A comparison of Autism Spectrum Disorder DSM-IV criteria and associated features among African American and white children in Philadelphia County

Article Outline

• Abstract
• Background
  • Race differences in diagnosis
  • Race differences in clinical presentation of ASD
  • Race difference in clinical presentation of other mental health conditions
• Methods
  • Design
  • Subjects and study area
  • Sources of data
  • Procedures for data collection
  • Outcome variables
  • Analyses
  • Power analysis (Table 3)
• Results
  • Sample (Table 2)
  • Age of first evaluation and diagnosis (Table 2)
  • DSM criteria for qualitative impairments in social interaction, communication, and repetitive/restrictive behaviors (Table 4)
  • Associated features (Table 6)
• Discussion
  • Limitations
• Conclusions
  • Recommendations for research
• References
• Copyright

Abstract 

Background

Racial differences are documented in the timing and type of autism spectrum disorder (ASD) diagnosis among white and African American children. Differences in clinical presentation by race may contribute to these disparities. This study explores documented differences in core ASD symptoms and associated behavioral features among African American and white children.

Methods

This project is a secondary data analysis from the Pennsylvania Autism and Developmental Disabilities Surveillance Program and utilized methodology that evaluates existing records, reviews, and codes for DSM-IV criteria for ASD and 12 associated behavioral features. The sample comprised 343 children meeting surveillance case definition for ASD, from 3 population-based cohorts of children in Philadelphia County.

Results

A higher frequency of white children compared to African American children with ASD have documented DSM-IV criteria of inflexible adherence to nonfunctional routines/rituals (92% vs 81%; p = .005) and persistent preoccupation with parts of objects (67% vs 50%; p = .002). A higher frequency of white children with ASD compared to African American children with ASD have documented abnormal motor development (74% vs 60%; p = .008) and odd responses to sensory stimuli (76% vs 51%; p < .001). There were no significant differences in externalizing behaviors or reciprocal social interaction.

Conclusions

This study suggests differences in the types of ASD symptoms and associated behavioral features exhibited by African American as compared to white children with ASD. Further research is needed to determine if these differences contribute to disparities in the timing or type of ASD diagnosis.

Keywords: Population based surveillance, Autism spectrum disorders, Race/ethnicity, Associated features, DSM criteria

Autism spectrum disorders (ASDs) are a group of developmental disabilities that include autism, Asperger’s syndrome, and pervasive developmental disorder–not otherwise specified (PDD-NOS). The category also includes Rett’s disorder and childhood disintegrative disorder, but as the clinical presentation and course of these disorders differ from the previously named ASDs, they are not included in the analysis. ASDs are estimated to affect 0.6% to 1% of children in the United States and the rate of diagnosis has increased significantly over the past 10 years [1], [2], [3]. ASDs are diagnosed based on impairment in 3 domains: social interaction, communication, and repetitive movements or restricted interests. The Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR) describes 4 characteristic symptoms for each of these domains and at least 1 to 2 symptoms in each domain must be present for a diagnosis [4].

In addition to the 3 domains of impairment considered for a diagnosis of ASD, many children have associated behavior problems. For example, children with ASDs may have symptoms of ADHD, externalizing behaviors such as aggression or oppositionality, mood disorders such as anxiety or depression, and unusual responses to sensory stimuli [5]. Often it is these associated behavior problems that lead families to seek a diagnostic evaluation.

Racial categorization is based on cultural and, to a lesser extent, biological factors. Race has been shown to influence patients’ access to health care resources prompting epidemiologists and clinicians to be mindful of potential racial disparities in the diagnosis and treatment of individuals with autism spectrum disorders (ASDs). Disparities in the timing of the ASD diagnosis and outcome of diagnostic evaluations have been reported for African American as compared to white children with ASD. This report explores whether there are racial differences in symptoms among cases of ASD in 3 birth cohorts in Philadelphia County: children 8 years of age in 2002, 2006, and 2008. We examine differences in reciprocal social interaction, communication and restricted interests/repetitive behaviors among African American and white children with ASD as documented in clinical records. We also examine if African American children with ASD have differences in documented associated behavioral difficulties.

Back to Article Outline

Background 

Some previous studies reported that the prevalence of ASD does not vary by race [6], [7], [8], [9], but these studies are in conflict with a more recent study by the CDC that did report differences in prevalence by race and/or ethnicity in the latest surveillance data among whites, African Americans, and Hispanics. In this study, the overall average prevalence of ASDs was higher among whites than African American or Hispanic children. According to these data for ASDs in Philadelphia County, for children aged 8 years the prevalence was 1% for white children and 0.75% for African American children [2].

