Medical research publications are no longer restricted to case reports and presentation of hypotheses. Until the second half of the 20th century, expert opinions were (more or less) the only available evidence that further research could build upon, but expert opinion is now considered the lowest level of evidence. The highest level comes from scientific experiments, i.e. from well designed and performed randomised trials, thereafter from observational studies and in turn from time series evaluations. Study design, data collection, analysis strategy, calculations and statistical inference play now crucial roles in research because all empirical research on humans and animals is based on samples, and sampling variation makes scientific generalisation of the findings in a sample uncertain. This uncertainty needs to be minimised, quantified and presented when reporting research results. The methodology for this is statistics (“the science of uncertainty”). Statistical inference has become an integrated part of modern medical research.
All parts of a research project, from the development of a study hypothesis and a study design to collecting and analysing data statistically and interpreting the results and reporting the findings, impose practical and theoretical limitations on the outcome of the research. It is the statistical reviewer’s responsibility to verify that these limitations are clearly presented to the reader. In this role, the statistician evaluates evidence, not the fulfilment of underlying assumptions or clinical significance. However, it is part of the job to make sure that the authors distinguish between assumptions, opinions, and evidence. The main principle is that the author is responsible for providing the reader with the information that is necessary to interpret the validity, the uncertainty and the clinical significance of the findings.
Shor S. The responsibilities of a statistical reviewer. Chest 1972;61:486-487.