A survey is reported of 113 systematic reviews of therapies in neonatology, based on 559 eligible randomized trials in total. These reviews were prepared by the CNRG and were published in the Cochrane Library, Issue 3,2001. The median number of included trials per review was 3 (range 0 to 32) and participants 207 (range 0 to 5460). Among 90 reviews with a categorical primary outcome, the median number of outcome events per review was 54 (range 1 to 1284). Among reviews finding a statistically significant benefit of treatment, the effect size was large (median relative risk 0.55, range 0.09 to 0.93). Reviews of surfactant for prevention and treatment of respiratory distress syndrome were able to detect moderate-sized treatment effects (median relative risk 0.85) because of the large number and size of trials in this field. Among many reviews finding no evidence of treatment effect, large and potentially important benefits or harms could not be excluded. Most CNRG reviews were current. There is a continuing need to prepare systematic review of therapies not yet covered and to keep an increasing number of reviews up-to-date.