How study designs influence comparative effectiveness outcomes: the case of oral versus long-acting injectable antipsychotic treatments for schizophrenia

This article reviews key methodological considerations for clinical trials that utilize explanatory and pragmatic trial designs and relates these contrasting approaches to the interpretation of results from comparisons of oral versus long-acting injectable (LAI) antipsychotics in schizophrenia. Explanatory randomized controlled trials (RCTs) generally measure the efficacy of a treatment in a homogeneous population with intensive, frequent, and often clinical trial-specific assessments. In contrast, pragmatic trials measure effectiveness in routine clinical practice and frequently aim to inform choices between treatments. Comparative effectiveness outcomes with pragmatic designs in naturalistic settings for schizophrenia treatments are of increasing interest to healthcare providers because outcomes of treatment (both efficacy and safety) may vary significantly when identified in an explanatory setting compared with a naturalistic pragmatic setting. Indeed, it has been suggested that the inconsistent outcomes observed in trials comparing oral and LAI antipsychotic medications may be a function of the use of explanatory or pragmatic trial designs. In practice, clinical trial designs are seldom purely explanatory or pragmatic. To identify the predominant orientation of a trial, one must consider multiple features. This paper reviews the relative impact of these features when comparing LAI and oral antipsychotic treatments and makes recommendations for improving these comparative designs.