Errors in Relative Risks Reported in Figure 3 in a Network Meta-analysis of Cognitive Behavior Therapy Delivery Formats in Adults with Depression

To the Editor: The authors regret to report that Figure 3B was not represented correctly in their Original Investigation, “Effectiveness and Acceptability of Cognitive Behavior Therapy Delivery Formats in Adults With Depression: A Network Meta-analysis,”1 published in the July 2019 issue. In reviewing the article for a presentation, a coauthor detected the errors. In the original Figure 3B, we had shown the dropouts of care as usual over each of the treatment formats, instead of dropouts of the formats over care as usual. But Figure 3B supposed care as usual to be the common comparator, so the correct representation should give the dropouts of the various formats over care as usual. We therefore had to reverse the relative risks. Because of this adjustment, we had to change a number in the text and add a phrase for context, but none of these changes affect the interpretations or conclusions of the study.

Thus, we have requested that our article be corrected.2 The authors apologize for any inconvenience caused. This article was previously corrected on July 17, 2019, to fix a label error in Figure 3B.3