Race differences in diagnosis 

While prevalence may or may not vary by race, there are reports of racial differences in the timing of diagnosis and in the type of ASD diagnosed. Differences in the timing of the diagnosis are significant as earlier treatment results in improved outcomes [10]. In 1 study in Philadelphia County, African American children were reported to be diagnosed an average of 2 years later than white children [11]. In another, African American children were more likely to be given an initial diagnosis other than an ASD at their first specialty care visit, usually ADHD (attention-deficit hyperactivity disorder) or conduct disorder (CD), before later being diagnosed with ASD [12]. Data from the Autism and Developmental Disabilities Monitoring Network (ADDM), a population-based public health surveillance network in the United States, reported that when African American children were diagnosed with an autism spectrum disorder, they were more likely to be classified as autistic, rather than having Asperger’s syndrome or PDD-NOS [13].

The reasons for the differences in the timing of diagnosis are not well understood. Differences in access to care and care seeking behaviors could contribute these disparities. This study will investigate the possibility that differences in symptoms of impairment may influence the approach, and possibly the expectations, of the evaluator and, thereafter, the results of diagnostic evaluation.

Race differences in clinical presentation of ASD 

Previous studies of racial differences in the core symptoms of ASD have produced conflicting results. Cuccaro and colleagues [14] found that African American children with ASD were more likely to have delayed language compared to white children. Yet, Lord and colleagues [15] did not find any difference by race in verbal measures on the Autism Diagnostic Interview (ADI) or on the Autism Diagnostic Observation Schedule (ADOS). The authors of this report have observed in clinical practice that African American children with ASD appear to have less frequent deficits in eye contact or nonverbal behavior than white children. These observations have not been confirmed in controlled studies and warrant a more systematic review of cases for evidence of racial differences in the presentation of core symptoms of ASD.

Differences in clinical presentation are potential factors that confound the relationship between race and diagnosis [16] such that core symptoms may appear at different ages. A 2008 report of ASD cases identified aggression and defiance as behavioral features identified more often in 8-year-old African American children with ASD compared to white children with ASD [17]. In this study, only 9% of African American children had a previous diagnosis of ASD compared to 65% of white children [17].

Race difference in clinical presentation of other mental health conditions 

Differences in the clinical presentation among African American and white children in other developmental and behavioral conditions have been reported. A study of racial differences of parent-reported ADHD using a parental questionnaire showed more parent ratings of symptoms of hyperactivity, impulsivity, and inattention in African Americans compared to white children [18]. A similar pattern was noted on teacher rating scales of ADHD symptoms, with African American children receiving higher scores on behavior rating scales of externalizing symptoms such as hyperactivity and conduct problems as well as inattention [19], [20], [21]. Arnold and colleagues [22] found that among children diagnosed with ADHD in the Multimodal Treatment of ADHD study sample, African American children had more teacher-rated symptoms of ADHD and ODD (oppositional defiant disorder) at baseline compared to white children.

There is also evidence for race differences in clinical presentation in psychotic disorders [23]. In a 2001 study of 1001 children, African American youth were more likely to be diagnosed with schizophrenic spectrum disorders than white youths [24]. This disparity was not explained by insurance status. In 2007, a large study of 2991 youths ages 4 to 22 years, reported that African American youths were more likely than white youths to be diagnosed with a psychotic disorder [25].

In summary, based on clinical observation and the literature to date, we hypothesize that (1) African American children with ASD will have less documented impairment in reciprocal social interaction compared to white children with ASD as measured by the presence of DSM criteria 1a through d; (2) African American children with ASD will have no differences in communication or repetitive, restrictive or stereotypical patterns of behavior; and (3)African American children with ASD will have more documented externalizing behaviors, only, as measured by the presence or absence of associated features of aggression, oppositionality, hyperactivity, and temper tantrums, compared to white children with ASD.

Back to Article Outline

Methods 

This study was conducted on 3 cohorts of children living in Philadelphia County. This geographic region is well suited to explore the question of differences in reported characteristics among white and African American children with ASD. According to data from the US Census Bureau, the proportion of children in the 5- to 9-year age range in Philadelphia County corresponds to the national average of approximately 12% to 13% of the total population [26]. In Philadelphia County, there are almost equal percentages of whites and African Americans (45% and 42 %, respectively) compared with national demographics (75% white and 12% African American). The 2002, 2006, and 2008 Pennsylvania vital statistics derived from the Unites States Census data shows consistent proportions of African Americans (13%) and whites (81%) in the 5- to 9-age range over the 3 study years [27], [28], [29]. The average population base for these 3 study years is approximately 22,000 children who were 8 years old. This provides a large and diverse pool for surveillance data collection. Philadelphia County is also home to several major children’s hospitals that are a resource for many children with developmental disorders including ASDs.

