Setting new research in the context of previous research: some options

If a new clinical trial is to be justifiable both scientifically and ethically it should be designed in the light of an assessment of relevant previous research, ideally a systematic review. When its findings are reported, these should be set in the context of updated reviews of other, similar research.1

When reading the report of a new controlled trial, interpretation will be greatly aided if the Discussion sets the results in the context of the results from available similar research.1 ,2 Indeed, the CONSORT statement3 (item 22) suggests including ‘a formal systematic review in the results or discussion section of the report’. However, CONSORT observes that ‘Such synthesis may be judged impractical for trial authors, but it is often possible to quote a systematic review of similar trials.’ For practical reasons, most authors are likely to choose the latter option, but this still leaves the reader to mentally weigh up the previous systematic review results and the new trial results. We have considered whether there might be better options and aimed to enumerate them here.

As an example, consider a meta-analysis4 of lower vs versus usual targets for blood pressure lowering containing 6 trials and 41 491 patients which found an overall relative risk (RR) of 0.91 with 95% CI 0.80 to 1.04 and a new trial5 with 8511 patients with an RR 0.74; 95% CI 0.60 to 0.92. What do these mean together? The figure 1 shows these results as well as the combined estimate. The addition of the new trial has increased the effect size in the meta-analysis from a 9% relative reduction to a 14% relative reduction and the CI of the new estimate now excludes the null result (RR: 1.00), so is statistically significant.