A methodological note on the CYP-guides randomized controlled trial for the treatment of major depression

Abstract

Genetically guided therapy for major depression has been recently recommended but has resulted in null effects. We hypothesized that a potential confounding variable for not finding differences in Length of Stay and Re-admission rate in a major depression clinical trial is the equivalence between treatment protocols in the standard treatment (S) and CYP2D6 genotype (G) treatment groups. The two treatment protocols (i.e., type and degree of substrate drugs) were contrasted using a latent class analysis (LCA) model. Specifically, an LCA model specifying the presence of two classes, namely, the G and S groups was estimated with the intercepts of the 30 prescribed drugs freely estimated. This model was compared to a constrained latent class model in which the two treatment protocols (intercept terms) were contrasted to be equivalent between the two treatment groups. Results indicated that there were no significant differences between G and S treatment groups in the types and number of drugs administered. Consequently, the lack of finding significant differences in length of stay and readmission rate may likely be attributed to the equivalence of the treatment protocols.