Study Notes on Standardized Screening and Diagnosis of Autism Spectrum Disorders

ABSTRACT

  • Objective:

    • Hypothesis: Routine, standardized screening for Autism Spectrum Disorder (ASD) may reduce racial/ethnic and socioeconomic status (SES) disparities in age of first diagnosis.

    • Study: Explored demographic differences in toddlers' age and performance on developmental measures at ASD assessment.

  • Method:

    • Participants:

    • Toddlers aged 16–39 months at evaluation, identified as at-risk for developmental delay using the Modified Checklist for Autism in Toddlers (M-CHAT) or M-CHAT-Revised (M-CHAT-R) and follow-up interviews.

    • Composition: 44.7% racial/ethnic minorities, 53.5% non-minorities.

    • Parent-reported variables:

      • Child race/ethnicity

      • Years of maternal education (MEd)

      • Household yearly income (YI)

  • Results:

    • Statistically significant but small correlations between MEd/YI and

    • Age at evaluation

    • Adaptive communication, socialization, and motor scores.

    • Minority racial/ethnic status did not significantly predict child's performance or likelihood of ASD diagnosis when controlling for MEd and YI.

    • Small effects on age at evaluation and receptive language skills were observed.

  • Conclusion:
    Routine screening and timely diagnostic evaluations can potentially reduce discrepancies in diagnosis timing among different demographic groups, but effects may be minimal.

AUTISM SPECTRUM DISORDERS (ASD)

  • Characteristics:

    • Serious impairments in communication and social interaction.

    • Presence of restricted interests and repetitive behaviors.

  • Prevalence: Current estimates show 1 in 88 children in the United States is diagnosed with an ASD.

  • Diagnosis Timing:

    • Strong evidence supports diagnosing ASD around the second birthday, with stable diagnostic outcomes.

    • Early interventions yield improved outcomes following diagnosis.

  • Demographic Variability:

    • ASD occurs across all racial/ethnic and socioeconomic groups.

    • Prevalence reports vary:

    • Similar prevalence in African-American and white children, but lower rates in Latino children.

    • With increasing SES, the prevalence of ASD also increases.

  • Factors Contributing to Disparities in Diagnosis:

    • Delays in well-child visit attendance

    • Limited English proficiency

    • Cultural differences in developmental milestone interpretations.

CURRENT STUDY AIMS

  • Expand understanding of demographic factors impacting age of ASD evaluation and developmental characteristics at diagnosis.

  • Hypothesis: Toddlers from underrepresented populations will have later evaluation ages due to existing disparities, although screening methods will mitigate these effects.

METHODS

Sample

  • Source: Ongoing study at University of Connecticut (UConn) and Georgia State University (GSU) focused on the psychometric properties of M-CHAT and M-CHAT-R.

  • Inclusion Criteria:

    • Completed M-CHAT or M-CHAT-R evaluations for ASD risk.

    • Provided at least one demographic variable (race/ethnicity, MEd, family income).

Participants Overview

  • Total screening participants: 35,204

    • M-CHAT: N = 18,989

    • M-CHAT-R: N = 16,215

  • At-risk toddlers requiring follow-up: 2,899

  • Diagnostic Evaluators:

    • M-CHAT(-R) screening could not assess refusals due to inconsistent documentation.

  • Missing Data:

    • Participants missing either race/ethnicity or development evaluation excluded.

    • Total diagnostic evaluations analyzed: N = 349.

Measures

  • M-CHAT (Modified Checklist for Autism in Toddlers):

    • 23-item parent questionnaire; identifies at-risk ASD behaviors.

    • Sensitivity: 0.97, Specificity: 0.95, Negative Predictive Value (NPV): 0.99.

    • Screening positive: 3 of 23 items or 2 of 6 critical items.

  • CARS (Childhood Autism Rating Scale):

    • Behavior rating scale to distinguish ASD from other delays.

    • Good interrater reliability and internal consistency values reported.

  • Mullen Scales of Early Learning:

    • Standardized assessment of cognitive abilities for ages up to 68 months.

    • High internal consistency and reliability.

  • Vineland Adaptive Behavior Scales:

    • Parent-report interview covering adaptive skills across various domains.

    • Established reliability and validity with moderate to high correlations.

Statistical Analyses

  • Power Analysis:

    • N = 349 provided adequate power for detecting medium effects.

  • Hierarchical Regression:

    • Examined child age at evaluation and scores based on demographic variables (MEd, YI, minority status).

  • Discriminant Function Analysis:

    • Assessed diagnosis likelihood between minority and non-minority children.

RESULTS

  • Mean Age at Evaluation: 25.72 months (SD = 4.48; Range 16.77–39.67).

  • Minority children presented for evaluation later than non-minority:

    • Mean: 26.28 months vs. 25.20 months; statistical significance at p = 0.025.

  • Correlation Findings:

    • MEd and YI significantly correlated (r = 0.60, p < 0.001).

    • Negative correlations with age evaluation:

    • YI: r = -0.14, p = 0.019

    • MEd: r = -0.14, p = 0.014

  • Regression Models:

    • Minority status predicted age at evaluation and receptive language measures while controlling for SES indicators.

  • No significant differences found in likelihood of ASD diagnosis based on minority status.

DISCUSSION

  • Investigation confirmed SES impacts on age at evaluation and child performance in measures of developmental abilities and symptom severity.

    • Minority status yields small, statistically significant relationships but likely not clinically meaningful.

  • Need for culturally aware diagnostic practices and continued attention to barriers minority families face during evaluations.

ACKNOWLEDGMENTS

  • Appreciation extended to participating families, physicians, and research teams from UConn and GSU.

  • Funding sources:

    • Current NIH Grant HD039961 and previous grants supporting the research.