Design 

This project is a secondary analysis of data from the Pennsylvania Autism and Developmental Disabilities Surveillance Program (PADDSP). PADDSP is part of the Autism and Developmental Disabilities Monitoring (ADDM) Network, which is the public health surveillance program established in 2000 by the Centers for Disease Control and Prevention to monitor the prevalence of ASDs [30]. Institutional review board approval of The Children’s Hospital of Philadelphia and The University of Pennsylvania was obtained prior to data collection.

Subjects and study area 

For inclusion a child must have resided in Philadelphia County with at least 1 parent or guardian during the study year and have been born between January 1 and December 31 in the years 1994, 1996, and 2000. Data from the ADDM network for Philadelphia County only were chosen for this study because the prevalence of ASDs in Philadelphia County is consistent with national surveillance data and this catchment area has one of the highest percentages of African Americans in the population base (42%) within the ADDM network.

Sources of data 

Data comprised information abstracted from evaluation summaries contained in the medical records of children who met eligibility criteria. The age of 8 years was selected as previous surveillance studies show that children who have ASD are likely to have received a diagnosis by this age [30].

Procedures for data collection 

To identify cases of ASD (administrative prevalence), the ADDM network utilizes a methodology originally implemented by the Metropolitan Atlanta Developmental Disability Surveillance Program (MADDSP) that applies a common case definition, standardized method of data abstraction and clinician review of abstracted records [31]. This surveillance methodology and validation are described in detail elsewhere [1], [32].

Data abstractors refer to a list of 34 “trigger phrases” describing social and communication behaviors associated with ASD. The abstractors consolidate records and abstracted comments containing these phrases into a summary that is stored in a secure database. These records are later de-identified for the process of clinician review. Summaries from multiple sources are combined to create a composite record for each child that is evaluated by an expert clinician reviewer to determine if the child meets criteria as a case of ASD (autistic disorder, ASD-NOS, or PDD-NOS).

The clinician reviewer classifies the children as having an ASD if they (1) have evidence of delays prior to age 3 and meet DSM IV criteria in social, communication, and behavior domains or (2) meet criteria in social and either communication or behavior domains for PDD-NOS (see Table 1).

Table 1. DSM-IV TR Criteria for Autism

Developmental delay <3 years of age along with qualitative impairments in:

1. Reciprocal Social Interaction

a. Nonverbal Behavior

b. Peer Relationships

c. Spontaneous Seeking

d. Emotional Reciprocity

2. Communication

a. Spoken Language

b. Conversational Deficit

c. Repetitive Language

d. Imaginative Play

3. Restricted, Repetitive& Stereotyped Patterns of Behavior, Interests & Activities

a. Restricted Interests

b. Routines and Rituals

c. Stereotyped Mannerisms

d. Preoccupation with Parts

In addition, 13 associated behavioral features are recorded if present. Behavioral features that were examined in this study include abnormalities in eating/drinking, abnormal sleep, abnormal mood, abnormal development of cognitive skills, aggression, oppositionality/defiance, delayed motor milestones, hyperactivity/inattention, abnormal level of fear, odd responses to sensory stimuli, self-injurious behavior, seizures or seizure-like behaviors, and temper tantrums. These behaviors are coded as present based on specific trigger phrases.

Outcome variables 

The main outcomes for analysis are DSM-IV-TR criteria for core domains, including impaired reciprocal social interaction (denoted by DSM criteria 1a, 1b, 1c, and 1d), impaired communication (denoted by DSM criteria 2a, 2b, 2c, and 2d) and restrictive, repetitive or stereotyped behaviors (denoted by DSM criteria 3a, 3b, 3c, and 3d). Please see Table 1 for specifics of criteria. In addition, we will examine the presence of the associated behavioral features including abnormalities in eating, sleep, cognitive and motor development, abnormal fear or sensory responses, seizure-like activity, self-injurious behavior and externalizing behaviors of aggression, oppositionality, hyperactivity, and temper tantrums.

Analyses 

Descriptive statistics (means, SD, frequencies, percents) are used to summarize the demographic characteristics of the sample (see Table 2). Frequencies and percents are used to describe DSM-IV-TR criteria for ASD.

Table 2. Sample Characteristics/Demographics

AUT: Case status of Autistic Disorder; ASD: Case status of Autism Spectrum Disorder.

