Comparing Colombian- and US-Based Cut Points for Measuring “High” Risk and “Low” Protection for Adolescent Drug Use

Introduction: Since 2012, the Nuevos Rumbos Corporation (Bogota, Colombia), in collaboration with researchers from the University of Washington (Social Development Research Group) and the University of Miami (Dept. of Public Health Sciences), has been implementing an adaptation of the Communities That Care (CTC) prevention system called Comunidades Que se Cuidan (CQC). As part of this system, school-based survey data have been collected and are used to identify community-level prevalence rates of “high” risk and “low” protection for youth drug use. These data are used to diagnose communities’ specific needs for preventive intervention and evaluate program outcomes. This study examined whether cut points in the distributions of risk and protective factors are equivalent between the United States and Colombia, and if not, what corrections need to be made in order to come up with a common assessment of “high” risk and “low” protection between the two countries.

Method: Data on risk and protective factor measures were collected from over 78,000 surveys of students between the ages of 12 and 19, located in 23 different Colombian municipalities. Following the methodology of Arthur et al., 2006, which identified the optimal cut-point as ± .15*median absolute deviation (MAD) in the distributions of risk/protective factors, we developed new cut points for the Colombian measures and compared them with cut points used in the United States.

Results: Comparison of grade-specific US and Colombian cut points for the Colombian data showed that some risk and protective factors (e.g., for Perceived Availability of Drugs and Parental Attitudes Favorable to Antisocial Behavior) had similar cut points in the distributions of the measures (e.g., MADs < .10), whereas other risk and protective factors (e.g., Low Commitment to School and Perceived Risks of Drug Use) exhibited very different cut points (e.g., MADs > .50).

Discussion: Colombia has currently a large number of surveys, but not all the areas of the country have been equally covered. Current survey efforts are underway to obtain a more complete national representation of Colombian youth. Moreover, additional analyses are in progress to determine the sensitivity and specificity of Colombian-generated cut points with regard to assessing the concurrent validity of risk/protective factors with measures of youth drug use and antisocial behaviors. Valid assessment of “high” risk and “low” protection for youth drug use is of paramount importance if prevention systems like CTC (in the US) and CQC (in Colombia) are to accurately diagnose communities’ needs for effective preventive intervention.