MOVING BEYOND RANDOMIZED CONTROLLED TRIALS

We concur with Victora et.al.1 and other contributors to the important evaluation issue of the Journal, which highlights a long-standing concern among many in the public health community about the overreliance on randomized controlled trials (RCTs) in research.2 Clearly, RCTs have an important role to play in medical research, although their strength may be limited to assessing well-controlled, narrow interventions, for example, comparing drug A against drug B or procedure A against procedure B. Increasingly, however, public health interventions and the funders who invest in those interventions acknowledge the multifactorial basis of many health outcomes and thus the need for more sophisticated, multidimensional, community-based designs. RCTs have limited capacity to assess such comprehensive initiatives. They also are limited in assessing aspects of quality3 as well as performance measurement.4

If we are to understand the effectiveness of complex community interventions, needed in addition are more robust approaches, ranging from Theory of Change5 to hybrid models combining several traditional approaches,6 modeling, anthropological and sociological tools (focus group dynamics, social network analysis), and syndemic relational analysis.7

Finally, from a population perspective, our experience with immigrants and refugees indicates that trust, choice (patient, family, even community), and client participation are critical prerequisites for successful program design and evaluation.8 Such patient-centeredness is inconsistent with the RCT “randomized and blinded” methodology. Interventions require research designs commensurate with their sophistication.