence intervals spanned the value of each corresponding odds ratio, many did so in the hundredths place decimal or beyond so some may not appear to span the odds ratio value.

References

  1. Akobirshoev I, Mitra M, Li FS, Dembo R, Dooley D, Mehta A, & Batra N (2020). The compounding effect of Race/Ethnicity and disability status on children’s Health and Health Care by Geography in the United States. Medical Care, 58(12), 1059–1068. 10.1097/MLR.0000000000001428. [DOI] [PubMed] [Google Scholar]
  2. American Psychiatric Association (2013). The Diagnostic and Statistical Manual of Mental Disorders: DSM 5. 10.1176/appi.books.9780890425787. [DOI]
  3. Anderson DK, Liang JW, & Lord C (2014). Predicting young adult outcome among more and less cognitively able individuals with autism spectrum disorders. Journal of Child Psychology and Psychiatry, 55(5), 485–494. 10.1111/JCPP.12178. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Baio J, Wiggins L, Christensen DL, Maenner MJ, Daniels J, Warren Z, Kurzius-Spencer M, Zahorodny W, Robinson C, Rosenberg, White T, Durkin MS, Imm P, Nikolaou L, Yeargin-Allsopp M, Lee LC, Harrington R, Lopez M, Fitzgerald RT, & Dowling NF (2018). Prevalence of Autism Spectrum Disorder among Children aged 8 years — Autism and Developmental Disabilities Monitoring Network, 11 Sites, United States, 2014. MMWR Surveillance Summaries, 67(6), 1–23. 10.15585/mmwr.ss6706a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Blacher J, Stavropoulos K, & Bolourian Y (2019). Anglo-latino differences in parental concerns and service inequities for children at risk of autism spectrum disorder. Autism, 23(6), 1554–1562. 10.1177/1362361318818327. [DOI] [PubMed] [Google Scholar]
  6. Chavez AE, Feldman MS, Carter AS, Eisenhower A, Mackie TI, Ramella L, Hoch N, & Sheldrick RC (2021). Delays in Autism diagnosis for U.S. Spanish-speaking families: The contribution of appointment availability. Evidence-Based Practice in Child and Adolescent Mental Health, 7(2), 275–293. 10.1080/23794925.2021.2001772. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Child and Adolescent Health Measurement Initiative (2016). 2016–2021 National Survey of Children’s Health (NSCH) [SAS Constructed DataSet]. Data Resource Center for Child and Adolescent Health supported by Cooperative Agreement U59MC27866 from the U.S. Department of Health and Human Services, Health Resources and Services Administration (HRSA), Maternal and Child Health Bureau (MCHB). Retrieved 10/27/2022 from www.childhealthdata.org. [Google Scholar]
  8. Constantino JN, Abbacchi AM, Saulnier C, Klaiman C, Mandell DS, Zhang Y, Hawks Z, Bates J, Klin A, Shattuck P, Molholm S, Fitzgerald R, Roux A, Lowe JK, & Geschwind DH (2020). Timing of the diagnosis of Autism in African American Children. Pediatrics, 146(3), 10.1542/PEDS.2019-3629. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Daniels AM, & Mandell DS (2014). Explaining differences in age at autism spectrum disorder diagnosis: A critical review. Autism: The International Journal of Research and Practice, 18(5), 583–597. 10.1177/1362361313480277. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Dawson JF (2014). Moderation in Management Research: What, why, when, and how. Journal of Business and Psychology, 29(1), 1–19. 10.1007/s10869-013-9308-7. [DOI] [Google Scholar]
  11. DeGuzman PB, Altrui P, Allen M, Deagle CR, & Keim-Malpass J (2017). Mapping geospatial gaps in early identification of children with Autism Spectrum Disorder. Journal of Pediatric Health Care, 31(6), 663–670. 10.1016/j.pedhc.2017.05.003. [DOI] [PubMed] [Google Scholar]
