Effect of Clinical Doses of Fluoxetine on Psychological Variables in Healthy Volunteers

To determine whether selective serotonin reuptake inhibitors (SSRIs) act by correcting a target deficit in depression or have a general thymoleptic action, we examined the effect of 6 weeks of fluoxetine administration on the mood, general well-being, quality of life, and personality characteristics of healthy volunteers. Previous reports on the psychological effects of antidepressants in normal subjects have been limited to short-term evaluations (1, 2).

METHOD

Fifteen volunteers (nine men and six women, mean age=38.5 years, SD=12.9, range=22–63) gave written informed consent for this study, which was approved by an internal review board. They received financial compensation for time and travel. They were interviewed with the Schedule for Affective Disorders and Schizophrenia—Lifetime Version (3), did not meet criteria for any DSM-IV axis I or II disorder, had no history of psychiatric illness in first-degree relatives, were in good physical health, were not taking medication, and did not use any substances (besides nicotine in seven cases). Thirteen were employed, one was retired, and one was a student; 12 were living with a partner.

For 10 weeks the subjects were given one capsule per day: a placebo capsule for the first 2 weeks, 10 mg of fluoxetine the next week, 20 mg of fluoxetine for the following 5 weeks, and a placebo capsule for the last 2 weeks. They did not know for how long or when placebo would be administered. They were followed up in person after 2 weeks (end of first placebo phase), 5 weeks (after 3 weeks of receiving fluoxetine), 8 weeks (after 6 weeks of receiving fluoxetine), and after the final 2 weeks of placebo administration. They were assessed with the Hamilton anxiety and depression scales (4) at each visit and filled in a 10-cm visual analogue mood scale (5, 6), the State-Trait Anxiety Inventory (7), the General Well-Being Schedule (8), and the Quality of Life Enjoyment and Satisfaction Questionnaire (9). Visual analogue mood scales have been extensively used for the assessment of subjective phenomena, including drug effects (5), and have been shown to be sensitive to individual differences in nonclinical settings (6). The General Well-Being Schedule (8) generates a score that reflects overall well-being and scores on six subscales. The Quality of Life Enjoyment and Satisfaction Questionnaire (9) provides a sensitive measure of the enjoyment and satisfaction experienced by the subject in various areas of daily functioning. The Comrey Personality Scales (10) provide scores on eight personality dimensions, a validity check for random scoring, and a response bias check. These scales were administered at baseline and after 6 weeks of fluoxetine administration. Information on adverse effects was obtained through a general inquiry and by a checklist at each visit and by telephone at the end of each week in which the subject was not seen in person.

Because the data were not normally distributed, Friedman's nonparametric analysis of variance with repeated measures was employed to evaluate changes in the rating scale scores over the course of the trial. If significant, the Wilcoxon signed ranks test was applied to compare time points. Bonferroni correction (α=0.05) was applied.

RESULTS

There was no significant change in visual analogue mood scale scores over the course of the trial (table 1). Four subjects deviated by more than two standard deviations at specific time points but inconsistently. Hamilton anxiety and depression scale scores were minimal at baseline and exhibited negligible change, as did the State-Trait Anxiety Inventory and the overall and subscale scores of the General Well-Being Schedule.

Overall scores (general and life satisfaction scales) of the Quality of Life Enjoyment and Satisfaction Questionnaire were stable over the entire study. Of the subscales, physical health (χ²=17.35, df=4, p=0.002), work (χ²=10.08, df=4, p=0.004), leisure activities (χ²=10.71, df=4, p=0.03), and social relationships (χ²=12.67, df=4, p=0.01) showed extremely small changes, but only physical health remained significant after Bonferroni correction (for 14 analyses, an alpha level of 0.05 requires a significance level of p<0.004). No comparisons of the five time points of the physical health subscale remained significant after Bonferroni correction (for 10 comparisons, an alpha level of 0.05 requires a significance level of p<0.005). Observed trends suggested change over the entire trial rather than a specific effect of the active drug.

The mean score for the Comrey Personality Scales validity check was well below the ceiling of 32. There was no response bias (mean=17.4, SD=2.7, range=12–23). Scores on the eight personality scales did not deviate from expected norms, were highly correlated at baseline and after 6 weeks of fluoxetine administration (r_s=0.72–0.91), and showed no significant change (p>0.10).

Three subjects reported adverse effects during the study, and two of them reported these effects during the fluoxetine phase (dizziness [first week only] and dyspepsia [throughout] for one subject and nausea [first to third weeks] for the other).

DISCUSSION

The striking finding of this study is that 6 weeks of fluoxetine administration at a “therapeutic” dose had no significant effect on mood, general well-being, or quality of life in normal volunteers. When asked to guess during which weeks active drug and placebo were administered, none of our subjects was able to differentiate the periods precisely. Our findings contrast with the observations of Kramer (11) (whose subjects all had sought treatment) and suggest that mood-enhancing and other psychoactive effects of SSRIs are not a general property of these agents but are manifest in the context of target symptoms.

Short-term treatment trials for patients with major depression have shown 6 weeks of fluoxetine to be effective and have failed to show a significant difference between 10 and 20 mg/day (12). Steady-state plasma levels of fluoxetine and norfluoxetine are achieved within 28 days (13). Nevertheless, effects of a longer period of fluoxetine cannot be definitively excluded. In a larger cohort, effects might have been detected in a subgroup, or the small trends that we noted might have emerged more strongly. Because of the 15-day elimination half-life of the active metabolite norfluoxetine (13), the placebo period at the end of the trial was too short to draw conclusions regarding withdrawal effects.

Fluoxetine induced minimal adverse effects. Since information was elicited weekly by general inquiry and a checklist, underreporting was unlikely. The graduated dose regimen could have reduced gastrointestinal complaints and anxiety, whereas effects on sexual function might have been limited by the length of the trial. Another placebo-controlled study of fluoxetine administration to normal subjects (60 mg/day for 14 days) also found a lower than expected rate of adverse effects (14).

TABLE 1

Enlarge table