Purpose <p>The relevance of CYP2D6 activity on serum levels and clinical response of fluoxetine remains unclear. The present study aim was to evaluate the impact of <i>CYP2D6</i> genotype on i) fluoxetine and norfluoxetine exposure, and ii) treatment switch from fluoxetine to alternative antidepressants in adults and children/adolescents.</p> Methods <p>Patients were included retrospectively from a therapeutic drug monitoring service. Patients were subgrouped by age, i.e. &lt; 18 yrs (young) and ≥ 18 yrs (adult), and divided into CYP2D6 genotype-predicted phenotype subgroups of poor metabolizers (PMs), intermediate metabolizers (IMs), normal metabolizers (NMs) and ultrarapid metabolizers (UMs).</p> Results <p>Among 1027 adult patients, the metabolic ratio was lowest in PMs and highest in UMs (<i>p</i> ≤ 0.04), but there was no difference in active moiety (fluoxetine + norfluoxetine) across CYP2D6 phenotype groups (<i>p</i> ≥ 0.1). Similar trends were observed in 196 young patients, both for metabolic ratio and active moiety. In adult patients, switching from fluoxetine to an alternative antidepressant had odds ratio of 2.9 in UMs (<i>p</i> = 0.004) and 2.3 in PMs (<i>p</i> = 0.007) compared with NMs. The number of switches per genotype group was too low for meaningful comparisons in young patients.</p> Conclusion <p>The fluoxetine and norfluoxetine active moiety was unaffected by <i>CYP2D6</i> genotype in adults and children/adolescents. Still, adult CYP2D6 PMs and UMs switched antidepressant treatment two to three times more often than NMs, indicating that relative levels of fluoxetine and norfluoxetine may affect treatment outcome rather than active moiety in relation to CYP2D6 phenotype.</p>