Predictors of Parent Expectations within an Intervention for Children with Autism

Background:  Past literature has emphasized the influence that family and child factors can have on parent expectations. Child variables, such as symptom severity, and family variables, such as ethnicity and socioeconomic status (SES), predicted parent expectations for children with oppositional behavior (Nock & Kazdin, 2001). These variables may be even more salient within ASD, as diagnosis rates and treatment access have been lower among ethnic minority families and families with low resources (Bhasin & Scendel, 2006; CDC, 2014; Liptak et al., 2008). Furthermore, parent expectations regarding the benefit of therapy can influence parent attendance and adherence (Nock et al., 2006; Nock & Kazdin, 2001). Therefore, understanding factors that may influence parents’ expectations may inform support strategies.

Objectives: The purpose of this study was to investigate how family characteristics (e.g., ethnicity, SES, intervention history) and child characteristics (e.g., ASD symptom severity) predict parent expectations at the start of intervention.

Methods:

Participants. Of 147 preschool children with ASD and their caregivers who were randomized into a larger intervention trial (Kasari et al., 2014), 113 with complete Parent Expectancies Questionnaire (PEQ: Kasari et al., 2015, adapted for this study from Nock & Kazdin, 2001) data were included in the present study (see demographics in Table1).

Measure. The PEQis a caregiver questionnaire that assesses parent expectations on a 5-point scale. Two outcome variables were derived based on factor analyses (Toolan, in prep) including: (a) belief in child improvement (BCI) including 3 items, and (b) treatment credibility (TC) including 8 items. Both BCI and TC are percentage scores where 100%=expectations for timely child improvement (BCI), and the treatment will lead to gains for my child (TC), while 0%=no expectation my child can improve (BCI), and the treatment will not lead to gains for my child (TC). One linear multiple regression model was conducted for each factor. The following parameters were included: SES score (Hollingshead Two-Factor Index of Social Position), ethnicity (dummy coded, white as reference group), MSEL developmental quotient, ADOS severity score, child’s early intervention experience (yes/no), and Parenting Daily Hassles frequency percentage score, controlling for site.

Results: Participants entered the intervention with high expectations for both BCI (M=72.27%, SD=15.65%) and TC (M=89.62%, SD=10.41%). No variables were found to predict TC scores. All racial groups demonstrated higher mean expectations scores than White participants. Participants reporting as African American or mixed/other had significantly higher BCI expectations than White participants (β=0.28, p=.030; β=0.35, p=.004 respectively). Early intervention experience was significantly and negatively related to BCI (β=-0.28, p=.010). There was no interaction between intervention experience and race/ethnicity.

Conclusions:

African American and mixed/other parents had higher expectations for intervention than White parents. Future research should not only examine factors that may influence these underlying differences, but also how these differences are related to treatment outcomes. Early intervention experience negatively predicted parents’ BCI. Future research may investigate the influence of parent satisfaction with intervention, type of service, and duration on expectancies.