In the article by Jani et al published in the August 2015 issue of Arthritis & Rheumatology (pages 2011–2019), typographical errors in two of the 95% confidence intervals shown in tables were introduced during the copyediting process. In Table 3, the regression coefficient (95% confidence interval) shown for adalimumab antidrug antibody status under “Multivariate model using GEE (ΔDAS28)” should have been 20.22 (20.78, 0.33). In Table 4, the regression coefficient (95% confidence interval) shown for etanercept level under “Univariate analysis using GEE” should have been 0.0080 (20.015, 0.031). The corrections do not change the messages of the report, as the above results were not reported to be significant.
We regret the errors.