  12. Diemer MC, Gerstein ED, & Regester A (2022). Autism presentation in female and Black populations: Examining the roles of identity, theory, and systemic inequalities. In Autism (Vol. 26, Issue 8, pp. 1931–1946). SAGE Publications Ltd. 10.1177/13623613221113501. [DOI] [PubMed] [Google Scholar]
  13. Dimian AF, Symons FJ, & Wolff JJ (2021). Delay to early intensive behavioral intervention and Educational Outcomes for a Medicaid-Enrolled cohort of children with autism. Journal of Autism and Developmental Disorders, 51(4), 1054–1066. 10.1007/s10803-020-04586-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Donohue MR, Childs AW, Richards M, & Robins DL (2019). Race influences parent report of concerns about symptoms of autism spectrum disorder. Autism: The International Journal of Research and Practice, 23(1), 100. 10.1177/1362361317722030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Durkin MS, Maenner MJ, Meaney FJ, Levy SE, di Guiseppi C, Nicholas JS, Kirby RS, Pinto-Martin JA, & Schieve LA (2010). Socioeconomic inequality in the prevalence of autism spectrum disorder: Evidence from a U.S. cross-sectional study. PloS One, 5(7), 10.1371/JOURNAL.PONE.0011551. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Eilenberg JS, Paff M, Harrison AJ, & Long KA (2019). Disparities based on race, ethnicity, and socioeconomic Status over the transition to Adulthood among Adolescents and Young adults on the Autism Spectrum: A systematic review. In current Psychiatry reports. Issue 5) Current Medicine Group LLC, 21, 1. 10.1007/s11920-019-1016-1. [DOI] [PubMed] [Google Scholar]
  17. Emerson ND, Morrell HER, & Neece C (2016). Predictors of age of diagnosis for children with Autism Spectrum Disorder: The role of a consistent source of Medical Care, Race, and Condition Severity. Journal of Autism and Developmental Disorders, 46(1), 127–138. 10.1007/S10803-015-2555-X. [DOI] [PubMed] [Google Scholar]
  18. Estes A, Munson J, Rogers SJ, Greenson J, Winter J, & Dawson G (2015). Long-term outcomes of early intervention in 6-Year-old children with Autism Spectrum Disorder. Journal of the American Academy of Child & Adolescent Psychiatry, 54(7), 580–587. 10.1016/J.JAAC.2015.04.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Fombonne E (2009). Epidemiology of pervasive developmental disorders. Pediatric Research, 65(6), 591–598. 10.1203/PDR.0b013e31819e7203. [DOI] [PubMed] [Google Scholar]
  20. Harrop C, Libsack E, Bernier R, Dapretto M, Jack A, McPartland JC, van Horn JD, Webb SJ, Pelphrey K, & the GENDAAR Consortium. (2021). Do Biological Sex and Early Developmental Milestones predict the age of first concerns and eventual diagnosis in Autism Spectrum Disorder? Autism Research, 14(1), 156–168. 10.1002/aur.2446. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Hus V, Taylor A, & Lord C (2011). Telescoping of caregiver report on the Autism Diagnostic Interview - Revised. In Journal of Child Psychology and Psychiatry and Allied Disciplines (Vol. 52, Issue 7, pp. 753–760). J Child Psychol Psychiatry. 10.1111/j.1469-7610.2011.02398.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hyman SL, Levy SE, Myers SM, Kuo DZ, Apkon CS, Davidson LF, Ellerbeck KA, Foster JEA, Noritz GH, O’Connor Leppert M, Saunders BS, Stille C, Yin L, Brei T, Davis BE, Lipkin PH, Norwood K, Coleman C, Mann M, & Paul L (2020). Identification, evaluation, and management of children with autism spectrum disorder. Pediatrics, 145(1), 10.1542/PEDS.2019-3447. [DOI] [PubMed] [Google Scholar]
