Cost-effectiveness of Different Models of the Red Light, Purple Light! Self-Regulation Intervention

• In Event: Evaluating Preschool and School-Age Interventions That Promote Self-Regulation and Executive Function Skills in Children

Abstract

Children’s school readiness skills including self-regulation have been found to help children make a successful transition to formal schooling and predict later academic success (McClelland et al., 2013; Robson et al., 2021). Early childhood interventions can improve self-regulation, but there are few economic evaluations of such interventions (Jones et al., 2019). This study analyzed the cost-effectiveness of an early childhood self-regulation intervention (Red Light Purple Light!; RLPL), which has been found to improve children’s early self-regulation and academic skills (McClelland et al., 2019; Schmitt et al., 2015). We compared three different models of implementation across stages of intervention development: Model 1: trained research assistants (RAs; graduate students) directly delivered the RLPL intervention to children; Model 2; RAs trained trainers, who then trained teachers to implement RLPL with children (e.g., train-the-trainer); and Model 3: program faculty trained teachers to deliver the RLPL intervention to children.

Data were collected as part of four intervention studies examining the RLPL intervention in young children (Duncan et al., 2018; McClelland et al., 2019; Schmitt et al., 2015; Tominey & McClelland, 2011). The four studies fit three models of RLPL implementation. Model 1 included Tominey & McClelland (2011) & Schmitt et al., (2015). Tominey & McClelland (2011) included 65 children (n=28 intervention; n=37 control; 50% participating in Head Start) from nine classrooms. Schmitt et al., (2015) included 276 children from 14 Head Start classrooms (7 intervention, 7 control) with 33% (N = 88) English language learners (ELLs) assessed in Spanish and who received the intervention in Spanish.
Model 2 included Duncan et al. (2018), with 125 children from four schools (3 intervention, 1 control) in a large public school district. Model 3 included McClelland et al., (2019) with 157 children (52% female) from 13 Head Start classrooms (10 intervention, 3 control). Fifty-eight percent of the sample identified as Latino and 33% (N = 62) were ELLs. These children were assessed in Spanish but received the intervention in English following best practices for supporting Dual Language Learners. For all studies, children’s self-regulation was assessed using the Head-Toes-Knees-Shoulders (HTKS) measure, which has been shown to be reliable and valid with diverse populations (McClelland et al., 2014).
We implemented a cost-effectiveness analysis by calculating the incremental cost-effectiveness ratio. We also conducted a series of sensitivity analyses to adjust for parameter uncertainty. Our base-case analysis suggests that Model 2 was the most cost-effective strategy, in that a cost of $23 per child was associated with a one-unit increase of effect size on self-regulation scores (see Table 1). The “train-the-trainer” model remained the optimal strategy across scenarios in our sensitivity analysis, suggesting that the program could be more cost effective by using a “train-the-trainer” model.

This study fills an important gap in cost-effectiveness analyses on early childhood self-regulation interventions. Our process and results can serve as a model for future cost-effectiveness analyses of early childhood intervention programs and may ultimately inform decisions related to intervention adoption that optimize resource allocation and improve program design.

Authors

• Shauna L Tominey, Oregon State University - Corvallis
  Presenting Author

• Tao Li

• Alexis Tracy

• Megan M McClelland, Oregon State University - Corvallis