Designed Cultural Adaptation and Implementation Quality in Substance Abuse Prevention: An Effectiveness Trial for the Keepin’ It REAL Curriculum

Introduction: Disseminating efficacious prevention programs is challenging, as taking programs to scale may involve less than ideal implementation practices as well as the need to adapt the program to new populations. Research in this area can help to elucidate a program’s impact in real world conditions and offer guidelines for adaptations that may be effective vs. ones to caution against. This study examines both designed cultural adaptation and implementation quality (IQ) processes during a trial of keepin’ it REAL (kiR), an evidenced-based middle school substance use prevention curriculum. The study examines IQ in urban (non-adapted) and rural (designed adaptation) versions of kiR as delivered to rural youth.

Methods: A total of 39 schools from rural school districts in Pennsylvania and Ohio participated and were randomly assigned to urban kiR treatment, rural kiR treatment, and control conditions. For the analysis, 2,178 survey responses were used (M = 14.71 years, SD = .60; 51% male; 97% European American) from students in participating schools at baseline in fall of 7th grade (2009) and again in 9th grade (2012).

Results: Based on observational ratings, each treatment group was coded as high or low IQ, yielding 5 conditions (e.g., control, low urban, high urban, low rural, and high rural). A series of analysis of covariance revealed significant differences in youth lifetime use of alcohol (p = .002), cigarettes (p = .000), marijuana (p = .018), and chewing tobacco (p = .002) among five conditions while controlling for baseline substance use. Pairwise comparisons showed that lifetime alcohol, cigarette, and chewing tobacco use were significantly lower among youth who received the designed adaptation (both low and high IQ rural version) than youth in the control condition. Further, the designed adaptation (rural) kiR, both low and high IQ versions, resulted in stronger effects on youth cigarette and chewing tobacco use than the non-adapted (urban) version, regardless of IQ, and stronger effects on alcohol and marijuana use than the low IQ urban version. There were iatrogenic effects for the low IQ urban condition with regards to lifetime marijuana use, as compared to youth in the control condition.

Conclusions: Findings highlight important issues related to programmatic changes both planned by the designer and those that occur during implementation. Advancing knowledge in these two domains contributes new evidence in prevention science that can aid both the program developer as well as the prevention community as interventions are scaled.