Webs of Persuasion: The Connections of Social Influences and Educational Prevention on Youth Illicit Drug Use

Background: Prior research highlights the importance of parental support and monitoring of activities in addition to preventive drug education as a way to reduce adolescent substance abuse.  Specifically, research on adolescent substance abuse has demonstrated the preventative efficacy of positive peer support, positive parental involvement, and certain drug education programs.  Such research is embedded in a sociocultural theoretical framework, suggesting a variety of factors may determine why individuals use drugs and which kind, and which factors may act as preventive deterrents.  While research examining the direct effects of preventive drug education on adolescent drug use has produced inconsistent results, little or no research has examined possible mediated effects of preventive drug education by the social influences of parents and peers onto illicit drug use of adolescents.  Because prevention efforts of drug education can operate across different levels of environmental systems, the potential for interaction among various factors has important implications on services and methods of delivery when designing and implementing adolescent substance prevention programs. The study has been approved by authors’ university IRB.

Methods: This study conducted secondary data analysis with adolescents, aged 12 to 17 years, using a public dataset (http://www.icpsr.umich.edu/icpsrweb/NAHDAP/studies/35509)  (N=17736) of the 2013 National Study on Drug Use and Health. The hypothesis guiding the study posits that social influences of parental support and monitoring and peer attitudes may mediate drug education when predicting adolescent illicit drug use while controlling for youth age and gender.  The use of illegal drugs in the current study is defined as illicit use of analgesics, cocaine, heroin, marijuana, hallucinogens, inhalants, sedatives, stimulants, and tranquilizers.  The hypothesized path model is tested with structural equation modeling in SPSS with Amos software.  The SEM analysis uses Full Maximum Likelihood to address missing values.  Model worthiness is evaluated by Normative Fit Index (NFI), Tucker-Lewis Index (TLI), and Benter’s Comparative Fit Index (CFI) with expected values > 0.95, and Root Mean Square Error of Approximation (RMSEA) with expected values < 0.06 to indicate a good model fit, and comparing models with and without missing values at df and p < 0.05 for model differences.

Results: The final SEM path analysis with FML produced a model with good results: Chi-Square (df 3) = 16.720, p < 0.001; NFI = 0.997, TLI = 0.979, CFI = 0.998, RMSEA = 0.016 with PCLOSE = 1.000.  The path model explained 16 percent of variance in adolescent illicit drug use.  Comparison of models with and without missing values produced equally strong results. Surprisingly, drug prevention education did not directly influence youth illicit drug use. However, parental monitoring, parental support, and peer attitude fully mediated drug prevention influence on youth illicit drug use.  Authors consider strengths and limitations of the correlational design and directions for future research.

Implications:  Findings suggest that exposure to drug prevention messages alone may not directly influence youth behavior.  When considering design and delivery of drug prevention education, educators and policy-makers may need to consider the importance of parental and peer influences on youth behavior in illicit drug use.