Engaging Toddlers with Autism: Effective Caregiver Strategies

Background: Young children with ASD demonstrate differences in their use of joint attention and social play when engaging with others. Such behaviors make it difficult for caregivers to engage their children.

Caregiver mediated interventions have shown success in teaching parents strategies to help their children increase their use of joint attention initiations and responses (Kasari et al., 2010; Rocha, Schreibman, & Stahmer, 2007) and their engagement with others (Kasari et al., 2010). Many of these interventions utilize an individualized approach within the framework of the treatment model.  One question is the extent to which caregivers enter treatment with mastery of specific strategies.  Thus the current study examined caregiver strategy use prior to intervention.

Objectives: The purpose of this study was to describe and compare caregiver strategies while interacting with their children with ASD within supported joint and object engagement states.

Methods: The sample included 85 toddlers with ASD (70 males, 15 females) and their caregivers. The mean CA of the children was 31.62 months. Mean receptive language age on the Mullen was 18.24 months, and the mean expressive language age was 17.33 months. Children were recruited from an early intervention program at UCLA. Coding: Caregiver-child interactions were coded using symbol-infused engagement states (Adamson et al., 2004). This system measures the percentage of time in various engagement states between caregivers and their toddlers. Two states (object and joint engagement) were calculated given their importance in previous studies of caregivers and toddlers with autism (Kasari et al., 2010). Caregiver strategies during the MCX were coded, capturing five different strategies used by caregivers during interactions with their children. These strategies included: 1) caregiver directiveness 2) play level 3) maintaining strategies 4) activity level 5) gestural behaviors. Each variable was coded within the longest period of supported joint engagement and object engagement for each dyad.

Results: Exact McNemar tests were used to identify differences in the proportion of caregivers using each strategy between the object engaged and supported joint engaged states. Caregiver play level (p<.001), maintenance of their child’s attention (p<.001), and caregiver activity level (p<.001) were significantly different between the two engagement states. A greater proportion of caregivers were able to match their child’s play level during supported engagement than in the object state. Second, few caregivers attempted to redirect their children’s attention during the supported joint state, while many tried to do so in the object state. Parents also tended to be less active in interactions with their object focused children. Verbal directiveness and gestural behaviors were not different between engagement states. Overall, children spent more time in object engaged states, and parents had difficulty modeling joint attention skills and playing at their child’s level.

Conclusions: These findings suggest that playing at the child’s level, having a balanced activity level, and maintaining the child’s attention may be important components of sustaining joint engagement. Additionally, tailoring the intervention to the existing skills of the caregiver will likely be more successful than using the same “one shoe fits all” approach to intervention.