Brief report: large individual variation in outcomes of autistic children receiving low-intensity behavioral interventions in community settings

Yoko Kamio, Hideyuki Haraguchi, Atsuko Miyake, Mikio Hiraiwa

Despite widespread awareness of the necessity of early intervention for children with autism spectrum disorders (ASDs), evidence is still limited, in part, due to the complex nature of ASDs. This exploratory study aimed to examine the change across time in young children with autism and their mothers, who received less intensive early interventions with and without applied behavior analysis (ABA) methods in community settings in Japan. Eighteen children with autism (mean age: 45.7 months; range: 28-64 months) received ABA-based treatment (a median of 3.5 hours per week; an interquartile range of 2-5.6 hours per week) and/or eclectic treatment-as-usual (TAU) (a median of 3.1 hours per week; an interquartile range of 2-5.6 hours per week). Children's outcomes were the severity of autistic symptoms, cognitive functioning, internalizing and externalizing behavior after 6 months (a median of 192 days; an interquartile range of 178-206 days). In addition, maternal parenting stress at 6-month follow-up, and maternal depression at 1.5-year follow-up (a median of 512 days; an interquartile range of 358-545 days) were also examined. Large individual variations were observed for a broad range of children's and mothers' outcomes. Neither ABA nor TAU hours per week were significantly associated with an improvement in core autistic symptoms. A significant improvement was observed only for internalizing problems, irrespective of the type, intensity or monthly cost of treatment received. Higher ABA cost per month (a median of 1,188 USD; an interquartile range of 538-1,888 USD) was associated with less improvement in language-social DQ (a median of 9; an interquartile range of -6.75-23.75). To determine an optimal program for each child with ASD in areas with poor ASD resources, further controlled studies are needed that assess a broad range of predictive and outcome variables focusing on both individual characteristics and treatment components.