Atomoxetine is mainly metabolized by CYP2D6 while CYP2C19 plays a secondary role. It is known that patients carrying genotypes encoding decreased/absent CYP2D6 metabolism obtain higher atomoxetine concentrations and are at increased risk of adverse effects. Here, we aimed to investigate the added effects of reduced-function CYP2C19 genotype on atomoxetine concentrations in real-world settings. Serum atomoxetine concentrations and CYP2D6/2C19 genotypes were included from a therapeutic drug monitoring service. Patients were first subgrouped according to CYP2D6 encoding normal, reduced or absent CYP2D6 metabolism, referred to as normal (NM), intermediate (IM) or poor metabolizers (PM). Then, the effect of reduced-function CYP2C19 genotypes was investigated. Genotyping of the CYP2D6 nonfuctional or reduced variant alleles comprised CYP2D6*3-*5, *9-*10 and *41. For CYP2C19, the CYP2C19*2 was analysed to define metabolizer phenotype. Dose-adjusted serum atomoxetine concentration was the exposure measure. Using a patient cohort (n=315), it was found that CYP2D6 IM and PM patients had 1.9-fold (95%CI: 1.4-2.7) and 9.6-fold (5.9-16) higher exposure of atomoxetine compared with CYP2D6 NMs. CYP2C19*2 carriers had 1.5-fold (1.1-2.2) higher atomoxetine exposure than non-carriers regardless of CYP2D6 genotype. CYP2D6 genotype has a great impact on atomoxetine exposure, where our real-world data suggest atomoxetine dose requirements to be around half and one-tenth in CYP2D6 IM and PM vs. NM patients, respectively. When adding CYP2C19 genotype as a factor of relevance for personalized atomoxetine dosing, CYP2C19*2 carriers should further reduce the dose by a third. These findings suggest that pre-emptive CYP2D6/CYP2C19 genotyping should be performed to individualize atomoxetine dosing and prevent adverse effects.