Treating Individuals: How Generalizable Are Trial Data?

The first article discusses the issues that can limit the generalisability of data from randomised trials and systematic reviews into clinical practice. Subsequent articles will reflect common clinician concerns about subgroups, meta-analyses, and risk.

Peter Rothwell (University of Oxford, UK), co-ordinator of the series and author of the first article, comments: "RCTs and systematic reviews cannot be expected to produce results that are directly relevant to all patients and all settings, but to be externally valid they should at least be designed and reported in a way that allows patients and clinicians to judge to whom they can reasonably be applied. While cognisant of the risks of over-regulation, some of the following recommendations might be worthwhile:

# Further research into the external validity of RCTs, particularly in relation to the measured treatment effect.

# Stricter requirements than previously for the external validity of RCTs submitted to pharmaceutical licensing authorities.

# Increased consideration of external validity in the CONSORT guidelines on the reporting of RCTs, and the Cochrane Collaboration guidelines on systematic reviews, and agreement on a checklist.

# The International Committee of Medical Journal Editors should require that all primary reports of RCTs or systematic reviews should contain a section entitled "To whom do these results apply?"

William Summerskill from The Lancet discusses the series in an accompanying commentary (p 13): "Let us not neglect the central role of patients as decision-makers in their own care. It is the responsibility of healthcare workers to communicate objective evidence in a manner which allows recipients to make an informed choice, and then to respect that choice. To empower consumers to make informed choices, the ability to interpret evidence and risk needs to be as much a life-skill for the public as for the profession. Towards this end the UK Department of Education's strategy to incorporate research design and data interpretation as part of the national science curriculum for secondary schools is to be applauded…Unless we improve our teaching, understanding, application, and communication of evidence, we risk denying patients the potential benefits derived from their peers who participated in research and of those who worked so hard to report and publish it."