IMPORTANCE A substantial proportion of women with schizophrenia experience debilitating treatment-refractory symptoms. The efficacy of estrogen in modulating brain function in schizophrenia has to be balanced against excess exposure of peripheral tissue. Raloxifene hydrochloride is a selective estrogen receptor modulator (mixed estrogen agonist/ antagonist) with potential psychoprotective effects and fewer estrogenic adverse effects. OBJECTIVE To determine whether adjunctive raloxifene therapy reduces illness severity in women with refractory schizophrenia. DESIGN, SETTING, AND PARTICIPANTS This 12-week, double-blind, placebo-controlled, randomized clinical trial with fortnightly assessments was performed at an urban tertiary referral center and a regional center from January 1, 2006, to December 31, 2014. Participants included 56 women with schizophrenia or schizoaffective disorder and marked symptom severity despite substantial and stable antipsychotic doses. Data were analyzed using intention to treat as the basis. INTERVENTIONS Adjunctive raloxifene hydrochloride, 120mg/d, or placebo for 12 weeks. MAIN OUTCOMES AND MEASURES The primary outcomewas the change in the Positive and Negative Syndrome Scale (PANSS) total score. Clinical response (defined as a-20% decrease in PANSS total score from baseline) and change in PANSS subscale scores, mood, cognition, reproductive hormone levels, and adverse events were also assessed. RESULTS Of the 56 participants (mean [SD] age, 53 [7.7] years; age range, 40-70 years; mean [SD] duration of psychotic illness, 24  years), 26 were randomized to raloxifene and 30 were randomized to placebo. Raloxifene produced a greater reduction in the PANSS total score relative to placebo (β = -6.37; 95%CI, -11.64 to -1.10; P = .02) and resulted in an increased probability of a clinical response (hazard ratio, 5.79; 95%CI, 1.46 to 22.97; P = .01). A significant reduction was found in the PANSS general symptom scores for the raloxifene compared with the placebo (β = -3.72; 95%CI, -6.83 to -0.61; P = .02) groups. For patients who completed the full 12-week trial, there was not a statistically significant treatment effect on PANSS positive symptom scores (β for change in raloxifene vs placebo, -1.92; 95%CI, -3.83 to 0.00; P = .05). Change in mood, cognition, and reproductive hormone levels and the rate of adverse events did not differ between groups. CONCLUSIONS AND RELEVANCE Raloxifene hydrochloride, 120mg/d, reduces illness severity and increases the probability of a clinical response in women with refractory schizophrenia. This large trial of raloxifene in this patient population offers a promising, well-tolerated agent that has potential application in clinical practice.