To the Editor: We share Dr. Horrobin’s disappointment that 3 g/day of ethyl EPA added to a current antipsychotic medication regimen was no more effective than placebo in reducing residual symptoms and cognitive deficits in patients with schizophrenia. In agreement with Dr. Horrobin, we indicate that potential explanations for our failure to find a therapeutic effect might include 1) the relative illness duration and severity of our study group, 2) inadequate or excessive doses of ethyl EPA, 3) an inadequate duration of treatment, or 4) ineffectiveness of the putative therapeutic agent. On the other hand, we do not concur that "it is unlikely that any standard drug would show benefit in a trial with this add-on design." Add-on designs such as this have demonstrated the benefit of many augmentation strategies, including lithium augmentation of antidepressants (1), d-cycloserine augmentation of conventional antipsychotic agents for negative symptoms in schizophrenia (2), and pindolol augmentation for patients with treatment-resistant panic disorder (3). Contrary to Dr. Horrobin’s contention, if the hypothesis under consideration is that supplemental therapy is beneficial to patients with residual symptoms and deficits despite adequate treatment with standard approaches, a double-blind placebo-controlled add-on study would appear to be the only appropriate experimental design to rigorously assess the hypothesis. Finally, in our view, our failure to find any correlation between changes in the arachidonic acid/EPA ratio and improvement in clinical dependent variables renders unlikely the hypothesis that a diet change in the placebo-treated group explains improvement in both groups. Given that the bulk of the placebo effect for both groups was seen in the first 2 weeks of treatment, however, we recommend that future studies consider using a single-blind placebo lead-in period. Further research will be needed to clarify the potential use of ethyl EPA in schizophrenia and other neuropsychiatric conditions.