Table 3.

A pragmatic dialogue approach to causal inferences about an agent or determinant A with respect to a disease D: Evidence increasing or decreasing confidence in a potential causal relation between A and D.

Type of evidence	Increasing confidence	Decreasing confidence
From prior knowledge	Results conform to predictions from theoretical considerations and/or prior knowledge about specificity of outcome, specificity of type of exposure, or specificity regarding the outcome in different subgroups of the population.	Although there are sound arguments for specificity of outcome, specificity of type of exposure, or specificity regarding the outcome in different subgroups of the population, data do not conform to these expectations.
	Association between A and D is coherent with biologic knowledge and/or a plausible mechanistic model of action can be delineated.	There is knowledge about mechanism of action that indicates lack of effect of A on D.
From epidemiology	Strength of association between A and D exceeds that of potential confounders.	There are known confounders not considered in existing investigations strong enough to explain the observed effect.
	Association between A and D is consistently observed in different populations, with different types of studies, or in different time intervals.	There is substantial heterogeneity in the effect of A on D in different populations, different study types, or different time intervals.
	Manipulating A in the population changes pattern and/or frequency of D.	Manipulating A in the population does not affect occurrence of D.
	In the case a meaningful meter of the “dose” of A can be defined, there exists a dose–response relationship.	A meaningful “dose” meter can be defined but the relationship between “dose” and response is not monotonous.
From animal studies	Long-term animal studies in different species indicate an association between A and D (or D′, an analogue of D in these species).	There exist animal models of the disease D, and in none of these models A is effective.
	A enhances the effect of a known pathogen B.	No promoting or antagonizing effect of A with a variety of other agents could be found in different exposure regimes relevant for human exposures.
	In animal experiments, intermediate steps of the pathogenic process can be evoked by exposure to A.	In different species that are sensitive to other exposures producing effects expected to be similar to those of A, the latter is ineffective.
From in vitro studies	Exposed cells or tissues react or get damaged by exposure to A consistent with the pathogenic process of D.	In cell lines or tissues sensitive to exposures similar to A, no effect of exposure to A is found.
	Upstream events can be observed by exposure to A that may lead to D in the intact organism.	No changes in cellular processes or alterations of signaling pathways can be evoked by exposure to A.
	A enhances the effect of a known cellular pathogen B.	No promoting or antagonizing effect of A with a variety of other agents could be found.