Location: Delta Obesity Prevention ResearchTitle: Assessment of program implementation for the Mississippi Communities for Healthy Living (MCHL) Nutrition Intervention) Author
Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 2/1/2014
Publication Date: 5/24/2014
Citation: Huye, H., Landry, A., Resavy, M., Crook, L., Connell, C. 2014. Assessment of program implementation for the Mississippi Communities for Healthy Living (MCHL)Nutrition Intervention [abstract]. Proceedings of the International Society of Behavioral Nutrition and Physical Activity. p. 381. Interpretive Summary:
Technical Abstract: The purpose of this study was to assess program implementation (PI) using an implementation score composed of process evaluation (PE) components reach, dose delivered, dose received, and fidelity. This 9-month, multi-site nutrition intervention consisted of two treatment groups (N=319). Sixteen sites in 11 counties of Mississippi met monthly for five nutrition education sessions and one review session delivered by three educators indigenous to the area. Educators were provided extensive training related to lessons and the theoretical foundation prior to and midway through the intervention. The implementation score was derived from attendance records, participant evaluations, educator fidelity evaluations, and educator self-reports that were maintained for each session. These measures composed the four PE component scores that were averaged to obtain a PI score; an a priori PI score of 80% was set as a target to indicate acceptable implementation. The overall PI score was 80% with 76% (n=243) of the participants reached, averaging three (M=3.4) out of five sessions attended. Dose delivered was scored at 95% while dose received was scored at 64% and fidelity at 87%. Process evaluation allows researchers to identify factors that contribute to the success or failure of multi-site interventions with significant distance between research staff and sites. While extensive educator training and feedback may have resulted in a higher PI score, educator bias related to dose delivered measures may have contributed to elevated scores. Furthermore, data analysis methods for multiple data sources used for several sessions with multiple sites should be pilot tested.