Assessing Autism in Mexico Through the Autism Behavior Checklist (ABC)

Background:

The alarming increase in the prevalence of autism spectrum disorders ASD (1 in every 80 in USA, 2006), makes early detection very important. Therefore, the use of easy screening methods, which do not require special training, with high reliability and sensitivity are needed. Even though Mexican versions of M-CHAT and SRS have been published,  instruments oriented towards the educational settings are yet to be validated (Albores et al. 2011, Fombonne, 2011).

The Autism Screening Instrument for Educational Planning (ASIEP) developed by Krug et al. (1980), has five components which measure: behavior, vocalization, interaction, learning rate and educational status. The Autism Behavior Checklist (ABC) is one of the five components of the ASIEP developed to assess symptoms in individuals with autism spectrum disorders (ASD) to be used in a school setting. The ABC provides different profiles for individuals from 18 months to 35 years of age may be answered by teachers or parents and does not require specialized training and has an adequate validity and reliability  (Krug, Arick, & Almond, 1980).

Objectives:

To evaluate the validity and reliability of the ABC Spanish version.

Specific Objectives:

To evaluate the Spanish ABC :

•  internal consistency (Cronbach alpha).
  • test- retest reliability.
  • convergent validity with the Autism Diagnostic Interview- Revised (ADI-R).
  • construct validity.

Methods:

Participants were 133 children (aged 2-17), with a presumptive diagnosis of autism. After translating and backtranslating the ABC all parents answered the checklist,  the ADI-R and a semi-structured interview to confirm an ASD diagnosis.  

The sample for the ABC test-retest analysis consisted of 19 parents with unaffected typically developing children between 2 and 17 years old.

Results:

A total of 133 children and adolescents with a mean age of 6.9 years (SD 3.7), age range of 2-17 years, 83.5% were males (n=111)  and 16.5% were females (n=22). The ABC total mean score was 69.3 (SD 25). Males had the highest scores in all subscales.  The Internal consistency was through the Cronbach Alfa coefficient which was: α= .83 <p=.001 for the 57 ABC items.

The 10 day test-retest reliability showed a Pearson correlation coefficient of r = .98, p < 0.001 and r < = 1.00., p< 0.001.

The Spearman correlation coefficients between the ABC subscales and the ADI-R ranged from (rˢ=.494) to (rˢ=.816).

Criterion validity with a cutoff of 30 resulted in a sensitivity of 87% and a specificity of 37%. The best Kappa was .285 between the ABC and the ADI-R.

Conclusions:

In this study we investigated the psychometric properties of the ABC Spanish adaptation version. As with other instruments the cutoff points proposed by the authors (Krug, Arick, & Almond, 1980) resulted in an underestimation of autism symptoms in women, leaving some females not classified as autistic.

Mexican parents show confusion when answering the checklist especially when a chosen item had a “high weight” that they thought did not reflect the “symptom severity.”  All difficulties were resolved by removing the assigned weight and asking them to rate each item in a true/false option.