During the writing, rewriting, and editing of this research paper by Helen Truby and colleagues (BMJ 2006;332:1309-11, 3 June), some small errors crept into table 3 of the full version on bmj.com (table 2 of the print version). In the control group the mean (SD) fall in cholesterol during 2-6 months should be 0.24 (0.6) (not 0.24 (0.24) as written) and during 0-6 months should be 0.18 (0.5) (not 0.5 (0.18)). The table footnote should have stated that the fall in total cholesterol at 2 months in the Weight Watchers, Rosemary Conley, and Slim-Fast groups was significantly different from that in the control group. However, the text is correct in explaining these differences, and the conclusions drawn are not affected by the errors.