International Meeting for Autism Research: M-CHAT & STAT: The Effectiveness of Multi-Level Screening for ASD

M-CHAT & STAT: The Effectiveness of Multi-Level Screening for ASD

Thursday, May 12, 2011
Elizabeth Ballroom E-F and Lirenta Foyer Level 2 (Manchester Grand Hyatt)
2:00 PM
M. Khowaja, D. L. Robins and L. B. Adamson, Georgia State University, Atlanta, GA
Background:  A variety of screening tools identify children who are at risk for autism.  Level 1 tools are used in unselected samples, but tend to have high false positive rates. Reducing the number of false positives cases will decrease the delay in receiving intervention services for true positive cases. 

Objectives:  This study seeks to measure whether a multilevel screening method, using the STAT (Level 2) following screen positive results on the M-CHAT and Follow-up Interview (FUI; Level 1), will reduce false positives without significantly increasing the number of missed cases.  Additionally, this study seeks to replicate the effectiveness of the STAT with children younger than 24 months of age.

Methods: Parents in the metro-Atlanta area completed the M-CHAT at their child’s well-baby visits (n=10,440); 830 screened positive and 118 continued to screen positive on the FUI.  A subsample of 54 children completed both a STAT (Level 2) and a diagnostic evaluation. STAT cutoffs of 2.00 for children ≥ 24 months and 2.75 for children < 24 months were based on recommendations in Stone, McMahon, and Henderson (2008). 

Results:  Thirty-four cases screened positive on the STAT and 25/34 received an ASD diagnosis, yielding a PPV of .74 for this multilevel screening method, compared to previously published data on the M-CHAT/FUI, whose PPV ranged from .57 to .65 (Kleinman et al., 2008; Robins, 2008).  Five of the 20 children who screened negative on the STAT had ASD, reducing the sensitivity of the multilevel method compared to M-CHAT/FUI alone.

ROC analysis for the subsample of children who completed a STAT before 24 months and also received a diagnostic evaluation (n=46) yielded an area under the curve (AUC) of .79.  An optimal STAT cutoff score for this age group was 2.00: sensitivity=.88 and specificity=.55.  Psychometric properties of the STAT for children ≥ 24 months was also high (AUC=.88), and the recommended cutoff of 2.00 was supported: sensitivity=.91 and specificity=.54. 

Reanalysis of the data for two-level screening using a STAT cutoff of 2.00 for all children resulted in 40/54 screen positive STATs.  Of these 40 cases, 27 received an ASD diagnosis, yielding a PPV of .68, similar to published data on the effectiveness of the M-CHAT alone for screening.  

Conclusions:  The lack of differential cutoffs by age in the current study compared to the Stone et al.’s (2008) study may be due to the qualitative differences in the sample (siblings of children with ASD vs. at-risk children from general population).  However, the higher PPV for two-level screening when using age-based STAT cutoffs (2.00 vs. 2.75) is promising, although it reduces sensitivity.  Efforts must continue to reduce the false positive rate without significantly increasing the number of missed cases.  For example, approaches to identify children who urgently need an evaluation and autism-specific early intervention may be combined with approaches to reduce the false positive rate for children whose scores are not as high. Empirical studies are needed to inform policy decisions on the early detection of ASD to shorten the delay in receiving appropriate treatment.

| More