Preventing injuries in children: cluster randomised controlled trial in primary care

This general practice paper by Denise Kendrick and colleagues (10 April, pp 980-3) contains three errors. The calculation of sample size was based on “a mean cluster size of 60 [not 60%]” (first sentence, p 981); in table 3 (p 982) the number of children who received advice at the 18-24 month check should have been 535 rather than 35; and in table 4 (p 982) the number of children in the control group who had any medically attended injury should have been 330 rather than 220.