Abstract
Latin America has high rates of drug and alcohol consumption by young people (Inter-American Drug Abuse Control Commission, 2015), with rates rapidly increasing during the adolescent years. In Mexico, the 2014 National Survey of Drug Use among Students (n=191,880; Villatoro et al, 2015) revealed that the prevalence of drug use increases exponentially from primary (3.3%) to middle (10.9%) and high school (26.2%). The same pattern is observed for alcohol binging (defined as 5 or more drinks in one sitting) in the last year: prevalence was 2.4% in primary, 8.9% in middle and 27.3% in high school. The social costs of addictions continue into adulthood. For example, according to the Addiction Attention and Prevention Institute (2012), four out of every five crimes in Mexico are committed under the influence of alcohol.
Governments have commonly sought to address this problem through primary prevention programs delivered in the school setting. The evidence of these programs’ effectiveness, mostly from developed countries, is mixed. Faggiano et al (2014) provide a comprehensive overview of various types of programs and a systematic review and meta-analysis of impacts measured through 51 RCTs. They find improvements in knowledge and small effects on behavior, though not always statistically significant. Moreover, these programs are generally quite resource intensive (most involve between 10 to 20 sessions), and for this reason, they cannot be easily scaled up in developing countries.
Many behavior change interventions are based on rational behavior models that greatly assume the provision of information should be sufficient to motivate individuals to adopt healthier behaviors. Rational and individual-centered models are increasingly being enriched with insights from the psychology and sociology literatures (DellaVigna 2007, World Bank World Development report “Mind, Society and Behavior” 2015). Decisions are often driven by emotions, systematic biases, and by an individual’s perception of social norms, or what others do or approve of. Mass media programs can support behavior change campaigns by targeting these non-rational components. Entertainment education (or edutainment) is a type of media that incorporates educational messages into an entertaining format with the end goal of improving knowledge, shifting attitudes and social norms and changing behavior (Singhal and Rogers, 2004). Entertainment education traces its theoretical foundations to Albert Bandura's (1976) social learning theory, which posits that individuals learn by observing others, especially if these are role models that observers can relate to. Narratives are inherently easier to observe, understand and remember than abstract concepts that lack a storyline to connect them (Fisher 1987). Pioneered by Mexican TV producer Miguel Sabido in the 70s, entertainment education has been used to address public policy issues primarily related to health, and increasingly in other areas as well.
The entertainment education production that is the focus of this evaluation, "Addicted to Life" (A2L), has been designed to prevent the consumption of psychoactive substances by young people. A2L was created by Life Changing Experiences, an Israeli company, with the help of national and international behavior specialists, and incorporates local content to make it more relevant and appealing to Mexican audiences. Screenings of A2L will be complemented by interventions organized in schools the following week: a workshop with students and a session with parents.
CinemaPark, the study’s implementation partner, distributes it in Mexico in Cinépolis movie theaters and in schools. Proof of concept evaluations have been conducted for A2L in the US, Israel and Mexico. However, these studies were based on exit surveys collected immediately after program exposure, so they only measured intended, and not realized, behavior change. The objective of this evaluation is to measure impacts on knowledge, attitudes and behaviors related to drug and alcohol use after six months of program exposure.
This study will investigate the potential role of an entertainment education production as a tool to prevent and reduce the use of alcohol and drugs among middle school students in Mexico City. The participants will be students in the second year (12-14 year-olds) of middle school. The selection of the target group is motivated by consumption patterns as measured by the 2012 Survey of Drug Use among Students in Mexico City. This survey revealed that students start experimenting with alcohol and drugs in the second and third year of middle school, hence the importance of implementing a prevention program with this group. We chose students in the second year to be able to measure effects in the following academic year (those in the third year would have already moved on to high school).
The study will employ a cluster randomized controlled trial design where schools will be randomly assigned to one of three groups: screenings of “Addicted to Life” in the movie theater plus complementary school based interventions (Treatment 1), screenings of “Addicted to Life” in schools plus complementary school based interventions (Treatment 2), and a comparison group that will participate in screenings of “Addicted to Life” only after the follow-up survey. This evaluation is part of the Narrating Behavior Change research program of the World Bank’s Development Impact Evaluation (DIME) Unit. In its first phase, this research program is launching randomized controlled trials in the entertainment hubs of Brazil, India, Mexico, and Nigeria.
The study aims to answer the following research questions:
1. Does the intervention lead to changes in knowledge about the effects of drugs, perceptions of risk, self-efficacy to resist peer pressure, beliefs about the consumption of peers?
2. Does the intervention lead to a reduction in or a delayed onset of alcohol and drug use?
3. Does the mode of delivery for the edutainment production matter: do screenings in movie theaters and in schools have similar effects?
4. Do the effects differ by gender?
5. Are there heterogeneous effects on youth at higher risk of consumption ex ante?