Examining Treatment Effects in Substance Use Disorder Treatments
Author Information
Author(s): Ryoko Susukida, Masoumeh Amin‐Esmaeili, Elena Badillo‐Goicoechea, Trang Q. Nguyen, Elizabeth A. Stuart, Michael Rosenblum, Kelly E. Dunn, Ramin Mojtabai
Primary Institution: Johns Hopkins Bloomberg School of Public Health
Hypothesis
The causal forest approach may reveal varying patterns in the conditional average treatment effect (CATE) linked to specific covariates in substance use disorder treatments.
Conclusion
The study found limited evidence of treatment effect heterogeneity across most psychosocial interventions, but identified potential subgroup-specific benefits in motivational interviewing.
Supporting Evidence
- The motivational incentive intervention displayed a statistically significant positive average treatment effect.
- The CATE variance was consistently lower than the total outcome variability across nine interventions.
- The RATE analysis revealed a statistically significant estimate only for the motivational interviewing intervention.
Takeaway
This study looked at how different treatments for drug problems work for different people, finding that some treatments might help certain groups more than others.
Methodology
The study used data from 12 randomized controlled trials and applied a causal forest approach to estimate the conditional average treatment effect on drug abstinence.
Potential Biases
The analysis was conducted within a single clinical network, which may limit the generalizability of the findings.
Limitations
The study's sample size for each intervention was limited, which may restrict the ability to detect subtle treatment effect heterogeneities.
Participant Demographics
Demographic variables such as age, sex/gender, race, and socioeconomic status were included in the analysis.
Statistical Information
Confidence Interval
95% CI: 0.02, 0.25 for motivational interviewing
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website