The Hill’s Criteria is one of the most cited frameworks for causal deduction in the field of epidemiology. The criteria consist of the strength of analogy, experiment, coherence, plausibility, the biological gradient, temporality, specificity, consistency, and association (Frank, Faber, & Stark, 2016).
The strength of association is the first criterion that posits that the larger the association between an exposure and disease, the more likely the causation (Frank et al., 2016). However, the determination of strength has evolved not only to take into account the magnitude of association but also statistical significance. Consistency refers to the reproduction of an unswerving association between two or more variables in line with the null hypothesis in numerous epidemiologic studies across varied settings, populations, and methods. Specificity, on the other hand, maintains that causality is only possible when the exposure leads to only one disease. Temporality has been termed as the only criterion that is universally agreeable. It holds that an exposure-disease association is only causal if the exposure precedes the onset of disease. The biological gradient criterion holds that an exposure-disease association is causal if a dose response is present. Biological plausibility criterion implies that causality can only be established if there is an interaction between epidemiology and biology (Frank et al., 2016). Coherence, just like plausibility, holds that the cause and its effect must concur with the basic knowledge and principles of science available to the researcher. The Hill’s experiment criterion explained that the risk of disease should decline with interventions or withdrawal of the exposure if a causal relationship is to be established. Finally, analogy implies that a strong causal relationship between an agent and disease should lower the threshold for acceptance of causality for a similar agent (Frank et al., 2016).
Strength of association correlates positively with consistency and temporality in the sense that when the strength of association increases, the consistency and temporality of an epidemiologic study increases too. Nonetheless, the correlation of temporality with consistency and strength of association cannot easily be established.
Zika Virus and Microcephaly
The characterization of relative risks and odds ratio from cohort and case-control studies respectively to determine the strength of association between Zika virus and Microcephaly is still underway. Significant increase in the incidence of microcephaly might not necessarily show a strong association because of other possible confounding cause or preconceptions (Frank et al., 2016). Moreover, a strong association would reveal a considerable increase in the cases of microcepahly in regions where the incidence of infection during pregnancy is significantly high.
Zika virus-microcephaly causal relationship has been inconsistent over the years. Even though the incidences of microcephaly relatively increased in countries that saw zika virus outbreak, such as Brazil and French Polynesia, this finding has not been established in zika virus endemic regions of Africa and Asia. Furthermore, other countries that experienced Zika virus outbreak in America have not recorded increased incidences of congenital microcephaly making the relationship inconsistent (Frank et al., 2016).
Microcephaly is a condition whose etiology is multifactorial which makes specificity a problematic criterion to determine. Nonetheless, only a specific impact of the virus on a developing brain can be determined. The temporality criterion in the Zika virus-microcephaly causal relationship from the Brazil and French Polynesia zika outbreaks show that indeed the infection precedes the condition (Frank et al., 2016). This, however, means that if the temporal pattern is upheld, the cases of congenital microcephaly in countries that experience outbreaks or endemicity will increase steadily.
Lastly, the zika virus-microcephaly causal relationship has not been shown to conform to the biological gradient criterion that anchor of dose response. Zika endemic regions of Africa and Asia have not been shown to exponentially record high incidences of microcephaly. Likewise, in other regions where zika virus has not been demonstrated, cases of congenital microcephaly still occur.
The criterion of strength of association was not met by the zika-microcephaly causal relationship as the connection is still being studied. This criterion is important because it seeks to establish the extent of causation between zika virus and microcephaly.
- Frank, C., Faber, M., & Stark, K. (2016). Causal or not: applying the Bradford Hill aspects of evidence to the association between Zika virus and microcephaly. EMBO Molecular Medicine, 8(4), 305–307.