A cluster-randomized trial comparing two SWITCH implementation support strategies for school wellness intervention effectiveness

Abstract

Background: The School Wellness Integration Targeting Child Health (SWITCH) intervention has demonstrated feasibility as an implementation approach to help schools facilitate changes in students' physical activity (PA), sedentary screen time (SST), and dietary intake (DI). This study evaluated the comparative effectiveness of enhanced (individualized) implementation and standard (group-based) implementation.

Methods: Twenty-two Iowa elementary schools participated, with each receiving standardized training (wellness conference and webinars). Schools were matched within region and randomized to receive either individualized or group implementation support. The PA, SST, and DI outcomes of 1097 students were assessed at pre- and post-intervention periods using the Youth Activity Profile. Linear mixed models evaluated differential change in outcomes by condition, for comparative effectiveness, and by gender.

Results: Both implementation conditions led to significant improvements in PA and SST over time (p < 0.01), but DI did not improve commensurately (p value range: 0.02‒0.05). There were no differential changes between the group and individualized conditions for PA (p = 0.51), SST (p = 0.19), or DI (p = 0.73). There were no differential effects by gender (i.e., non-significant condition-by-gender interactions) for PA (p_{for interaction} = 0.86), SST (p_{for interaction} = 0.46), or DI (p_{for interaction} = 0.15). Effect sizes for both conditions equated to approximately 6 min more PA per day and approximately 3 min less sedentary time.

Conclusion: The observed lack of difference in outcomes suggests that group implementation of SWITCH is equally effective as individualized implementation for building capacity in school wellness programming. Similarly, the lack of interaction by gender suggests that SWITCH can be beneficial for both boys and girls. Additional research is needed to understand the school-level factors that influence implementation (and outcomes) of SWITCH.

Keywords: Children; Implementation science; Obesity prevention; Physical activity; School health.

Graphical abstract

Fig. 1

Consolidated Standards of Reporting Trials diagram showing the flow of participants through stages of cluster-randomized trial. avg. = average.

Fig. 2

YAP-estimated daily MVPA across 2 time points by condition (using multiple imputation analysis). Data are shown as mean ± SE. Indiv. = individualized; MVPA = moderate-to-vigorous physical activity; T1 = Time 1, baseline assessment; T2 = Time 2, post-test assessment.

Fig. 3

YAP-estimated daily sedentary behavior across 2 time points by condition (using multiple imputation analysis). Data are shown as mean ± SE. Indiv. = individualized; T1 = Time 1, baseline assessment; T2 = Time 2, post-test assessment; YAP = Youth Activity Profile.

Fig. 4

School-level association between tracking and behavior change (comprising physical activity, sedentary screen time, and dietary intake). YAP = Youth Activity Profile.