Symptomatic benefit and metabolic cost of adding olanzapine to risperidone in schizophrenia: an 8-week retrospective cohort study

A second antipsychotic is commonly added when monotherapy response is judged insufficient, but the symptomatic value of such combinations has rarely been weighed against their metabolic cost in the same patients. In this single-centre retrospective cohort study, 131 hospitalised adults with DSM-5 schizophrenia, confirmed by independent re-review of each record against the DSM-5 criteria, were treated for 8 weeks with risperidone plus olanzapine ( n = 67) or risperidone alone ( n = 64); 124 had complete week-8 data. Co-primary outcomes were change in SAPS and SANS total scores. Secondary outcomes were response rates and change in anthropometric, glycaemic and lipid measures; MoCA was exploratory, being a brief screen readministered in the same form at each visit. Total antipsychotic exposure was expressed as olanzapine-equivalent daily dose and used to estimate striatal dopamine D2 receptor occupancy. Analyses used t-tests, chi-square tests and linear mixed-effects models with Hochberg adjustment. Baseline characteristics, including treatment history, were balanced, but total antipsychotic exposure was not: the combination group received 77% more in olanzapine-equivalent terms (30.71 vs. 17.35 mg/day), with hypothetical D2 receptor occupancies of 87.3% versus 83.0% under the common Emax = 100% assumption. Symptom reduction was greater with combination for SAPS (− 33.35 vs. − 24.16 points; difference − 9.19, 95% CI − 13.18 to − 5.20) and SANS (− 27.11 vs. − 18.33; difference − 8.78, 95% CI − 12.48 to − 5.08; both adjusted P < 0.001). Response rates were 57.1% versus 32.8% for SAPS and 46.0% versus 13.1% for SANS. All eight metabolic parameters deteriorated more with combination (weight + 3.58 vs. + 1.42 kg; difference + 2.16 kg; all adjusted P ≤ 0.004), and weight gain ≥ 7% occurred in 39.7% versus 9.8% (number needed to harm 3). MoCA change did not differ, and weight gain was unrelated to symptom improvement within either group. The combination was associated with greater symptomatic improvement and with a consistent metabolic penalty. Because it was compared with a one-drug regimen without randomisation, and delivered 77% more total antipsychotic, the efficacy signal cannot be separated from total antipsychotic exposure. These data do not establish an advantage of combining two agents over adequate dosing of one. Not applicable.