ABSTRACT
One of the exciting developments around evidence-based medicine has been the emergence of new concepts that help to graphically illustrate and quantify how well a medicine works. Confidence intervals figure largely in many systematic reviews, in meta-analyses and for describing numbers needed to treat. Patients want to know their chance of getting better or being harmed, and that includes influences from all sources. People were randomly allocated to active vaccine or control, which were given in double-blind conditions in weeks 44–46 (November) before a peak of influenza incidence in Holland. The use of quality-adjusted life years in resource allocation decisions does mean that choices between patient groups competing for medical care are made explicit. A funnel plot aims to plot the trial's effect estimate against sample size and relies on the hypothesis that precision in the underlying treatment effect will increase as the sample size of the individual studies increases.
