It's who you know: Caregiver social networks predict service use among under-resourced children with autism

Background: Numerous studies have shown that racial/ethnic minority and under-resourced families face barriers that delay timely access to autism services. These barriers include lack of resources and information about autism, financial hardship, mistrust in the service system, cultural and language mismatch, and other factors that have yet to be identified. Method: The current study aimed to examine additional caregiver and system-level factors that could be associated with early service access using a diverse sample from four study sites (Los Angeles, CA; Philadelphia, PA; Sacramento, CA; and Rochester, NY). Partnering with community agencies that serve traditionally underrepresented groups, the research team recruited 118 caregivers of young children with autism who were low-income, English, Spanish or Korean speaking and had not accessed autism-specific services. Results: Regression analyses revealed that the total number of services accessed were associated with caregiver social network size (p = 0.011) but not by race, autism knowledge and caregiver agency. Among families receiving at least one non-autism specific service, a marginally significant interaction effect of site and primary language on total services received was observed (p = 0.06). Conclusion: Findings suggest that caregivers' social network connections are crucial in early service access, and future interventions could target increasing social networks to improve families' service engagement. More attention for non-English speaking families, especially those living in areas with few supports in their native languages, is needed.

Automated summary

The study found that the size of caregivers' social networks is crucial for early service access, and future interventions should aim to increase social networks to improve families' service engagement. More attention is needed for non-English speaking families, especially those living in areas with few supports in their native languages.