Can Robots Do Epidemiology? Machine Learning, Causal Inference, and Predicting the Outcomes of Public Health Interventions

Abstract

This paper argues that machine learning (ML) and epidemiology are on collision course over causation. The discipline of epidemiology lays great emphasis on causation, while ML research does not. Some epidemiologists have proposed imposing what amounts to a causal constraint on ML in epidemiology, requiring it either to engage in causal inference or restrict itself to mere projection. We whittle down the issues to the question of whether causal knowledge is necessary for underwriting predictions about the outcomes of public health interventions. While there is great plausibility to the idea that it is, conviction that something is impossible does not by itself motivate a constraint to forbid trying. We disambiguate the possible motivations for such a constraint into definitional, metaphysical, epistemological, and pragmatic considerations and argue that “Proceed with caution” (rather than “Stop!”) is the outcome of each. We then argue that there are positive reasons to proceed, albeit cautiously. Causal inference enforces existing classification schema prior to the testing of associational claims (causal or otherwise), but associations and classification schema are more plausibly discovered (rather than tested or justified) in a back-and-forth process of gaining reflective equilibrium. ML instantiates this kind of process, we argue, and thus offers the welcome prospect of uncovering meaningful new concepts in epidemiology and public health—provided it is not causally constrained.

Dengue as a Disease Threatening Global Health: A Narrative Review Focusing on Latin America and Brazil

“…Another important issue is technological advances and their association with epidemiology and public health. For example, data analysis mechanisms (through computer tools) in projection and prediction studies have been used systematically, aiming to maximize resources and optimize results in action planning, especially in countries with limited financial resources, such as those in Latin America [ 142 ]. The study by Cabrera et al [ 95 ] demonstrated that machine learning techniques could be successfully employed in dengue prediction models, being an important tool for Latin countries.…”

Malpractice Claims and Incident Reporting: Two Faces of the Same Coin?

“…These instruments have shown some limitations; as already reported in the literature regarding examples of A.I. in the medical field, algorithms may suffer from the poor reproducibility of the data set on which they were built and tested [ 6 , 7 ]. Moreover, although electronic medical records are now widely used, digitalization is not yet complete; this is why adverse events, such as falls, may not be noted, thus escaping A.I.…”