Radiation biology. Radioecology
ISSN: 0869-8031

Archives → 2020 → V. 60. №5. → pp. 453–480

Article

Hill’s Criteria “Biological Plausibility”. Integration of Data from Various Disciplines for Epidemiology and Radiation Epidemiology

A. N. Koterov, L. N. Ushenkova, and A. P. Biryukova

A.I. Burnasyan Federal Medical Biophysical Center of Federal Medical Biological Agency, Moscow, Russia

Abstract

In the review the various aspects of «Biological plausibility» criterion, sometimes replaced by the «Coherence» criterion (consistency with well-known medical and biological knowledge) are considered. The importance of the criterion for epidemiological evidence of causation, especially for such disciplines (ecology, toxicology, carcinogenesis), in which there are difficulties not only to perform of experiments, but even to observe the effect, is noted. Only statistical approaches in epidemiology are incapable to prove the true causality for association (possibly the effect of chances, confounders, biases and reverse causation). Without knowledge of the biological mechanism and a plausible model, such relationship (especially for weak associations) cannot be regarded as confirmation of the true causation of the effect of the factor. The essence of the criterion is mean the integration of data from various biomedical disciplines, including experiments on animals and in vitro. There are three (Weed D. L., Hursting S. D., 1998) and four (Susser M., 1977; 1986) levels of attaining biological plausibility and coherence with existing knowledge. The methodologies for integrating data from various disciplines through Bayesian analysis (Bayesian meta-analysis) based on Weight of Evidence (WoE) and teleoanalysis are considered. The latter is a combination of data from different types of studies to quantify the causal relationship between two such associations, each of which can be proved, but to determine the relationship between the causality of the first and the final effect of the second is difficult for several reasons, including an ethical plan. The approach by teleoanalysis seems doubtful. Despite the need for the criterion «Biological plausibility», it, like almost all of Hill’s criteria (except for «Temporal») is neither necessary nor sufficient for evidence. Examples are presented (including the effects of radiation) which show first,
that «Biological plausibility» depends upon the biological knowledge on the day, and, second, that there are real, but seemingly implausible associations, as well as vice versa. This is base for criticism by some authors (A.R. Feinstein; K.J. Rothman and S. Greenland; B.G. Charlton; K. Goodman and C.V. Phillips) both specifically for the criterion «Biological plausibility» and the entire inductive approach based on causal criteria. However, the «Biological plausibility» criterion remains important for proving causality in epidemiological
studies, especially for those disciplines that public health relies on in making preventive decisions and developing safety standards.

Keywords

causality criterion ‘Biological plausibility’, Weight of Evidence, integration of data from biomedical disciplines, Bayesian meta-analysis, teleoanalysis