However, improvements with tocilizumab monother apy were greater than aTNF monotherapy in terms of pain and self reported disease activity. Tocilizumab was at least molarity calculator as efficacious as aTNF regarding functional ability . In combination with MTX, aTNF, abatacept and tocilizumab showed comparable improvements in pain, self reported disease activity, and physical health as measured with the SF36 PCS component, whereas aTNF and tocilizumab showed the greatest improvements in HAQ DI. An interesting finding was that aTNFs as mono therapy seem less effective than aTNFs in combination with MTX. With tocilizumab as monotherapy, PROs similar to that of tocilizumab in combination with MTX were observed. The difference between aTNF as monotherapy and aTNF in combination with MTX can be considered clinically meaningful according to the defined MCID for pain, PGA and HAQ DI.

In addition to pain, self reported disease activity, functional ability, and physical health, we aimed to perform an analysis for fatigue as well. Fatigue is common in RA. Given the differences in fatigue scales used across studies we did not perform a network meta analysis for this endpoint. However, since fatigue is strongly associated with pain, and secondary associated with disease activity, it can be expected to find a similar pattern of efficacy across biologics for fatigue as obtained for pain and PGA. A limitation of the current analysis is that the study did not explicitly address differences in risk due to adverse events among treatments.

However, an analysis of relative short term RCT data would not provide a valid picture of the adverse event risk associated with long term use of biologics. The evidence of efficacy for all interventions was obtained from RCTs identified by means of a systematic literature review, which is a strength from an internal Brefeldin_A validity point of view. It is important to realize that the value of randomization holds within trials but not across trials. As such, there is the possibility that differences in study and patients characteristics across studies are modifiers of the treatment effects. This is a source of heterogeneity across studies comparing the same interventions, and a source of bias in the indirect comparison of treat ments.

There was some variation in duration of disease, lower swollen and tender joint count, and CRP across studies, but we did not observe systematic differences in the distribution of disease table 5 duration across different types of direct comparisons. As such, these factors can be a cause of heterogeneity but are likely not biasing the indirect comparisons. Of course, we can never exclude the possibility of unmeasured differences in patient characteristics across different comparisons. Although other network meta analysis of biologic treatments for RA have been published in the past few years, they focus on clinical outcomes such as the ACR response rates.