Limitations of our analyses include sample size limitations on statistical power, RCT participant heterogeneity, and assumptions about variable effects required by our pooled regression analyses. Sample size limitations on statistical power (See Table, Supplemental Digital Content 11: Odds Ratio (OR) detectable with 80% power) may underlie our inability to make statements about chr15q25.1 nAChR SNP pharmacogenetic effects in individuals randomized to BUP, VAR or combined therapies. Increasing PG sample sizes in future analyses will increase power to detect pharmacogenetic effects, but will still require integrated data analysis choices to be made. While there are differences in baseline, treatment and outcome variables among the RCTs, ascertainment characteristics of the RCTs are similar and there are no significant differences in nAChR SNP allele or genotype frequencies among the RCTs. In the analyses reported here, we utilize one approach to performing integrated data analysis, namely, pooled regression analysis. Heuristically, all of the studies contributed to the estimation of the regression coefficients for each demographic and dependence variable, which were assumed to have the same value across arms; each individual arm contributed to estimation
