Levoprotiline

To the Editor, in the abstract to their paper, Katz et al. (1993) comment that oxaprotiline significantly resembled amitriptyline and that levoprotiline significantly resembled placebo in antidepressant potential, ! take issue with the authors on their phraseology. Statistical tests are constructed to test significance of differences (among other things) and not of resemblances. What Katz et al. actually found was that oxaprotiline did not differ significantly from amitriptyline and that levoprotiline did not differ significantly from placebo. There's no splitting of hairs hero--the issue is an important one, with bearing on the power of the applied test, which is in turn dependant on (for example) sample size. Thus, if 4 of 8 depressed patients respond to placebo while 7 of 8 respond to fiuoxetine, statistically the drug groups do not differ. This is not to say that fiuoxetine significantly resembles placebo in antidepressant potential for, if the sa:,~o proportions are maintained and if 40 of 80 depressed patients respond to placebo while 70 of 80 respond to fluoxetine, the 2 drugs will indeed differ stat.:stically. It is convenient to use phrases like "significantly resembles"

We describe a double-blind study involving 58 in-patients with major depression (DSM-III). After one week on placebo, the patients were randomly assigned to either levoprotiline or maprotiline treatment for three weeks. In the next three weeks, responders were maintained on the same medication and non-responders were shifted to treatment with the complementary drug. After the initial three weeks' treatment, 31% of levoprotiline patients and 58% of maprotiline patients had responded. Both in the initial three-week treatment period and after shifting non-responders to the complementary drug, there were significant differences in favour of maprotiline. The comparison of properties of different outcome criteria in prediction analyses shows that the final score gives the best agreement with global evaluation; using the delta score (final minus baseline) or ratio score (final/baseline) as the outcome criterion may yield paradoxical results.