Process evaluation of the IDEFICS school intervention: putting the evaluation of the effect on children's objectively measured physical activity and sedentary time in context

Abstract

SummaryObjectives: The first aim was to describe the intervention implementation and reception with specific regard to physical activity (PA) within Belgian schools participating in the IDEFICS (Identification and prevention of Dietary-and lifestyle-induced health EFfects In Children and infantS) intervention. Second, we examined if a higher intervention process score in schools showed more favourable effects on children's objectively measured PA and sedentary time (ST). Methods:The IDEFICS intervention focused on improving children's health behaviour (including PA) at the community, school/kindergarten and family level. Five process evaluation questionnaires completed by school working groups, and (physical education) teachers were used for the process evaluation of four intervention modules (school working groups, healthy weeks, school environment, health-related physical education and PA). Items were summed to calculate a total intervention process score per school. Schools were subdivided into three groups (low, medium and high score). Multilevel models were used to examine if changes in PA or ST differed between schools with a low, medium or high score. In total, 333 children (54.1% girls, mean age 6.0 ± 1.5 years) from 34 intervention schools (18 pre-schools and 16 primary schools) in the town of Geraardsbergen, Belgium, provided valid accelerometer data for two weekdays and one weekend day.Results: Mean intervention process score (maximum value = 44) was 20.9 ± 5.8 for schools. The breakdown per module showed which components were implemented and received well and which components could have been improved. After the intervention, the decrease in light PA and increase in ST was much higher in children from primary schools with a low intervention process score, whereas the behaviours remained relatively stable in children from primary schools with a medium or high score. The change in moderate to vigorous PA did not differ significantly between schools with a low, medium or high score.Conclusion: The IDEFICS intervention was relatively able to prevent unfavourable changes in ST and light PA in schools with a medium and high intervention process score. Further process evaluation research is needed to obtain a more profound picture of the IDEFICS intervention process.

Effect and process evaluation of a kindergarten-based, family-involved cluster randomised controlled trial in six European countries on four- to six-year-old children’s steps per day: the ToyBox-study

“…The results from the current study are very relevant, as our results emphasise the major importance of linking process to effect evaluation. These results are comparable to what has already been found in a recent study by Verloigne et al (2015) in Belgian children (mean age: 6.0 years old) in which favourable effects were found for primary schools with medium and high process evaluation scores of the IDEFICS-intervention, but these effects were not found in preschool children [ 37 ]. Other studies found similar results with favourable effects in the targeted behaviours in adolescents with a higher level of intervention implementation [ 38 – 40 ].…”

“…Although it was based on a method used in previous studies [ 37 , 49 , 50 ], there are still some limitations linked to the method used to calculate the process evaluation scores as not all components of Saunders et al (2005) were used to calculate the score, and equal weights for each component were used. In addition, there is not yet a consensus on how to operationalise the different process evaluation components [ 37 , 51 ]. Consequently, the process evaluation scores provide a more general idea of the level of process evaluation of the kindergarten teachers and parents/caregivers.…”

Effect and process evaluation of a kindergarten-based, family-involved intervention with a randomized cluster design on sedentary behaviour in 4- to 6- year old European preschool children: The ToyBox-study

“…The calculation of the process evaluation scores was theory-based, as key elements from the process evaluation model of Saunders et al [32] were used. This method was also used elsewhere [49, 50]. However, it should be acknowledged that there is no standardized method to calculate process evaluation scores.…”

Assessing the acceptability of an adapted preschool obesity prevention programme: ToyBox‐Scotland

“…Participants' responses to post intervention feedback surveys were coded using a similar approach to that employed by Verloigne et al () and Pinket et al (). Specifically, positive Likert scale responses (responses of either agree or strongly agree ) were coded as 1, whereas nonresponses or negative responses ( neither agree nor disagree , disagree , or strongly disagree ) were coded as 0.…”