Description
- Abstract:
- Background & Aims. Alcohol and other drug (AOD) use disorders are a significant public health concern. Cognitive-Behaviorally Based Interventions (CBIs) are frontline interventions for AOD treatment. This thesis study pursued three aims related to the efficacy of CBIs in terms of key population and study context variables. Firstly, we assessed the reporting of demographic information in these studies, to determine how demographic characteristics such as age, race, ethnicity, sex, gender, sexual orientation, and socioeconomic status are being reported and whether there is need for increased consideration of diversity in this research. Secondly, we examined CBI efficacy for different client sub-populations as defined by demographic characteristics. Finally, we examined CBI efficacy in different service contexts as defined by recruitment source (i.e., community samples, medical setting, specialty substance use or mental health setting, college setting, or criminal justice setting), treatment type (i.e., inpatient vs. outpatient), and treatment funding source (i.e., public vs. private). We also examined whether CBI efficacy differed by other study characteristic variables including drug type (i.e., alcohol, poly drug, cannabis, and other drugs [e.g. opiates and stimulants]), study inclusion of one or more co-occurring mental health conditions (i.e., yes or no), use of biological assay outcome (e.g., urine-screening) measures (i.e., yes or no), treatment length (i.e., scheduled number of sessions), CBT format (i.e. individual therapy, group, or mixed), and risk of bias (i.e. low risk vs. some concerns and high risk). Methods. Study characteristic data were collected and summarized using mean and standard deviation estimates as well as percentile values. Inverse-variance weighted standardized mean differences (effect sizes) were calculated for within- and between-group comparisons (i.e., CBI pre-test vs CBI post-test; CBI post-test vs control post-test, respectively) in each study and synthesized with random effects models. The demographic and study context variables used in the moderator analysis were determined during the data acquisition phase based on what was reported in the primary studies. Sensitivity analyses included assessing for study risk of bias using Cochrane guidelines, heterogeneity, influential studies, and publication bias. Results. The sample included K= 29 AOD clinical trials conducted in the United States (US). Of the racial/ethnic categories, the percentage of White participants was reported in most (26 studies [68% White on average]). Other racial/ethnic categories were less frequently reported and Native American, Asian, and Hawaiian Pacific Islander participants were underrepresented. The percentage of transgender participants was reported in three studies and among those studies, 20% of participants were transgender. Four studies reported sexual orientation and among those studies, 59% of participants were heterosexual. The mean between-group effect size was small and moderately heterogenous (d = 0.158, 95% CI = 0.079, 0.238, p = 0.000, I2 = 46.296) and the mean within-group effect large and very heterogenous (dz = 1.147, 95% CI = 0.811, 1.482, p = 0.000, I2 = 96.203). Drug type and biological assay outcome measures moderated between-group CBI efficacy and co-occurring disorders and study date moderated within-group CBI efficacy. Conclusions. Key demographic characteristics such as gender identity, sexual orientation, race, and ethnicity were underreported and/or were reported in non-inclusive ways. Non-alcohol studies had larger effects which may have been driven by large effects in two opioid studies. Studies with biological measures had larger effect sizes which may be explained by the superior precision of measurements or the overall robustness of these studies. Newer studies having smaller effects could potentially be a result of increased efforts to reduce publication bias. Co-occurring disorder studies having smaller effects may be due to the lower treatment adherence and higher drop-out often seen in these studies. Future research in the field of AOD treatment should seek to include more gender, sexual, racial, and ethnic minority participants, and report demographic information as inclusively and comprehensively as possible.
- Notes:
- Thesis (M. P. H.)--Brown University, 2023
Citation
Burnett, Don-Pierre,
"A Descriptive Review and Meta-Regression of Demographic and Study Context Factors in US Clinical Trials of CBT Efficacy for Substance Use"
(2023).
Public Health Theses and Dissertations.
Brown Digital Repository. Brown University Library.
https://repository.library.brown.edu/studio/item/bdr:duvf5a85/
Relations
Collection:
-
Public Health Theses and Dissertations
Theses and Dissertations for the Public Health department....