  23. Jo H, Schieve LA, Rice CE, Yeargin-Allsopp M, Tian LH, Blumberg SJ, Kogan MD, & Boyle CA (2015). Age at Autism Spectrum disorder (ASD) diagnosis by race, ethnicity, and primary Household Language among children with Special Health Care needs, United States, 2009–2010. Maternal and Child Health Journal, 19(8), 1687–1697. 10.1007/S10995-015-1683-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kogan MD, Vladutiu CJ, Schieve LA, Ghandour RM, Blumberg SJ, Zablotsky B, Perrin JM, Shattuck P, Kuhlthau KA, Harwood RL, & Lu MC (2018). The prevalence of parent-reported autism spectrum disorder among US children. Pediatrics, 142(6), 20174161. 10.1542/PEDS.2017-4161/76865. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kogan, Blumberg SJ, Schieve LA, Boyle CA, Perrin JM, Ghandour RM, Singh GK, Strickland BB, Trevathan E, & van Dyck PC. (2009). Prevalence of parent-reported diagnosis of autism spectrum disorder among children in the US, 2007. Pediatrics, 124(5), 1395–1403. 10.1542/peds.2009-1522. [DOI] [PubMed] [Google Scholar]
  26. Lee CM, Snyder G, Carpenter L, Harris LA, Kanne J, Taylor S, Sarver CM, Stephenson DE, Shulman KG, Wodka LH, Esler EL, & SPARK Consortium. (2023). Agreement of parent-reported cognitive level with standardized measures among children with autism spectrum disorder. Autism Research: Official Journal of the International Society for Autism Research. 10.1002/aur.2934. [DOI] [PubMed] [Google Scholar]
  27. Li Q, Li Y, Liu B, Chen Q, Xing X, Xu G, & Yang W (2022). Prevalence of Autism Spectrum Disorder among Children and Adolescents in the United States from 2019 to 2020. JAMA Pediatrics, 176(9), 943–945. 10.1001/JAMAPEDIATRICS.2022.1846. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Liptak GS, Benzoni LB, Mruzek DW, Nolan KW, Thingvoll MA, Wade CM, & Fryer GE (2008). Disparities in diagnosis and access to health services for children with autism: Data from the national survey of children’s health. Journal of Developmental and Behavioral Pediatrics, 29(3), 152–160. 10.1097/DBP.0B013E318165C7A0. [DOI] [PubMed] [Google Scholar]
  29. Lockwood Estrin G, Milner V, Spain D, Happé F, & Colvert E (2021). Barriers to Autism Spectrum Disorder diagnosis for Young Women and Girls: A systematic review. Review Journal of Autism and Developmental Disorders (Vol, 8(4), 454–470. 10.1007/s40489-020-00225-8. Springer. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Loomes R, Hull L, & Mandy WPL (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 & Adolescent Psychiatry, 56(6), 466–474. 10.1016/j.jaac.2017.03.013. [DOI] [PubMed] [Google Scholar]
  31. Maenner MJ, Shaw KA, Baio J, EdS1, Washington A, Patrick M, DiRienzo M, Christensen DL, Wiggins LD, Pettygrove S, Andrews JG, Lopez M, Hudson A, Baroud T, Schwenk Y, White T, Rosenberg CR, Lee LC, Harrington RA, & Dietz PM. (2020). Prevalence of Autism Spectrum Disorder among Children aged 8 years — Autism and Developmental Disabilities Monitoring Network, 11 Sites, United States, 2016. MMWR Surveillance Summaries, 69(4), 1. 10.15585/MMWR.SS6904A1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Maenner MJ, Shaw KA, Bakian A, Bilder DA, Durkin MS, Esler A, Furnier SM, Hallas L, Hall-Lande J, Hudson A, Hughes MM, Patrick M, Pierce K, Poynter JN, Salinas A, Shenouda J, Vehorn A, Warren Z, Constantino JN, & Cogswell ME (2021). Prevalence and characteristics of Autism Spectrum Disorder among Children aged 8 years — Autism and Developmental Disabilities Monitoring Network, 11 Sites, United States, 2018. MMWR Surveillance Summaries, 70(11), 1–16. 10.15585/MMWR.SS7011A1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Maenner MJ, Warren Z, Williams AR, Amoakohene E, Bakian A, Bilder DA, Durkin MS, Fitzgerald RT, Furnier SM, Hughes MM, Ladd-Acosta CM, McArthur D, Pas ET, Salinas A, Vehorn A, Williams S, Esler A, Grzybowski A, Hall-Lande J, & Shaw KA (2023). Prevalence and characteristics of Autism Spectrum Disorder among Children aged 8 years — Autism and Developmental Disabilities Monitoring Network, 11 Sites, United States, 2020. MMWR Surveillance Summaries, 72(2), 1–14. 10.15585/mmwr.ss7202a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Magaña S, Parish SL, Rose RA, Timberlake M, & Swaine JG (2012). Racial and ethnic disparities in quality of Health Care among children with autism and other Developmental Disabilities. Intellectual and Developmental Disabilities, 50(4), 287–299. 10.1352/1934-9556-50.4.287. [DOI] [PubMed] [Google Scholar]
