Outcomes of Early Intervention Services for Families of Children with Autism Spectrum Disorders

Background:  State early intervention programs (EIPs) are charged with delivering evidence-based interventions for infants/toddlers with developmental delays and with providing supports and services to their families. Since 2005, states have been required to report annually on outcomes for families in their EIPs. Outcomes for families of children with autism spectrum disorder (ASD) are particularly important in that research indicates that these families may experience significant stressors related to parenting their child with ASD. The New York State Bureau of Early Intervention (NYSBEI) is one of approximately 25 state EIPs that collects EIP family outcome data using the Impact on Family Scale (IFS) developed specifically for this purpose by the National Center for Special Education Accountability Monitoring. The IFS was validated on an ethnically and geographically representative sample of families participating in EIPs across the U.S.; however, scale developers did not have diagnostic information needed for validation of the instrument for families of a child with ASD.

Objectives:  The primary purpose of this study was to validate the IFS for families of a child with ASD.

Methods:  Each year, NYSBEI mails a survey that includes the 35 IFS items (e.g., “EI services have helped me/ my family understand my child's special needs,” “. . . take part in typical activities for children and families in my community,” “. . . advocate for my child”) to representative samples of families exiting the EIP. Respondents rate their agreement with each item on a scale of 1 (very strongly disagree) to 6 (very strongly agree). Families’ responses were linked to the state’s EIP database to provide de-identified demographic and diagnostic information. Data were analyzed through the Rasch measurement framework using Winsteps software.

Results:  Valid item-response data were available for 2,877 families, including 165 families of a child with ASD. The IFS exhibited high item and person reliabilities (all greater than r=.94) both for families of a child with ASD and those whose child had a different diagnosis or developmental delay(s). Item calibrations estimated independently for the two subgroups correlated r =.98, indicating identical ordering of item difficulties. Only one IFS item, related to families’ ability to manage their child’s behavior, exhibited statistically significant differential item functioning (DIF), suggesting a greater challenge on the part of EIPs in helping families of a child with ASD in this area. A t-test of the difference between mean IFS measures for families of a child with and without ASD was not statistically significant, suggesting that NY’s EIP has similar levels of positive impact on families of children with ASD and those with children who have other developmental delays.   

Conclusions:  Results of this study indicate that measures on the IFS can be interpreted identically for families of a child with ASD and other families participating in EIPs, providing strong support for use of the IFS for statewide program evaluations as well as for planned studies that will model the mediating effect of family outcomes on outcomes of EIPs for children with and without ASD.