Commentary: Tempering expectations of screening: what is the most authoritative advice we can give, given the data that we have?

The most authoritative basis for supporting a medical intervention is a meta-analysis of all sufficiently rigorous relevant randomized controlled trials. In this issue Saquib, Saquib and Ioannidis present an unprecedentedly thorough survey of 9 meta-analyses and 48 trials representing the best available evidence for the effectiveness of a range of screening interventions.1 Some of the evidence reviewed has been argued over before. In the case of breast cancer, probably the most debated screening intervention, a series of large trials in the 1970s and 1980s provided what seemed to be clear evidence that screening saved lives, and countries across the developed world introduced programmes. Yet, in 2000, a meta-analysis concluded that there was no reliable evidence for breast screening.2 The authors, Olsen and Gøtzsche, had identified eight trials but argued that the results of the six more or less favourable trials could not be trusted and that only the two more equivocal trials were sound.