Improving interventional causal predictions in regulatory risk assessment

Challenging unverified assumptions in causal claims: Do gas stoves increase risk of pediatric asthma?

The use of unverified models for risk estimates and policy recommendations can be highly misleading, as their predictions may not reflect real-world health impacts. For example, a recent article states that NO2 from gas stoves "likely causes ∼50,000 cases of current pediatric asthma from long-term NO2 exposure alone" annually in the United States. This explicitly causal claim, which is contrary to several methodology and review articles published in this journal, among others, reflects both (a) An unverified modeling assumption that pediatric asthma burden is approximately proportional to NO2; and (b) An unverified causal assumption that the assumed proportionality between exposure and response is causal. The article is devoid of any causal analysis showing that these assumptions are likely to be true. It does not show that reducing NO2 exposure from gas stoves would reduce pediatric asthma risk. Its key references report no significant associations - let alone causation - between NO2 and pediatric asthma. Thus, the underlying data suggests that the number of pediatric asthma cases caused by gas stoves in the United States is indistinguishable from zero. This highlights the need to rigorously validate modeling assumptions and causal claims in public health risk assessments to ensure scientifically sound foundations for policy decisions.

An AI assistant to help review and improve causal reasoning in epidemiological documents

Drawing sound causal inferences from observational data is often challenging for both authors and reviewers. This paper discusses the design and application of an Artificial Intelligence Causal Research Assistant (AIA) that seeks to help authors improve causal inferences and conclusions drawn from epidemiological data in health risk assessments. The AIA-assisted review process provides structured reviews and recommendations for improving the causal reasoning, analyses and interpretations made in scientific papers based on epidemiological data. Causal analysis methodologies range from earlier Bradford-Hill considerations to current causal directed acyclic graph (DAG) and related models. AIA seeks to make these methods more accessible and useful to researchers. AIA uses an external script (a "Causal AI Booster" (CAB) program based on classical AI concepts of slot-filling in frames organized into task hierarchies to complete goals) to guide Large Language Models (LLMs), such as OpenAI's ChatGPT or Google's LaMDA (Bard), to systematically review manuscripts and create both (a) recommendations for what to do to improve analyses and reporting; and (b) explanations and support for the recommendations. Review tables and summaries are completed systematically by the LLM in order. For example, recommendations for how to state and caveat causal conclusions in the Abstract and Discussion sections reflect previous analyses of the Study Design and Data Analysis sections. This work illustrates how current AI can contribute to reviewing and providing constructive feedback on research documents. We believe that such AI-assisted review shows promise for enhancing the quality of causal reasoning and exposition in epidemiological studies. It suggests the potential for effective human-AI collaboration in scientific authoring and review processes.