How to Score the M-CHAT-R
The M-CHAT-R is scored by counting the number of failed items (responses indicating concern), with a cutoff of ≥3 failed items on the initial screen triggering the follow-up interview, and ≥2 failed items after follow-up indicating high risk requiring referral for diagnostic evaluation 1, 2.
Scoring Algorithm
Initial M-CHAT-R Scoring
- Count failed items (responses that indicate potential ASD risk)
- The M-CHAT-R contains 20 yes/no questions completed by parents
- Each item has a designated "at-risk" response (either yes or no depending on the question)
Three-Tiered Risk Classification
Low Risk (0-2 failed items):
- No further action needed
- Continue routine developmental surveillance
- Represents 98.8% of screened children 3
Medium Risk (3-7 failed items):
- Administer the M-CHAT-R/F (Follow-Up interview) immediately
- The follow-up interview clarifies ambiguous responses and reduces false positives
- If ≥2 items remain positive after follow-up → refer for diagnostic evaluation
- If <2 items after follow-up → rescreen at next well-child visit
- This group represents approximately 6% of screened children 1
High Risk (≥8 failed items):
- Bypass the follow-up interview
- Refer directly for diagnostic evaluation and early intervention services
- This represents approximately 1% of screened children 1
- These children have 93% likelihood of requiring evaluation 1
Critical Implementation Points
The Follow-Up Interview is Essential
The follow-up M-CHAT-R/F interview is not optional for medium-risk children (3-7 items). Without it, the positive predictive value drops dramatically from 0.475 to 0.058 1. The interview:
- Clarifies parental understanding of questions
- Reduces false positives by 65-80% 1
- Can be administered by phone or web-based platform 4
- Should occur immediately after positive screen to avoid delays 4
Age-Specific Performance
Optimal screening window: 18-30 months 2
- 16-23 months: Lower PPV (0.28), higher false positives 1
- 24-30 months: Improved PPV (0.61), better accuracy 1
- 30+ months: Sensitivity increases but outside validated range 5
The tool performs poorly outside this window compared to age-specific alternatives 6.
Performance Characteristics
With Proper Two-Stage Administration (M-CHAT-R + Follow-Up):
- Sensitivity: 0.83 (95% CI: 0.75-0.90) 5
- Specificity: 0.51 (95% CI: 0.41-0.61) in high-risk populations 5
- PPV: 0.475 for ASD alone; 0.946 for any developmental disorder 1
- NPV: Consistently high (>0.90), reliably rules out ASD 6
Clinical Utility
When a child scores ≥3 initially and ≥2 after follow-up:
This means even "false positives" for ASD typically identify children needing intervention.
Common Pitfalls to Avoid
1. Skipping the Follow-Up Interview
- Increases false positives by 8-fold 1
- Creates unnecessary referrals and parental anxiety
- Only bypass for scores ≥8
2. Using Outside Optimal Age Range
- Screening at 12 months: sensitivity drops to 0.44 1
- Screening after 30 months: consider alternative tools 6
3. Misinterpreting Positive Screens
- Low PPV (0.06) in truly low-risk populations 5
- Consider co-occurring conditions: externalizing behaviors (ADHD) elevate scores independent of ASD 7
- Internalizing symptoms do not significantly affect scores 7
4. Lost to Follow-Up
- 41.2% of positive screens may refuse diagnostic follow-up 3
- 39.3% of screen-positive children were not evaluated in large studies 1
- Requires robust tracking systems and family education
Practical Scoring Summary
- Administer M-CHAT-R at 18 and 24-month well-child visits
- Count failed items (at-risk responses)
- 0-2 items: Low risk, routine surveillance
- 3-7 items: Conduct follow-up interview immediately
- ≥2 items remain positive → refer for evaluation
- <2 items → rescreen at next visit
- ≥8 items: Direct referral for diagnostic evaluation
The total score is more effective than alternative scoring methods (e.g., critical items only) 2. This two-stage approach reduces age of diagnosis by approximately 2 years compared to clinical detection without screening 2.