What to Make of CATIE: Are We Better Off With Atypical Antipsychotics?: Page 5 of 6

What to Make of CATIE: Are We Better Off With Atypical Antipsychotics?: Page 5 of 6

For every pair-wise comparison, with the exception of olanzapine versus risperidone, avoidance of a hospitalization event would be encountered more frequently than weight gain in excess of 7% of baseline (number needed to treat is lower, ie, stronger, than number needed to harm). Ratios of number needed to harm to number needed to treat have been used to describe the likelihood of being helped or harmed.14 However, some patients may gain substantially more than 7% of their baseline weight. Weight gain and robust efficacy in the same patient poses a clinical dilemma.

Moreover, even though number needed to treat and number needed to harm can help predict how often events can occur, there are no guarantees of weight gain or loss, or of a drug’s effectiveness. The patient’s history remains the best predictor of treatment success or failure.


After unsuccessful trials of several antipsychotics, both Simone and Sabella are fed up. They want to try clozapine. What does CATIE tell us about switching from a first-line atypical antipsychotic to clozapine?

CATIE phase 2E: efficacy

The “clozapine pathway” of phase 2 of CATIE, known as phase 2E, was designed to test what advantages there might be for switching to clozapine for patients who discontinued their phase 1 or phase 1B atypical antipsychotic because of poor efficacy. Fewer patients randomized to olanzapine in phase 1 entered phase 2E than 2T (19% vs 26%), but more patients who were randomized to quetiapine in phase 1 entered phase 2E than 2T (37% vs 31%). A possible confounder is that patients who entered phase 2E knew they could be randomized to clozapine, and once randomized knew if they were actually getting open-label clozapine versus double-blinded olanzapine, quetiapine, or risperidone. Hence, if they really wanted clozapine and did not get randomized to it, they may have been inclined to discontinue their anti-psychotic. The number needed to treat and the 95% CI for all-cause discontinuation are summarized in Table 9.

Statistically significant differences are evident for the comparisons of clozapine with risperidone or quetiapine; the number needed to treat showed an advantage for clozapine every 4 or 3 patients, respectively, in terms of all-cause discontinuation. The number needed to treat for clozapine versus olanzapine was 7, which indicates a respectable effect size difference that could have been statistically significant if only the sample size had been larger.


Simone and Sabella have heard that clozapine may be associated with hypersalivation. They ask how frequent a problem this is.

The importance of context when looking at treatment advantage

Sialorrhea is a common side effect of clozapine. Inspection of the clozapine product label reveals a reported frequency of 31% for hypersalivation.15 CATIE provides enough information to calculate the number needed to harm for sialorrhea, comparing clozapine with quetiapine, olanzapine, and risperidone, as outlined in Table 10.

For every 3 patients randomized to clozapine versus quetiapine, one additional case of sialorrhea was encountered. This is a strong effect size and emphasizes how common this adverse effect is with clozapine. However, if good efficacy is achieved, patients who take clozapine will tol-erate hypersalivation and will not stop their medication.

Loading comments...

By clicking Accept, you agree to become a member of the UBM Medica Community.