The frequency of DSM criteria 1a through d, 2a through 2d and 3a through 3d are compared in white versus African American children with a classification of ASD. We also compare by race the frequencies of all associated behavioral features including the key outcome variables of aggression, oppositionality, hyperactivity and temper tantrums. Simple logistic regression modeling and z-statistics are used to determine whether African American or white race is independently associated with outcome variables (DSM IV criteria for ASD and associated features including aggression, oppositionality, hyperactivity, and temper tantrums). In the interest of identifying clinical implications, we conducted analyses at a p-value of .05 to not overlook underlying trends. Significant differences between the 2 groups are concluded on the basis of a p-value < .01. Odds ratios adjusted for gender and diagnostic category were calculated for DSM Criteria and Associated Features.

Power analysis (Table 3) 

Table 3 provides estimates for minimum detectable group differences in core ASD symptoms and associated behavioral features in the setting of 202 African American children and 141 white children with 80% power and varying proportions (range, 50% to 95%) associated with ASD symptoms and behaviors.

Table 3. Power analysis

Back to Article Outline

Results 

Sample (Table 2) 

Data from 3 study years were combined for analysis. The 2002, 2006, and 2008 study years were combined for a sample size of 343 surveillance cases of ASD (see Table 2).

The racial distribution is 59% (n = 202) African American and 41% (n = 141) white. The African American group was composed of 82% males (n = 166) compared to 87% (n = 123) white males. In our sample, there were no more African American than white children diagnosed with autistic disorder vs other ASDs. This is noted because autistic disorder case status requires the presence of more DSM criteria and is generally regarded as being on the more symptomatic end of the spectrum. The sex ratio was 5 males to 1 female, which is slightly greater than previous studies (generally 4 males to 1 female) [17], [33], [34]. Using the Pearson χ² test, there was no significant association between the distribution of sex or ASD diagnostic type and race.

Age of first evaluation and diagnosis (Table 2) 

There were no statistically significant differences between white and African American children with ASD regarding the age of first evaluation, the age of earliest ASD diagnosis or the age of earliest evaluation confirming ASD diagnosis in this sample population.

DSM criteria for qualitative impairments in social interaction, communication, and repetitive/restrictive behaviors (Table 4) 

Contrary to our hypothesis, we did not find any significant differences in symptoms between the 2 groups under the DSM-IV criteria for qualitative impairments in social interactions. The p-values were >.01 for all DSM-IV criteria related to social impairment. Using simple logistic regression and the z-statistic, there are no significant differences in the presence of symptoms of impairments in social interaction among African American and white cases of ASD in our sample.

There was a trend toward significance in DSM-IV criteria 2a, Qualitative Impairments in Spoken Language, with 95% of African Americans and 88% of whites meeting this criterion with a p-value of .032. All other DSM criteria of qualitative impairments in communication (2b-2d) showed no significant differences between the 2 groups.

DSM-IV criteria 3 involve restricted interests and repetitive or stereotyped behaviors. In our sample population there were significant differences in criteria 3b (presence of nonfunctional routines/rituals) with 92% of whites meeting criteria versus 81% of African Americans with a p-value of .005. Criteria 3d (preoccupation with parts) also showed an increased proportion of whites (67%) than African American (50%) with a p-value of .002. There were no significant differences in criteria 3a (restricted interests) or 3c (stereotypical behaviors).

Odds ratios adjusted for gender and type of diagnosis showed no effect on significance of DSM-IV criteria for ASD based on gender or diagnostic type Table 5.

Table 4. Frequency of DSM Criteria Recorded as Present

DSM criteria present (all cases)	African American N = 202	White N = 141	p-value#
Social Impairment
1a–nonverbal behavior	169 (84%)	120 (85%)	.718
1b–peer relationships	159 (79%)	118 (84%)	.251
1c–spontaneous seeking	106 (52%)	77 (55%)	.697
1d–reciprocity	181 (90%)	132 (94%)	.200
Communication Impairment
2a–spoken language	191 (95%)	124 (88%)	.032
2b–receptive/pragmatic	176 (87%)	127 (90%)	.405
2c–repetitive language	146 (72%)	107 (76%)	.455
2d–imaginative play	103 (51%)	77 (55%)	.509
Restricted/repetitive behavior
3a–restricted interests	115 (57%)	88 (62%)	.310
3b–routines/rituals	164 (81%)	130 (92%)	.005∗
3c–stereotypies	134 (66%)	106 (75%)	.080
3d–preoccupation w/parts	101 (50%)	94 (67%)	.002∗

Table 5. Significant Odds Ratios Adjusted for Gender and Diagnostic Type

Table 6. Associated Behavioral Features Recorded as Present

Associated features (Table 6) 

Contrary to our hypothesis, race did not predict the presence of externalizing behaviors including oppositionality, hyperactivity, aggression, and temper tantrums. Logistic regression analysis showed no significant differences among African American and white children with ASD on the associated features of hyperactivity, oppositionality, aggression, or temper tantrums that are our key outcome variables.

