The Impact of Study Design Comparative Effectiveness Research in Schizophrenia
Noam Y. Kirson, PhD; Bruce J. Wong, MD; Yermakov Sander, MS; Wayne Huang, MPP; Thomas Samuelson; Steve Offord, PhD; and Paul E. Greenberg, MS
From the Analysis Group, Inc, Boston, Massachusetts (Dr Kirson, Mr Greenberg, Mr Huang, Mr Samuelson, and Mr Yermakov); the University of Pennsylvania, Philadelphia (Dr Wong); and Otsuka America Pharmaceutical, Inc., Princeton, New Jersey (Dr Offord).
This poster presentation was supported by Otsuka America Pharmaceutical, Inc.
Background: Randomized controlled trials (RCTs) and observational studies (prospective and retrospective) have yielded inconsistent results when comparing oral and depot formulations of antipsychotics, highlighting the question of how research design affects estimates of comparative effectiveness (CE). We build on a cross-design synthesis methodology to quantify the effect of study design on the CE of antipsychotic formulations.
Methods: A search of English literature since 2000 for studies with depot and oral antipsychotic treatment arms for schizophrenia with relapse, hospitalization or all-cause discontinuation as key endpoints. Average baseline characteristics were used to adjust endpoints for age and gender. Adjusted endpoints were converted to risk ratios (RR) [depot/oral; RR<1 favors depot] and pooled by study design (RCT, prospective, and retrospective). Meta-analysis with random effects was used to estimate the pooled RR and 95% confidence interval (CI) of all endpoints combined within each study design. The ratios of resulting point estimates were used to calculate average conversion factors between study designs.
Results: 13 studies (5 RCTs and 8 observational studies) with 19 depot-oral comparisons were included. Meta-analysis of adjusted endpoints resulted in RR [CI] of 0.88 [0.64-1.20] for RCTs, 0.62 [0.48-0.81] for prospective and 0.56 [0.44-0.71] for retrospective studies. These imply conversion factors of 1.41 and 1.57 between RCTs and prospective and retrospective designs, respectively.
Conclusions: Depot antipsychotics display increasing effectiveness compared with oral formulations as study design shifts from RCTs towards real-world clinical settings. Differences in control over compliance exerted in the respective designs could be a contributing factor. The estimated conversion factors quantify the average effect of study design on CE and facilitate comparison across studies.
Understand effect of study design on comparative effectiveness in schizophrenia
Understand application of cross-design synthesis methodology
Droitcour J, Silberman G, Chelimsky E. A new form of meta-analysis for combining results from randomized clinical trials and medical-practice databases. Int J Technol Assess Health Care. 1993;9(3):440–449. PubMed
DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7(3):177–188. PubMed