What are the appropriate age‑specific screening questions for autism spectrum disorder?

Screening Questions for Autism Spectrum Disorder

For children aged 18-30 months, use the Modified Checklist for Autism in Toddlers (M-CHAT) or its revised version (M-CHAT-R/F), which consists of 23 parent-completed questions, with any positive screen requiring a follow-up interview before proceeding to full diagnostic evaluation. 1

Age-Specific Screening Approach

18-24 Months (Primary Screening Window)

The M-CHAT-R/F is the most extensively studied and recommended tool for this age group 1, 2. The screening focuses on seven key discriminating questions that address:

• Joint attention behaviors:

  • Protodeclarative pointing (pointing to show interest)
  • Gaze monitoring (following where you point)
  • Showing objects to parents

• Social relatedness:

  • Interest in other children
  • Imitation abilities
  • Response to name

• Pretend play abilities 3, 4

Critical implementation detail: A positive screen requires a structured follow-up interview. Only after a positive follow-up interview should you refer for comprehensive diagnostic evaluation 1. This two-stage approach dramatically improves positive predictive value from 0.06 in low-risk populations to 0.53 in high-risk populations 5.

12-17 Months (Earlier Detection)

While the USPSTF found insufficient evidence for universal screening before 18 months 1, emerging tools show promise for earlier identification:

• First Year Inventory (FYI) can be used at 12 months, though sensitivity is modest (detecting 4 of 9 children later diagnosed) 2
• Communication and Symbolic Behavior Scales (CSBS) developmental checklist detects broader developmental delays including ASD risk 2
• Recent validation of screening tools in Chinese populations suggests feasibility of 12-24 month screening 6

Important caveat: Screening before 18 months has higher false-negative rates due to the gradual emergence of symptoms across the first two years 2.

30 Months and Beyond

For children not previously screened or with emerging concerns:

• M-CHAT-R/F maintains good performance, with sensitivity actually improving at 30 months compared to 24 months 5
• Combined use of Social Communication Questionnaire (SCQ) and Social Responsiveness Scale-2 (SRS-2) shows excellent accuracy in children 30-60 months (sensitivity 83.3%-94.9%) 7

Practical Screening Algorithm

Step 1: Administer M-CHAT-R/F at 18 and 24-month well-child visits

Step 2: Score immediately using these failure criteria:

• Failing ≥2 of the 7 key questions (highest sensitivity: 93.1%) 3, OR
• Failing ≥6 of all 23 questions (better specificity: 84.8%) 3

Step 3: If either criterion met, conduct structured follow-up interview (M-CHAT-R/F includes this)

Step 4: If follow-up interview positive, add brief observational assessment:

• Eye contact during appointment
• Ability to follow a point
• Pretend play demonstration
• Protodeclarative pointing to objects

Failing ≥2 of these 4 observational items has 73.6% sensitivity and 91.2% specificity 3

Step 5: Refer for comprehensive diagnostic evaluation if both parent questionnaire AND observation are positive

Critical Pitfalls to Avoid

Do not rely on M-CHAT alone in low-risk populations without the follow-up interview—positive predictive value drops to only 6% 5. This means 94% of positive screens would be false positives without the follow-up step.

Account for age when interpreting scores: The same cutoff performs differently at different ages. Sensitivity is lower at 24 months compared to 30 months 5.

Recognize cultural and language variations: Screening tools validated in Western populations may require adjusted cutoffs for other populations 3, 6.

The USPSTF assigns an "I" (insufficient evidence) grade for universal screening in children without parental or clinician concerns 1. However, this reflects lack of outcome studies on screened populations, not evidence of harm. Given that harms are minimal and early intervention improves outcomes, screening remains standard practice despite the insufficient evidence grade 1.

For children with developmental concerns or high-risk factors (siblings of children with ASD, developmental delays), screening accuracy improves substantially and should definitely be performed 5, 4.