LAUSR

not with infliximab alone or in combination, suggesting a potential bias or lack of statistical power. A retrospective population-based observational cohort study (n = 889) from the French Midi-Pyr en ees region found no statistically significant higher risk of serious infection in patients with psoriasis exposed to biological agents compared with traditional systemic agents, probably because of a lack of statistical power. In a previous publication from the BADBIR, no significant increases in the risk of serious infection were observed for etanercept, adalimumab or ustekinumab compared with nonbiological systemic therapies or methotrexate only. The authors concluded that the risk of serious infection should not be a key discriminator for patients and clinicians when choosing between nonbiological systemic therapies, etanercept, adalimumab and ustekinumab for the treatment of psoriasis. Thus, the study by Yiu et al. provides strong evidence, with practical implications on therapeutic decisions, that infliximab is associated with an increased risk of serious infection compared with nonbiological and methotrexate-only treatment in a U.K. and Ireland-based psoriasis registry. Generalization of these results to wider geographical areas might require confirmation for several reasons, including different practices regarding definition of serious infections, indications for hospitalization and systemic treatment of infections, and differing criteria regarding the indications for infliximab, which might introduce a selection bias and confounding by indication (patients with ‘very severe disease’, defined as a total Psoriasis Area and Severity Index of 20 or more and a Dermatology Life Quality Index of more than 18, according to National Institute for Health and Care Excellence guidelines) not present in other populations. Marked differences in infection rates occur between individual countries. Furthermore, the potential effect of combination treatment with methotrexate should be addressed, eventual comparison with emerging biological treatments will be required, and development and validation of a multivariable prediction model for serious infection in patients treated with biologics will provide a much needed wider picture.