There were also no significant differences between the 2 groups on the associated behavioral features of self-injury or abnormalities of sleep, mood, or cognitive development. There were significant differences in the 2 groups on the behavioral features of abnormal motor development and odd response to sensory stimuli. There were higher proportions of whites with ASD with abnormal motor development and odd responses to sensory stimuli noted as associated behavioral features compared to African American children with ASD in this sample population. There was a trend for significance in African American children with the behavioral feature abnormal level of fear with 66% of African American children and 55% of whites noted to have this feature. Data collection and coding does not specify whether behavioral features are abnormal due to hyporesponsiveness or hyperresponsiveness. Odds ratios adjusted for gender and type of diagnosis showed no effect on significance of behavioral features associated with ASD based on gender or diagnostic type (Table 5).

Back to Article Outline

Discussion 

This study investigated differences in core ASD symptoms and associated behavioral characteristics among African American and white children in a population-based cohort. Specifically, we found that compared to African American children with ASD, white children with ASD had more documented DSM-IV criteria of Restricted Interests and Repetitive/Stereotyped Behaviors including inflexible adherence to nonfunctional routines and preoccupation with parts. Compared to African American children with ASD, white children with ASD were also found to have more documentation of abnormal motor development and odd responses to sensory stimuli. Interestingly, these are all behaviors and criteria that relate to motor skills and atypical behaviors that are easier to observe by caregivers and clinicians. This could be attributed to either a difference in clinical symptom presentation or documentation differences among these 2 groups.

We found no differences in any of the recorded core social symptoms of ASD. This finding is similar to that of Cuccaro et al. [14] and Lord’s findings [15] in previous studies. Also similar to Cuccaro et al.’s first study, our population had more African Americans with documented impairments in communication and spoken language, although this was not statistically significant.

Our study methods applied the DSM-IV criteria for ASD to the documented symptoms of children in the abstracted records of this population-based sample. With application of DSM-IV criteria for ASD, it is possible to identify core social symptoms in both racial groups and use these to make a diagnosis of ASD. While clinicians’ records describe core symptoms (social, communication, and other), they do not always specify the experience of the clinician or if they are using a standardized tool to determine the presence of core symptoms, and it is possible that social impairment may be exaggerated or underappreciated. Although standardized diagnostic tools for ASD exist, they require specialized training and may be time-consuming. Clinicians may use more subjective, anecdotal criteria for diagnosis of ASD and may not be fully trained in the use of the DSM-IV or other diagnostic tools for ASD. This study provides a standardized assessment of recorded core social symptoms of ASD using objective methods in a larger population sample.

Since clinician observation plays a role in the assessment for ASDs, one may assume that attribute predilection might play a role in the diagnosis of ASD. By this we mean that a clinician may have inherent expectations that influence their evaluation and documentation of symptoms of ASDs and behaviors related to stereotype or bias. However, a study by Cuccaro and colleagues, which is germane to the notion of attribute predilection by the clinician, found no effect of race or ethnicity on professional perceptions of autism when clinicians were presented with case vignettes for diagnosis [35]. In their study, the variable that led to diagnostic confusion was found to be SES, not race. In their study, there were expectations that affected diagnosis of ASD that were related to SES, supporting the idea that clinicians may have expectations that affect diagnosis. But when the diagnostic criteria were applied correctly, with regards to race, clinicians were able to reach an accurate conclusion on the presence or absence of an ASD.

While continuing to address ongoing issues of racial disparities in care, it is important that we examine how we can identify children earlier. If there are differences in clinical symptom presentation among racial/ethnic groups we should incorporate these into our medical evidence base and use them to influence screening and diagnosis. In order to prevent bias in measurement and correctly document clinical symptoms, clinicians should develop and use standardized protocols and diagnostic tools for the evaluation of a possible diagnosis of an ASD. This will eliminate potential subjective differences among clinicians and lead to greater uniformity of diagnosis.

Despite these differences in some of the core symptoms of ASD, there was no significant difference in the age of diagnosis or first clinical evaluations among whites and African Americans in our sample population. This is a very positive change from previous studies where significant disparities in timing of diagnosis between African Americans and whites were notable. This change may have occurred as more clinicians become trained and skilled in the diagnosis of autism and evaluation services become more widely available due to a nationwide focus on autism awareness. However, it may be possible that this is due to the increased availability of resources and experienced clinicians in Philadelphia County compared to other regions.

