Sedation is an often unanticipated but clinically significant side effect in some patients treated with fluoxetine. To date, no clinical or pharmacokinetic factors have been identified in association with sedation in fluoxetine-treated patients. We hypothesized that patients experiencing sedation might have elevated plasma concentrations of fluoxetine and/or norfluoxetine compared with patients without sedation.Plasma samples from eight patients reporting new-onset sedation following initiation of treatment with fluoxetine and from 14 consecutive patients also receiving fluoxetine but who experienced no sedation were analyzed for fluoxetine and norfluoxetine concentrations.Patients in the two groups did not differ significantly in age, sex distribution, diagnoses, mean dose or duration of fluoxetine treatment, or mean combined plasma concentrations of fluoxetine and norfluoxetine. However, the plasma ratio of fluoxetine to norfluoxetine was less than 1.0 in all patients reporting sedation and greater than 1.0 in all patients without sedation. Also, there was a significant difference between the mean +/- SD ratio of fluoxetine-to-norfluoxetine plasma concentrations in the group reporting sedation (0.6 +/- 0.1) and the group without sedation (1.7 +/- 0.8).The results of this study suggest an association between sedation and increased plasma concentrations of norfluoxetine compared with fluoxetine. Individuals reporting sedation may metabolize the parent drug differently than patients who do not experience sedation, leading to a relative increase in the proportion of the active metabolite, norfluoxetine.