  35. Magaña S, Parish SL, & Son E (2015). Have racial and ethnic disparities in the quality of Health Care Relationships changed for children with Developmental Disabilities and ASD? American Journal on Intellectual and Developmental Disabilities, 120(6), 504–513. 10.1352/1944-7558-120.6.504. [DOI] [PubMed] [Google Scholar]
  36. Mandell DS, Listerud J, Levy SE, & Pinto-Martin JA (2002). Race differences in the age at diagnosis among medicaid-eligible children with autism. Journal of the American Academy of Child and Adolescent Psychiatry, 41(12), 1447–1453. 10.1097/00004583-200212000-00016. [DOI] [PubMed] [Google Scholar]
  37. Mandell DS, Ittenbach RF, Levy SE, & Pinto-Martin JA (2007). Disparities in diagnoses received prior to a diagnosis of autism spectrum disorder. Journal of Autism and Developmental Disorders, 37(9), 1795–1802. 10.1007/s10803-006-0314-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Mandell DS, Wiggins LD, Carpenter LA, Daniels J, DiGuiseppi C, Durkin MS, Giarelli E, Morrier MJ, Nicholas JS, Pinto-Martin JA, Shattuck PT, Thomas KC, Yeargin-Allsopp M, & Kirby RS (2009). Racial/ethnic disparities in the identification of children with autism spectrum disorders. American Journal of Public Health, 99(3), 493–498. 10.2105/AJPH.2007.131243. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Mandell DS, Morales K, Xie M, Lawer L, Stahmer AC, & Marcus S (2010). Age of diagnosis among Medicaid-Enrolled children with autism, 2001–2004. Psychiatric Services, 61(8), 822–829. 10.1176/PS.2010.61.8.822. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Martin AM, Ciccarelli MR, Swigonski N, & McNally Keehn R (2022). Evaluation of race and ethnicity across a statewide system of early autism evaluation. Journal of Pediatrics, 0(0), 10.1016/j.jpeds.2022.10.023. [DOI] [Google Scholar]
  41. Milner V, McIntosh H, Colvert E, & Happé F (2019). A qualitative exploration of the female experience of Autism Spectrum disorder (ASD). Journal of Autism and Developmental Disorders, 49(6), 2389–2402. 10.1007/s10803-019-03906-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Onaiwu MG (2020). They don’t know, don’t show, or don’t Care”: Autism’s White Privilege Problem. Https://Home Liebertpub Com/Aut, 2(4), 270–272. 10.1089/AUT.2020.0077. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. 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, 22(7), 891–896. 10.1177/1362361317710798. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Pearson JN, & Meadan H (2018). African american parents’ perceptions of diagnosis and services for children with autism. Education and Training in Autism and Developmental Disabilities, 53(1), 17–32. [Google Scholar]
  45. Penner M, Senman L, Andoni L, Dupuis A, Anagnostou E, Kao S, Solish A, Shouldice M, Ferguson G, & Brian J (2023). Concordance of diagnosis of autism spectrum disorder made by pediatricians vs a multidisciplinary specialist team. JAMA Network Open, 6(1), e2252879. 10.1001/jamanetworkopen.2022.52879. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Ratto AB, Anthony BJ, Kenworthy L, Armour AC, Dudley K, & Anthony LG (2016). Are non-intellectually disabled Black Youth with ASD Less impaired on parent report than their White Peers? Journal of Autism and Developmental Disorders, 46(3), 773. 10.1007/S10803-015-2614-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. SAS Institute Inc. (2016). SAS/STAT ® 9.4 user’s guide. SAS Institute Inc. [Google Scholar]