Associated behavioral features are not part of the diagnostic criteria for ASD; however, they can complicate the clinical picture. As such it is important to explore whether there were any significant differences among our study sample that could influence clinical diagnosis of ASD. Our study did not show any association between race and symptoms of hyperactivity, aggression, oppositionality, or tantrum behavior. This is not consistent with previous studies that reported that African American children are diagnosed with ADHD or CD prior to being given a diagnosis of ASD [12]. What the literature has reported as potentially leading to diagnostic confusion does not hold true in this population based study as symptoms of ADHD and CD did not have any significant predominance in our sample of African American children with ASD. It is also becoming more established that ADHD can co-occur clinically with ASD and that these are not mutually exclusive diagnoses. In fact, a recent study using surveillance data found a 21% report of co-occurring ADHD [36]. Other studies have shown rates from 40% to 75% in smaller samples [37], [38], [39]. These conditions may be thought to be independent and viewed as coexisting conditions. Proposed revisions to the DSM 5 include removing ASD as an exclusionary criterion for the diagnosis of ADHD [40]. As more clinicians become aware of the presence of target externalizing behaviors presenting with ASDs, we hope that the diagnostic confusion with ADHD and delayed diagnosis of ASD will become less common.

Interestingly, we found that some behavioral features differed by race. Our data analysis did not specify the direction of some of these behavior abnormalities. The findings of increased documentation of odd responses to sensory stimuli and abnormal motor development in white children with ASDs should be explored to see if this is a documentation difference, whether it brings them to an evaluation sooner or has an effect on the presentation of their ASD. Children with motor delays may present earlier to a medical professional who may screen them for symptoms of ASD if clinically indicated. Children with odd responses to sensory stimuli may have behaviors that appear atypical or that are problematic and may lead to an earlier evaluation for an ASD. Future research can clarify if there are any specific sensory issues that are more clearly associated with an ASD.

Limitations 

One limitation of this study is that it utilized a surveillance dataset that relies on documentation and is limited to record review. We did not use direct evaluation performed on potential cases using standardized instruments such as the Autism Diagnostic Observation Schedule (ADOS) or the Autism Diagnostic Interview (ADI). However, while obtaining diagnostic information by direct observation is the gold standard, it is impractical and financially cumbersome in a large, population-based sample. If standardized direct observation tools were used for diagnostic evaluations in larger populations, good interrater reliability would need to be ensured. Some of the diagnostic tools for ASDs rely on parental reporting of symptoms. It is possible that reporting of symptoms may be affected by race as has been demonstrated in other developmental disorders such as ADHD. Further study is needed to explore whether there are racial differences in symptoms reported by parents of children with ASDs.

Our overall sample size was 439 surveillance cases of ASD, and included cases with incomplete data. Because we compared only whites and African Americans with ASD over the 3 study years that had full data available, our sample size was reduced to 343 surveillance cases of ASD. This sample population is still sizable compared to many studies that have examined racial differences in ASD. Neither sex nor diagnostic type was a moderator in the significance of ASD core symptoms or associated behavioral features as odds ratios adjusted for these variables had similar results in our analysis. Due to the multiple comparisons made in this study, these findings should be replicated to confirm significant differences.

Another limitation is that our data analysis did not specify the direction of some of these behavior abnormalities that were documented as associated behavioral features. For example, “odd responses to sensory stimuli” were coded as present or absent but did not specify if children had sensory preferences or aversions. Similarly, “abnormal level of fear”, which trended toward significance in African American children with ASD compared to white children with ASD did not specify if the level of fear was high or low. Future research should aim to clarify these associations between race and associated features. For example, if levels of fear are low, children may engage in more risk-taking behaviors and be considered impulsive, which is a diagnostic criterion for ADHD. If levels of fear are higher, these children may appear overly anxious to an evaluator, and anxiety may be noted as an associated feature.

Back to Article Outline

Conclusions 

In conclusion, this study suggests that race may have become less of a factor over time in accessing diagnostics services, at least in Philadelphia County, as age of diagnosis and type of diagnosis were similar in our 2 groups. There are some interesting differences in DSM-IV criteria and behavioral features that relate to atypical behaviors and motor skills that may suggest these are present at a higher frequency in white children with ASD compared to African American children with ASD. It may be that white children with ASD with these motor and behavioral abnormalities have more observable symptoms compared to African American children with ASD. However, with carefully applied DSM-IV criteria and use of appropriate protocols and tools, it appears that, in general, the diagnostic services that are available are appropriate for all.

Recommendations for research 

In addition to clarifying behavioral differences that may differ by race, future research should examine clinician-specific or testing environment issues that could influence racial differences noted in this study. For example, looking at the diagnostic tools used by clinicians or the type of clinician making a diagnosis among these 2 groups may reveal differences that may have an impact on the clinical diagnosis.

