Background: Pharmacogenetics is considered a promising avenue for improving treatment outcomes, yet evidence arguing for the use of pharmacogenetics in the treatment of psychotic disorders is mixed and clinical usefulness is under debate. Many patients with psychosis use multiple medications, which can alter the metabolic capacity of CYP enzymes, a process called phenoconversion. In clinical studies, treatment outcomes of drugs for psychosis management may have been influenced by phenoconversion. Aim: Here we evaluate the impact and predictive value of CYP2D6 phenoconversion in patients with psychotic disorders under pharmacological treatment. Method: Phenoconversion-corrected phenotype was determined by accounting for inhibitor strength. Phenoconversion-corrected and genotype-predicted phenotypes were compared in association with side effects, subjective well-being and symptom severity. Results: Phenoconversion led to a large increase in poor metabolizers (PMs; 17–82, 16% of sample), due to concomitant use of the serotonin reuptake inhibitors fluoxetine and paroxetine. Neither CYP2D6-predicted nor phenoconversion-corrected phenotype was robustly associated with outcome measures. Risperidone, however, was most affected by the CYP2D6 genotype. Conclusion: Polypharmacy and phenoconversion were prevalent and accounted for a significant increase in PMs. CYP2D6 may play a limited role in side effects, symptoms and well-being measures. However, due to the high frequency of occurrence, phenoconversion should be considered in future clinical trials.