Dissertation
EVALUATING A COMMUNITY COALITION MODEL FOR ADOLESCENT SUBSTANCE ABUSE PREVENTION USING PROPENSITY SCORE ANALYSIS
Doctor of Philosophy (PhD), Washington State University
01/2019
Handle:
https://hdl.handle.net/2376/17890
Abstract
Introduction: Community Prevention and Wellness Initiative (CPWI) is a strategic, data-informed, community coalition model aimed at bringing together key stakeholders to reduce underage alcohol use, other substance use, and related risk factors among adolescents through evidence-based prevention programming. In this cross-sectional study, I evaluated the effectiveness of CPWI in reducing 8th and10th grade substance use and related risk factors.
Method: The sample consisted of students who participated in the statewide Healthy Youth Survey from 18 CPWI communities and approximately 160 control communities. I used multilevel level modeling on propensity-score adjusted data to examine whether CPWI communities differed significantly from control communities in outcomes of interest in 2016 compared to baseline. I conducted sensitivity analysis for unmeasured confounding by adding prevention readiness variables in the propensity score model. I also conducted sensitivity analysis for variable selection by adding a wider range of baseline covariates to the propensity score model.
Results: At baseline, CPWI communities were at significantly higher risks compared to control communities in 15 outcomes. In 2016, CPWI communities closed all baseline gaps in alcohol use, family risk factors, and community risk factors. However, new gaps in levels of risks were created in 2016 with CPWI communities reporting significantly higher rates of cigarette use. Similarly, marijuana use as well as alcohol and marijuana co-use showed mixed results; there were some improvements but CPWI communities continued to remain at increased levels of risk in 2016 for some outcomes. Results from the unmeasured confounding sensitivity analysis indicate that the addition of readiness variables in the propensity score model led to improvements in results for some marijuana use outcomes and peer-individual risk factors. Results of the variable selection sensitivity analysis did not differ from the results of the primary analysis.
Conclusion: CPWI was effective in reducing adolescent alcohol use and some related risk factors. Inclusion of readiness data in statistical analysis showed that CPWI is also effective in reducing youth marijuana use. Results of the current study indicate that CPWI is a promising model for disseminating evidence-based substance use prevention programs for adolescents.
Metrics
26 File views/ downloads
23 Record Views
Details
- Title
- EVALUATING A COMMUNITY COALITION MODEL FOR ADOLESCENT SUBSTANCE ABUSE PREVENTION USING PROPENSITY SCORE ANALYSIS
- Creators
- Gitanjali Shrestha
- Contributors
- Laura G Hill (Advisor)Brittany R Cooper (Committee Member)Robert Rosenman (Committee Member)Sterling McPherson (Committee Member)
- Awarding Institution
- Washington State University
- Academic Unit
- Department of Human Development
- Theses and Dissertations
- Doctor of Philosophy (PhD), Washington State University
- Number of pages
- 223
- Identifiers
- 99900581813501842
- Language
- English
- Resource Type
- Dissertation