The comparative effectiveness of eight phenothiazines

Summary

In a double-blind study of the comparative effectiveness of eight phenothiazines, 322 newly-admitted state psychiatric hospital patients requiring tranquilizing drug therapy were randomly assigned to treatment with one of the following drugs: chlorpromazine, thioridazine, triflupromazine, prochlorperazine, perphenazine, thiopropazate, trifluoperazine, and fluphenazine. The design of the study permitted flexible dosages, and treatment, unless prematurely terminated, lasted 30 days. Psychological and/or psychiatric assessments were made prior to treatment and at the end of 5, 15, and 30 days.

Although results again substantiated the effectiveness of the phenothiazine compounds as a whole, evidence of differential drug effectiveness was equivocal using raw scores derived from standard rating scales and traditional methods of statistical analysis. Such analyses suggested negligible treatment differences in terms of means. Non-constant, non-additive treatment effects were greatly in evidence, however; and the high incidence of heterogeneity of variance and of regression indicated that the drugs were neither equivalent nor interchangeable.

Subsequent analyses of MSRPP Total Morbidity, involving the more severely ill patients and using a scale transformation to achieve homogeneity of variance, yielded significant treatment differences at every evaluation period. In these analyses, the drugs found to be consistently effective over the 30-day treatment period were prochlorperazine, perphenazine, and fluphenazine. Those found to be less effective were thioridazine, triflupromazine, and thiopropazate. For the less severely ill patients, significant mean differences among the eight drugs were not obtained; however, there was a tendency for the rank order of drug efficacy in this group to be the reverse of that obtained with the severely ill patients.