Mantente REAL (Keepin' it REAL) in Guatemala City: A Test of a Middle School-Based US Model Program in Vulnerable Neighborhoods

Purpose: This study tested the applicability and effectiveness of the keepin it REAL (kiR) manualized universal substance use prevention program, originally developed in the USA, for use in middle schools in neighborhoods of the capital of Guatemala with high levels of violence. The main goals of keepin’ it REAL are to increase youth’s drug resistance skills, enhance risk assessment and decision-making skills, and promote anti-substance use norms and attitudes.

Methods: A local Guatemalan non-profit NGO partnered with the original team that developed and tested kiR to identify and recruit Guatemala City schools in relatively more neighborhoods, train teacher-implementers, gauge fidelity, and collect pretest-posttest data on intervention outcomes.  Data come from 676 middle school children in 6^th and 7^th grade, from 12 middle schools.  Seven schools were recruited to implement a linguistically adapted version of the kiR substance use prevention program, Mantente REAL, and five similar schools served as a non-implementing comparison group. Teachers in implementing schools received a day-long training. Students in all schools completed pretests before the first kiR lesson was developed and a post-test after the 10^th and last lesson. Measures of relative intervention effectiveness included last 30 day frequency and amount of use of alcohol, cigarettes and marijuana; other anti-social and prosocial behaviors, substance use intentions, permissive norms, and positive expectancies; and use of the drug resistance strategies taught in the curriculum—Refuse, Explain, Avoid, Leave [REAL]. Tests of the relative effectiveness of kiR versus the comparison group were analyzed through baseline adjusted regression models in Mplus using FIML estimation to adjust for attrition (13%).

Results: The lifetime prevalence of substance use was low: 24% for alcohol, 11% for cigarettes and 3% for marijuana. Nevertheless, kiR students reported significantly smaller pretest-to-posttest increases in cigarette and marijuana use, relative to the comparison group, and non-significant changes in the same direction for alcohol use.  The kiR group also reported relatively more frequent use of the REAL drug resistance strategies, more prosocial behavior, and less positive expectations of substance use, but there were no significant differences by intervention condition on permissive substance use norms, use intentions, or antisocial behavior. 

Conclusions: The linguistically adapted version of kiR shows substantial promise in achieving notable prevention effects on certain substance use behaviors, attitudes and resistance skills in Guatemala City.  Further refinement may be needed to amplify and extend these effects to alcohol use and other attitudinal antecedents of substance use.