Background Several multiple treatment comparison meta-analysis (MTCs) have been conducted over the past years to establish comparative effectiveness between biologic disease modifying anti rheumatic drugs (bDMARDs). Unfortunately, major inconsistencies exist in the findings across these MTCs.
Objectives To methodologically review the published literature on rheumatoid arthritis MTCs. To identify methodological issues that can explain the substantial discrepancies in the findings of these MTCs.
Methods We searched MEDLINE for rheumatoid arthritis multiple treatment comparisons. Following the PRISMA guidelines, we extracted a large set of methodological items from the identified reviews. These included, but were not limited to, inclusion/exclusion criteria, information sources (e.g., MEDLINE), approaches to dealing with monotherapies versus combination therapies, approaches to dealing with potential covariate effect modifiers (i.e., sources of heterogeneity).
Results We identified 12 published MTC, of which 7 were published since 2009. We identified major discrepancies in the inclusion of trials, despite highly similar eligibility criteria and literature searches. The total number of trials covered among all MTCs was 61. However, the number of trials included in the individual MTCs published since 2009 spanned from 15 to 31 – i.e., 25% to 50% of all available trials. We identified inconsistencies in approaches to dealing with monotherapy and combination therapy trials. Most MTCs lumped the two sets of trials without either controlling for the effect of concomitant use of disease modifying anti rheumatic drugs (DMARDs) or separately comparing the effectiveness estimates in the two. Approximately half of the identified MTCs did not explore potential sources of heterogeneity. Among those that did, the explored sources were inconsistent.
Conclusions Major methodological shortcomings and inconsistencies exist throughout published rheumatoid arthritis MTCs. There are many lessons to be learned from these previous publications which can potentially strengthen the current evidence base.
Disclosure of Interest None Declared