Abstract:
Over the past year I have been involved in examining the research practices of the Bandim Health Project, who for 40+ years have studied the non-specific effects (NSEs) of vaccines. More than 90,000 children have been included in a long series of Randomized Controlled Trials, but questions have arisen about what these trials have actually found. In my presentation, I will present our approach to reevaluating their evidence from a statistical point of view with a sharp focus on the need to distinguish between primary and secondary findings and the need to control the rate of false positives. I will finally discuss what this means for the way we make inference, first with respect to trials, but second with respect to observational studies as well – how do we investigate whether an effect is real or not?