Assessing Family Outcomes of Early Intervention: Utility of Items Specific to Families of a Child with an Autism Spectrum Disorder

Background: The mission of state early intervention (EI) programs under IDEA is to provide services and supports to both children 0-3 with developmental delays and their families. Since 2005, states have been required to report on families’ perceptions of the extent to which EI helped the family achieve desired family outcomes. Families of a child with an autism spectrum disorder (“ASD families”) may have specific needs that are not addressed by currently used outcome measures.

Objectives: To compare factor structures and scores associated with responses of ASD and non-ASD families to two sets of items inquiring about the helpfulness of EI to families served. Using Concept Mapping, one set of items was developed with families of children with a range of developmental delays, the other set with families of a child with an ASD.

Methods: 134 ASD families and 75 families of a child with a developmental delay other than an ASD responded to a survey on the impact of EI services in New York. All survey items had the same 6 response options, ranging from “Very strongly disagree” to “Very strongly agree.” The 86 family-focused items to which all participants responded included 35 “general items” developed with stakeholder groups representing families of children with a variety of developmental delays, e.g., “EI services helped me and/or my family understand how to change what I'm doing to help my child as he/she grows,” and 51 “ASD items” developed by ASD-focused stakeholder groups, e.g., “Early intervention services helped me and/or my family learn how to explain my child's unusual behaviors to others.” Single- and multiple-factor models were fit to the survey response data for ASD and non-ASD families, separately, and for the general items and ASD items, separately. Additional analyses examined differences between ASD and non-ASD families in terms of mean item scores, item difficulty parameters, and item factor loadings.

Results: For both ASD and non-ASD families, for both the general and ASD items, a 1-factor solution showed good fit to the data, GFIs > .982. Increases in GFI for the multi-factor models were exceedingly small. Comparing item factor loadings for ASD and non-ASD families, the congruence coefficient was .993 for the general items and .994 for the ASD items, indicating high similarity of factor structures across the two groups of respondents. Both mean item scores and IRT item difficulties were highly similar across the ASD and non-ASD families.

Conclusions: No meaningful differences were found between ASD and non-ASD families in their responses to sets of general vs. ASD-focused items addressing the extent to which EI services helped families to achieve desired family outcomes. The findings suggest that a common set of items can be used with both ASD and non-ASD families to report on their perceptions of the helpfulness of EI to their family. Future analyses will address the range of perceived EI impact within each group in order to better understand how variables such as type and intensity of services provided affect family outcomes of EI.