In the absence of direct evidence from randomized controlled trials (RCTs), real-world evidence (RWE) can play an important role in healthcare decision making. As part of a health technology assessment, we assessed the comparative risk of tuberculosis (TB) associated with using infliximab and etanercept in patients with rheumatoid arthritis.

We performed a systematic literature search using the PubMed database to identify relevant meta-analyses.

We located two relevant meta-analyses: one based on RCTs and one based on observational studies. Evidence from seven RCTs on infliximab (2,686 patients; 12 TB events) and two RCTs on etanercept (663 patients; 2 TB events) suggested no significant differences in the risk of TB between the two treatments, compared with placebo. In contrast, evidence from ten observational studies that directly compared the two treatments (443,941 patients; 253 TB events) indicated a significantly higher risk of TB with infliximab than with etanercept.

Although RWE is prone to confounding and bias, in this case it had the advantage of providing direct comparisons with larger sample sizes and longer follow up than evidence from RCTs. As a result, RWE was used to inform decision making on the risk of TB with infliximab and etanercept in patients with rheumatoid arthritis.