  48. Shenouda J, Barrett E, Davidow AL, Sidwell K, Lescott C, Halperin W, Silenzio VMB, & Zahorodny W (2023). Prevalence and disparities in the detection of Autism without Intellectual disability. Pediatrics, 151(2), 10.1542/PEDS.2022-056594. [DOI] [PubMed] [Google Scholar]
  49. Solomon M, Miller M, Taylor SL, Hinshaw SP, & Carter CS (2012). Autism symptoms and internalizing psychopathology in girls and boys with autism spectrum disorders. Journal of Autism and Developmental Disorders, 42(1), 48–59. 10.1007/s10803-011-1215-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. South M, Beck JS, Lundwall R, Christensen M, Cutrer EA, Gabrielsen TP, Cox JC, & Lundwall RA (2020). Unrelenting depression and suicidality in women with autistic traits. Journal of Autism and Developmental Disorders, 50(10), 3606–3619. 10.1007/s10803-019-04324-2. [DOI] [PubMed] [Google Scholar]
  51. Strang JF, van der Miesen AI, Caplan R, Hughes C, daVanport S, & Lai MC (2020). Both sex- and gender-related factors should be considered in autism research and clinical practice. Autism, 24(3), 539–543. 10.1177/1362361320913192. [DOI] [PubMed] [Google Scholar]
  52. Volkmar FR, Szatmari P, & Sparrow SS (1993). Sex differences in pervasive developmental disorders. Journal of Autism and Developmental Disorders, 23(4), 579–591. 10.1007/BF01046103. [DOI] [PubMed] [Google Scholar]
  53. Von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, & Vandenbroucke JP (2007). The strengthening the reporting of Observational Studies in Epidemiology (STROBE) statement: Guidelines for reporting observational studies. PLoS Wallis, [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. K. E., Adebajo T, Bennett AE, Drye M, Gerdes M, Miller JS, & Guthrie W. (2023). Short report: Prevalence of autism spectrum disorder in a large pediatric primary care network. Https://DoiOrg/,136236132211473. 10.1177/13623613221147396. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Wang X, Weng X, Pan N, Li X, Lin L, & Jing J (2023). Prevalence of Autism Spectrum Disorder in the United States is Stable in the COVID-19 Era. Journal of autism and developmental disorders, 1–4. Advance online publication. 10.1007/s10803-023-05915-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Whitlock A, Fulton K, Lai MC, Pellicano E, & Mandy W (2020). Recognition of Girls on the Autism Spectrum by Primary School educators: An experimental study. Autism Research, 13(8), 1358–1372. 10.1002/AUR.2316. [DOI] [PubMed] [Google Scholar]
  57. Wiggins LD, Durkin M, Esler A, Lee LC, Zahorodny W, Rice C, Yeargin-Allsopp M, Dowling NF, Hall-Lande J, Morrier MJ, Christensen D, Shenouda J, & Baio J (2020). Disparities in documented Diagnoses of Autism Spectrum Disorder based on demographic, individual, and Service factors. Autism Research, 13(3), 464–473. 10.1002/aur.2255. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Zhang Y, Li N, Li C, Zhang Z, Teng H, Wang Y, Zhao T, Shi L, Zhang K, Xia K, Li J, & Sun Z (2020). Genetic evidence of gender difference in autism spectrum disorder supports the female-protective effect. Translational Psychiatry, 10(1), 10.1038/s41398-020-0699-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Zhang W, Thompson KL, Watson LR, & LaForett DR (2022). Health Care utilization for privately and publicly insured children during Autism Insurance Reform. Journal of Autism and Developmental Disorders, 52(11), 5042–5049. 10.1007/S10803-021-05370-5/TABLES/2. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Tables
NIHMS1952092-supplement-Supplemental_Tables.docx (23.9KB, docx)

RESOURCES