At what age, 18 months or 2 years, is the Modified Checklist for Autism in Toddlers (MCHAT) most effective for screening Autism Spectrum Disorder (ASD) in a child?

M-CHAT Screening: 18 Months vs 2 Years

Screen at both 18 and 24 months, but expect significantly better performance at 24 months, with the 24-month screening showing more than double the positive predictive value (PPV 0.61) compared to 18-month screening (PPV 0.28) in low-risk populations. 1

Recommended Screening Schedule

The American Academy of Pediatrics recommends universal ASD-specific screening at both 18 and 24 months using the M-CHAT-R/F. 2 This dual-timepoint approach is critical because:

• 18-month screening detects early-onset cases and children who may regress between 18-24 months 3
• 24-month screening captures children with later symptom emergence and provides more reliable diagnostic accuracy 2

Age-Specific Performance Characteristics

At 18 Months (16-23 months)

• Positive predictive value: 0.28 (only 28% of screen-positive children have ASD) 1
• Sensitivity ranges from 0.44-0.93 depending on population 2
• Higher false-positive rates due to overlapping behaviors between typically developing toddlers and early ASD 2
• Limited evidence supporting accuracy at this age 4

At 24 Months (24-30 months)

• Positive predictive value: 0.61 (61% of screen-positive children have ASD) 1
• Diagnosis at ≥24 months is well-established and stable 2
• Sensitivity improves at 30 months compared to 24 months 4
• Meta-analysis shows pooled sensitivity of 82.6-83% and specificity of 45.7-94% when follow-up interview is used 1

Critical Two-Stage Screening Protocol

Stage 1: Initial M-CHAT-R Questionnaire

• Score ≥3 on initial screening triggers follow-up interview 1, 5
• Score ≥7 indicates high risk: skip follow-up and refer immediately for diagnostic evaluation 1, 2

Stage 2: Follow-Up Interview (Essential)

• Score ≥2 after follow-up interview indicates 47.5% risk of ASD and 94.6% risk of any developmental delay 5
• The follow-up interview is never optional for medium-risk scores (3-6 points), as it dramatically improves PPV from 5.8% to 43-65% 1

Key Clinical Pitfalls

Common Error: Skipping 18-Month Screening

• While 24-month screening performs better, 18-month screening captures regression cases and allows earlier intervention 3
• Children screened with M-CHAT-R/F were diagnosed 2 years younger than the national median age of diagnosis 5

Common Error: Not Performing Follow-Up Interview

• Without follow-up, PPV drops to only 5.8% in low-risk populations 1
• Specificity improves from moderate to approximately 0.99 when follow-up is completed 2

Common Error: Relying on 18-Month Results Alone

• Meta-analysis shows lack of evidence supporting M-CHAT performance at 18 months in low-risk children 4
• Clinicians must account for child's age when interpreting scores 4

High-Risk Populations Require Intensified Surveillance

For siblings of children with ASD (14-18% risk):

• Screen at minimum at both 18 and 24 months 2
• Consider additional screening timepoints given elevated baseline risk 2

Evidence Quality Note

The USPSTF (2016) assigned an "I" grade (insufficient evidence) for universal screening in children without parental or clinician concerns, noting lack of studies on clinical outcomes of screen-detected children. 6 However, the American Academy of Pediatrics continues to recommend universal screening at both timepoints based on the tool's ability to detect developmental delays broadly (94.6% of screen-positive children) and facilitate earlier intervention. 5