Comparing antipsychotic treatments for schizophrenia: a health state approach

The overall impact of first and second generation antipsychotics on quality of life and symptoms of people with schizophrenia remains controversial. We applied health state modeling to data from the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) Schizophrenia study, a randomized trial of antipsychotic medications, and evaluated the likelihood of patients moving to more favorable health states over time. We applied K-means clustering to the data to create discrete groupings of patients with symptom and side effect characteristics that were then validated using quality of life measures. We compared cluster distributions across medications at baseline and 6 months after randomization. 1,049 patients were included in the initial cluster analysis. Five health states were identified: (1) low symptoms and low side effects (LS + LSE) (2) low symptoms and obesity (LS + Ob) (3) high symptoms and low side effects (HS + LSE) (4) high symptoms with depression and akathisia (HS + Dp + Ak) and (5) moderate symptoms and high side effects (MS + HSE). Six-month outcomes among patients randomly assigned to perphenazine, olanzapine, risperidone and quetiapine were compared. At baseline, almost 20% of patients were in the worst health state (HS + Dp + Ak), with greater decreases at 6 months in this health state for perphenazine (9.2% decrease) and olanzapine (11.1%) groups compared to risperidone (4.7%) and quetiapine (6.7%). This study demonstrated that health state analysis can provide insight into the overall clinical state of patients beyond the mere comparison of average scores and largely confirmed original CATIE findings.