Factors Associated with Intervention Uptake in Community Practice: Acceptability, Feasibility, and Implementation Climate

Background: A research-to-practice gap exists in the use of evidence-based interventions for children with autism in community practice (e.g., Lord et al., 2005). Critical to disseminating evidence-based interventions from ‘bench-to-bedside’ is understanding the delivery context. Provider ratings of an intervention’s acceptability and feasibility have been found to influence intervention use (Proctor et al., 2011). Implementation climate, or the extent to which an intervention is perceived by users as supported and rewarded in their work setting (Klein & Sorra, 1996) is also associated with intervention use (e.g., Dingfelder, 2012). By identifying factors associated with intervention uptake, we can bridge the research-to-practice gap and equip community providers with interventions that are effective andappropriate for the settings in which they are delivered.

Objectives: The purpose of the study was to examine factors contributing to intervention uptake among community providers. Specific objectives were to: 1) compare provider ratings of intervention acceptability, feasibility, and work setting implementation climate from post-training to 3-month follow-up; 2) compare ratings of feasibility, acceptability, and implementation climate between providers who are and are not using the intervention, and 3) examine predictors of intervention use.  

Methods: Community providers from geographically and ethnically diverse communities across Washington State attended one-day workshops to learn Reciprocal Imitation Training (RIT), an evidence-based intervention for children with autism (Ingersoll, 2008). Providers (n=66) rated the acceptability and feasibility of RIT (URP-I, selected items; Chafouleas, 2009) and the implementation climate of their work setting (PICS, selected items; Dingfelder 2012) immediately post-training and at a 3-month follow-up. At follow-up, providers also reported whether they had used RIT with children in their caseload.

Results: Three separate repeated measures ANOVAs were conducted to examine intervention feasibility, acceptability, and climate. All three ANOVAs revealed significant main effects for time (post-training vs. 3-month follow-up), RIT use (Use vs. No-Use), and their interaction, ps<.02 (see Table 1). Post-hoc analyses were conducted using Bonferroni’s adjustment. At post-training, the Use and No-Use groups were comparable on feasibility, acceptability, and climate ratings, ps=.11-.16. From post-training to 3-month follow-up, all three ratings declined significantly for the No-Use group, and acceptability ratings also declined for the Use group, ps<.01. At the 3-month follow-up, the Use group reported significantly higher feasibility, acceptability, and climate ratings than the No-Use group, ps<.01.

Logistic regression indicated that higher intent to use RIT at post-training increased the likelihood of RIT use at the 3-month follow-up, OR=4.07, p=02. A significant relation was found between provider background and use of RIT; therapists such as speech-language pathologists were more likely to use RIT than were early childhood/special educators, X²(2,n=62)=6.03, p<.05. 

Conclusions: Immediately following training, all community providers perceived RIT as acceptable, feasible, and supported by their workplace.  While these ratings remained high for those providers who were using RIT, they decreased over the subsequent 3 months for those who were not using RIT.  Intent to use RIT following the training was a significant predictor of intervention use 3 months later.