| Criterion | Description |
|---|---|
| Strength | Whether those with the exposure are at a higher risk of developing disease and if so, how much more risk? This criterion suggests that a larger association increases the likelihood of causality. |
| Consistency | The credibility of findings increases with repetition of findings, including consistency of study findings across different populations and geographical locations. |
| Specificity | Causality is more likely if the exposure causes only one specific disease or syndrome, or if a specific location or population are being affected. |
| Temporality | This criterion requires that the exposure must occur before the disease, and not after a latency period that is too long. This criterion must always be fulfilled for causality to be concluded. |
| Biological gradient | The argument for causality is stronger in the presence of a dose–response relationship, where higher or longer exposure leads to an increased risk of disease. |
| Plausibility | A conceivable mechanism for causation between disease and exposure should exist for there to be a causal relationship. |
| Coherence | The current association should not contradict any previous knowledge available about the disease and/or exposure. |
| Experiment | This criterion can involve scientific experiments and addresses the association of exposure with disease. However, ‘experiment’ relates to the decrease in disease risk when the exposure is removed and often involves animal models. |
| Analogy | This criterion uses previous evidence of an association between a similar exposure and disease outcome to strengthen the current argument for causation. |
Source: [32]. |