TY - JOUR T1 - Multitreatment comparison meta-analysis: promise and peril JF - Evidence Based Medicine JO - Evid Based Med SP - 201 LP - 203 DO - 10.1136/eb-2013-101455 VL - 18 IS - 6 AU - Denise Campbell-Scherer Y1 - 2013/12/01 UR - http://ebm.bmj.com/content/18/6/201.abstract N2 - The multiple treatment comparison meta-analysis (MTC-MA) by Cahill et al1 described in this issue2 focuses on pharmacological interventions for smoking cessation. This MTC-MA provides a reminder of how useful this methodological approach can be. The efficacy of treatments for tobacco dependence is an excellent question for MTC-MA as smoking is prevalent and is a critical risk factor for many diseases with high morbidity and mortality; there is clinical interest in the relative efficacy of pharmacological interventions to aid cessation and direct head-to-head comparisons between each pair of interventions do not exist. Comparisons not based on direct research evidence are referred to as ‘indirect comparisons’. MTC-MA, also known as network meta-analysis, creates a network in which each node is a different intervention. MTC-MA provides a measure of treatment effect between all pairs of possible interventions in the network, including both direct and indirect comparisons, based on the existing data from controlled pair-wise comparisons in clinical trials.3 Practical interest in indirect comparisons has led to the development of rigorous approaches to conducting MTC-MA, and they are appearing more frequently in the clinical literature. An excellent summary for clinicians of how to use MTC-MA has been published recently in JAMA.4 Awareness of the criteria to look … ER -