Quantitative Bioethics: The Promise and Challenges of Using Advanced Quantitative Methods in Empirical Bioethics

The increased scale and complexity of health data collected by hospitals and public health organizations has coincided with an explosion in advanced quantitative techniques. Tools like machine learning, simulation models, and quasi-experimental causal inference present opportunities and challenges for the field of empirical bioethics. This panel will examine using data science as a method for bioethical inquiry.

The first presenter, a philosophical bioethicist, will identify the bioethical questions amenable to quantitative bioethics – including resource allocation, health equity measurement, and equipoise in research ethics. They will demonstrate how advanced quantitative techniques can test bioethical interventions, rigorously assessing the intervention's performance per its stated goals, and quantify important tradeoffs (e.g. benefit versus risk of harm, individual versus population).

The second presenter, a health services researcher and bioethicist, will discuss the simulation of life-support allocation in crisis standards of care. They will discuss how advanced empirical methodologies can inform and influence normative ethical reasoning, specifically how algorithm design forces the designer to make specific normative choices that in turn leads to a re-examination of the underlying bioethical reasoning.

The third presenter, a clinician and healthcare delivery scientist, will discuss how implementing prediction models in healthcare generates new empirical bioethical questions. They will explain how advanced quantitative methods are often opaque to key stakeholders and how many technical assumptions by the data scientists encode value judgments. Finally, they will discuss the practical challenges of attempting to embed normative values and into machine learning predictors.