Step-by-step causal analysis of Electronic Health Records to ground decision making

Abstract

Accurate predictions, as with machine learning, may not suffice to provide optimal healthcare for every patient. Indeed, prediction can be driven by shortcuts in the data, such as racial biases. Causal thinking is needed for data-driven decisions. Here, we give an introduction to the key elements, focusing on routinely-collected data, electronic health records (EHRs) and claims data. Using such data to assess the value of an intervention requires care: temporal dependencies and existing practices easily confound the causal effect. We present a step-by-step framework to help build valid decision making from real-life patient records by emulating a randomized trial before individualizing decisions, eg with machine learning. Our framework highlights the most important pitfalls and considerations in analysing EHRs or claims data to draw causal conclusions. We illustrate the various choices in studying the effect of albumin on sepsis mortality in the Medical Information Mart for Intensive Care database (MIMIC-IV). We study the impact of various choices at every step, from feature extraction to causal-estimator selection. In a tutorial spirit, the code and the data are openly available.

Variation in monitoring: Glucose measurement in the ICU as a case study to preempt spurious correlations

“…Furthermore, one key assumption of causal inference models is positivity, i.e. subjects from the treatment and control group should have enough overlap in the distribution of their confounding variables (44). Usually researchers check this assumption by tabulating the data across the most pertinent variables which can become unwieldy in case of many strata.…”