Future research can also look at whether treatment options may differ based on differences observed in this study. For example, African American children with ASD may require more language intervention and less intervention for atypical behaviors compared to white children with ASD. Along these lines, future research can test whether there are racial differences in follow up to diagnosis or type of treatment offered. If there are true differences in symptoms, there should be some differences in treatment and perhaps outcomes. If there are no differences, treatments and outcomes should be comparable.

Back to Article Outline

References 

1. CDC. Prevalence of Autism Spectrum Disorders–Autism and Developmental Disabilities Monitoring Network, 14 sites, US, 2002. Surveillance Summaries, February 9, 2007. MMWR. 2007;56(SS-1):
   • View In Article
2. CDC. Prevalence of Autism Spectrum Disorders–Autism and developmental disabilities Monitoring Network United states, 2006. In: Surveillance Summaries, December 18, 2009. MMWR. 2009;58(SS-10):2–21
   • View In Article
3. Kogan MD, et al. Prevalence of parent reported diagnosis of autism spectrum disorder among children in the US, 2007. Pediatrics. 2009;124(5):1395–1403
   • View In Article
   • CrossRef
4. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision. Washington, DC: American Psychiatric Association; 2000;
   • View In Article
5. Levy SE, Mandell DS, Schultz RT. Autism. Lancet. 2009;374:1627–1638
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (473 KB)
   • CrossRef
6. CDC. Mental health in the United States: Parental report of diagnosed autism in children aged 4-17 years: United States, 2003-4. May 5, 2006. MMWR. 2006;55(17):481–486
   • View In Article
7. Fombonne E, Simmons H, Ford T, et al. Prevalence of developmental disorders in the British nationwide survey of child mental health. J Am Acad Child Adolesc Psychiatry. 2001;32:12–23
   • View In Article
8. Liptak GS, Benzoni LB, Mruzek DW, et al. Disparities in diagnosis and access to health services for children with autism: data from the National Survey of Children’s Health. J Dev Behav Pediatr. 2007;29:152–160
   • View In Article
   • CrossRef
9. Yeargin-Allsopp M, Rice C, Karapurkar T, et al. Prevalence of autism in a US metropolitan area. JAMA. 2003;289:49–55
   • View In Article
   • MEDLINE
   • CrossRef
10. Lord C, McGee JP National Research Council, Committee on Educational Interventions for Children with Autism. Educating Children with Autism. Washington, DC: National Academies Press; 2001;
    • View In Article
11. Mandell D, Lesterud J, Levy S, et al. Race differences in the age at diagnosis among Medicaid eligible children with autism. J Am Acad Child Adolesc Psychiatry. 2002;41(12):1447–1453
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (116 KB)
    • CrossRef
12. Mandell D, Ittenbach R, Levy S, et al. Disparities in diagnosis received prior to a diagnosis of autism spectrum disorder. J Autism Dev Disord. 2007;37(9):1795–1802
    • View In Article
    • CrossRef
13. Wiggins L, Baio J, Rice C. Examination of the time between first evaluation and first autism spectrum diagnosis in a population-based sample. Dev Behav Pediatr. 2006;27(2):79–87
    • View In Article
14. Cuccaro M, Brinkely J, Abramson R, et al. Autism in African American families. Am J Med Gen. 2007;144B:1022–1026
    • View In Article
15. Lord C, Risi S, DiLavore PS, et al. Autism from 2 to 9 years of age. Arch Gen Psychiatry. 2006;63:694–701
    • View In Article
    • CrossRef
16. Mandell D, Novak M, Zubritsky C. Factors associated with age of diagnosis among children with autism spectrum disorders. Pediatrics. 2005;116:1480–1486
    • View In Article
17. Giarelli E, Wiggins L, Rice C, et al. Sex differences in the evaluation and diagnosis of autism spectrum disorders among children. Disabil Health J. 2010;3(2):107–116
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (359 KB)
    • CrossRef
18. Hillemeier M, Foster EM, Heinrichs B, et al. Racial Differences in parental reports of attention-deficit/hyperactivity disorder behaviors. J Dev Behav Pediatr. 2007;28(5):353–361
    • View In Article
    • CrossRef
19. Epstein J, Arch J, Conners CK, et al. Racial differences on the Conners Teacher Rating Scale. J Abnorm Child Psychol. 1998;26(2):109–118
    • View In Article
    • MEDLINE
    • CrossRef
20. Epstein J, et al. The role of children’s ethnicity in the relationship between teacher ratings of attention-deficit /hyperactivity disorder and observed classroom behavior. J Consult Clin Psychol. 2005;73(3):424–434
