Predictors of Enacted Stigma Following Disclosure Among People in Recovery From Opioid Use Disorder: A Machine...

Area of Science:

• Addiction research
• Psychology
• Machine learning applications

Background:

• Enacted stigma negatively impacts treatment retention and recovery for individuals with opioid use disorder (OUD).
• Identifying individuals at risk of enacted stigma is crucial for developing targeted interventions.
• Machine learning can uncover complex patterns in predicting stigma experiences.

Purpose of the Study:

• To apply a machine learning framework to identify predictors of enacted stigma in individuals recovering from OUD.
• To inform interventions aimed at reducing stigma and improving well-being and treatment outcomes.

Main Methods:

• Longitudinal study with 112 participants completing surveys at baseline and 3-month follow-up.
• Tested three machine learning models to predict enacted stigma.
• Evaluated model performance using various metrics, including R-squared.

Main Results:

• A random forest model achieved an R-squared of 0.85, explaining 85% of the variance in enacted stigma.
• Key predictors identified: shorter recovery duration, younger age, non-disclosure, current substance use issues, and commitment to sobriety.
• Younger individuals, those with shorter recovery times, non-disclosure, and ongoing substance use issues were at higher risk.

Conclusions:

• Shorter recovery duration, younger age, non-disclosure, and current substance use issues are significant risk factors for enacted stigma in OUD recovery.
• Interventions should target individuals with these characteristics to mitigate stigma.
• Reducing enacted stigma can enhance well-being and treatment outcomes for individuals in OUD recovery.