Table 3

Examples of arguments for convergent lines of evidence in communicable disease control

Example of outbreakExample of quantitative evidence supporting causal estimationExample of qualitative or case-study evidence supporting causal explanationCase for convergence
Cholera, 1854 45 46‘Snow (…) observed that two water companies served this area (and) that households receiving water from the two companies were intermingled on the same streets. These insights provided the basis for assuming that allocation of contaminated water occurred as-if at random, thus justifying the (quantitative) natural experiment (that) yielded an unusually decisive confirmation of his main hypothesis (…) that cholera spread through human contact and was waterborne.’‘Early on, he abandoned the established explanation that cholera spread through “miasmas”(…) Snow observed that (…) the first case in London involved the death of a sailor just arrived from Hamburg, where there was a cholera epidemic. The second was an individual who subsequently slept in the same boarding house room as this sailor.’‘Snow was ‘intimately familiar with the (…) area because of his medical practice’’, which contributed to astute inferences about the incidence of cholera in particular neighbourhoods, workplaces and households (…) Snow arrived at these insights prior to the confirmation provided by this experiment (and) the construction of this remarkable study was heavily dependent on (qualitative) causal process observations.
COVID-19, 202047Absence of trial evidence is partly due to the fact that experimental studies of mass public health measures are usually impractical (…) There are now many natural experiments of the wearing of masks or face coverings in COVID-19.’‘Another piece of evidence that covering the face could make a big difference is super-spreader events (…) Perhaps the most dramatic is the choir practice in Seattle, in which (…) 45 of 60 people became infected and 2 (so far) have died.’‘I am struck by the stories they did not examine (the COVID-stricken choir, (etc)) But these stories (…) pull together complex chains of influence and remind us that causality (…) is rarely linear (…) All these various streams of evidence contribute, in different ways and at different levels, to strengthen the argument (…) As with other public health measures, we should make a decision based on an assessment of the full body of evidence.’
  • Emphases added.