    • View In Article
    • CrossRef
21. Miller TW, Nigg JT, Miller RL. Attention deficit hyperactivity disorder in African American children: What can be concluded from the past ten years?. Clin Psychol Rev. 2009;29:77–86
    • View In Article
    • CrossRef
22. Arnold LE, Elliot M, Sachs L, et al. Effects of ethnicity on treatment attendance, stimulant response/dose, and 14 month outcome in ADHD. J Consult Clin Psychol. 2003;71(4):713–727
    • View In Article
    • CrossRef
23. Adebimpe VR, Chu CC, Klein HE, et al. Racial and geographic differences in the psychopathology of schizophrenia. Am J Psychiatry. 1982;139:888–891
    • View In Article
24. Delbello MP, Lopez-Larson MP, Soutullo CA, et al. Effects of race on psychiatric diagnosis of hospitalized adolescents: a retrospective chart review. J Child Adolesc Psychopharmacol. 2001;11(1):95–103
    • View In Article
    • MEDLINE
    • CrossRef
25. Muroff J, Edelsohn GA, Joe S, et al. The role of race in diagnostic and disposition decision making in a pediatric psychiatric emergency service. Gen Hosp Psychiatry. 2008;30:269–276
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (159 KB)
    • CrossRef
26. US Census Bureau. American Fact Finder. Available at: www.census.gov: http://factfinder.census.gov/servlet/QTTable?_bm=y&-qr_name=DEC_2000_SF1_U_DP1&-qr_name=DEC_2000_SF1_U_QTP1&-geo_id=01000US&-ds_name=DEC_2000_SF1_U&-_lang=en&-format=&-CONTEXT=qt
    • View In Article
27. Pennsylvania Department of Health. Health Statistics and Research. Vital Statistics 2002: Available at: http://www.portal.state.pa.us/portal/server.pt?open=514&objID=596032&mode=2
    • View In Article
28. Pennsylvania Department of Health. Health Statistics and Research. Vital Statistics 2006: Available at: http://www.portal.state.pa.us/portal/server.pt?open=514&objID=596032&mode=2
    • View In Article
29. Pennsylvania Department of Health. Health Statistics and Research. Vital Statistics 2008: Available at: http://www.portal.state.pa.us/portal/server.pt?open=514&objID=596032&mode=2
    • View In Article
30. Rice C, Baio J, Braun KV, et al. A public health collaboration for the surveillance of autism spectrum disorders. Paediatr Perinat Epidemiol. 2007;21:179–190
    • View In Article
    • MEDLINE
    • CrossRef
31. Van Naarden Braun K, Pettygrove S, Daniels J, et al. Evaluation of a methodology for a collaborative multiple source surveillance network for autism spectrum disorders-autism and developmental disabilities monitoring network, 14 sites, US, 2002. Surveillance Summaries February 9, 2007. MMWR. 2007;56(SS1):29–40
    • View In Article
32. Avchen RN, Wiggins LD, Devine O, et al. Evaluation of a records-review surveillance system used to determine the prevalence of autism spectrum disorders. J Autism Dev Disord. 2010;
    • View In Article
33. Tsia L, Beilser JM. The development of sex differences in infantile autism. Br J Psychiatry. 1983;142:373–378
    • View In Article
    • MEDLINE
    • CrossRef
34. Wing L. Sex ratios in early childhood autism and related conditions. Psychiatric Res. 1981;5:129–137
    • View In Article
35. Cuccaro ML, Wright HH, Rownd CV, et al. Brief Report: Professional perceptions of children with developmental difficulties: the influence of race and socioeconomic status. J Autism Dev Disord. 1996;26(4):461–469
    • View In Article
    • MEDLINE
    • CrossRef
36. Levy SE, Giarelli E, Lee LC, et al. Autism spectrum disorder and co-occurring developmental, psychiatric, medical conditions among children in multiple populations of the United States. J Dev Behav Pediatr. 2010;31(3):1–9
    • View In Article
    • CrossRef
37. Gillberg C. Deficits in attention, motor control, and perception: a brief review. Arch Dis Child. 2003;88:904–910
    • View In Article
    • CrossRef
38. Goldstein S, Schwebach AJ. The comorbidity of pervasive developmental disorder and attention deficit hyperactivity disorder: results of a retrospective chart review. J Autism Dev Disord. 2004;34:329–339
    • View In Article
    • MEDLINE
    • CrossRef
39. Holtmann M, Bolte S, Poustka F. Attention deficit hyperactivity disorder symptoms in pervasive developmental disorders: association with autistic behavior domains and co-existing psychopathology. Psychopathology. 2007;40:172–177
    • View In Article
    • MEDLINE
    • CrossRef
40. American Psychiatric Association. www.DSM5.org. Available at: http://www.dsm5.org/ProposedRevision/Pages/proposedrevision.aspx?rid=383
    • View In Article