Toward practical causal epidemiology

Population attributable fraction (PAF), probability of causation, burden of disease, and related quantities derived from relative risk ratios are widely used in applied epidemiology and health risk analysis to quantify the extent to which reducing or eliminating exposures would reduce disease risks....

Descripción completa

Guardado en:
Detalles Bibliográficos
Autor principal: Louis Anthony Cox, Jr
Formato: article
Lenguaje:EN
Publicado: Elsevier 2021
Materias:
Acceso en línea:https://doaj.org/article/bb2207714bbc48ba8dcec8671c215be9
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:bb2207714bbc48ba8dcec8671c215be9
record_format dspace
spelling oai:doaj.org-article:bb2207714bbc48ba8dcec8671c215be92021-11-14T04:35:20ZToward practical causal epidemiology2590-113310.1016/j.gloepi.2021.100065https://doaj.org/article/bb2207714bbc48ba8dcec8671c215be92021-11-01T00:00:00Zhttp://www.sciencedirect.com/science/article/pii/S2590113321000195https://doaj.org/toc/2590-1133Population attributable fraction (PAF), probability of causation, burden of disease, and related quantities derived from relative risk ratios are widely used in applied epidemiology and health risk analysis to quantify the extent to which reducing or eliminating exposures would reduce disease risks. This causal interpretation conflates association with causation. It has sometimes led to demonstrably mistaken predictions and ineffective risk management recommendations. Causal artificial intelligence (CAI) methods developed at the intersection of many scientific disciplines over the past century instead use quantitative high-level descriptions of networks of causal mechanisms (typically represented by conditional probability tables or structural equations) to predict the effects caused by interventions. We summarize these developments and discuss how CAI methods can be applied to realistically imperfect data and knowledge – e.g., with unobserved (latent) variables, missing data, measurement errors, interindividual heterogeneity in exposure-response functions, and model uncertainty. We recommend that CAI methods can help to improve the conceptual foundations and practical value of epidemiological calculations by replacing association-based attributions of risk to exposures or other risk factors with causal predictions of the changes in health effects caused by interventions.Louis Anthony Cox, JrElsevierarticleCausalityCausal artificial intelligencePopulation attributable fractionProbability of causationRisk analysisStatistical methodsInfectious and parasitic diseasesRC109-216ENGlobal Epidemiology, Vol 3, Iss , Pp 100065- (2021)
institution DOAJ
collection DOAJ
language EN
topic Causality
Causal artificial intelligence
Population attributable fraction
Probability of causation
Risk analysis
Statistical methods
Infectious and parasitic diseases
RC109-216
spellingShingle Causality
Causal artificial intelligence
Population attributable fraction
Probability of causation
Risk analysis
Statistical methods
Infectious and parasitic diseases
RC109-216
Louis Anthony Cox, Jr
Toward practical causal epidemiology
description Population attributable fraction (PAF), probability of causation, burden of disease, and related quantities derived from relative risk ratios are widely used in applied epidemiology and health risk analysis to quantify the extent to which reducing or eliminating exposures would reduce disease risks. This causal interpretation conflates association with causation. It has sometimes led to demonstrably mistaken predictions and ineffective risk management recommendations. Causal artificial intelligence (CAI) methods developed at the intersection of many scientific disciplines over the past century instead use quantitative high-level descriptions of networks of causal mechanisms (typically represented by conditional probability tables or structural equations) to predict the effects caused by interventions. We summarize these developments and discuss how CAI methods can be applied to realistically imperfect data and knowledge – e.g., with unobserved (latent) variables, missing data, measurement errors, interindividual heterogeneity in exposure-response functions, and model uncertainty. We recommend that CAI methods can help to improve the conceptual foundations and practical value of epidemiological calculations by replacing association-based attributions of risk to exposures or other risk factors with causal predictions of the changes in health effects caused by interventions.
format article
author Louis Anthony Cox, Jr
author_facet Louis Anthony Cox, Jr
author_sort Louis Anthony Cox, Jr
title Toward practical causal epidemiology
title_short Toward practical causal epidemiology
title_full Toward practical causal epidemiology
title_fullStr Toward practical causal epidemiology
title_full_unstemmed Toward practical causal epidemiology
title_sort toward practical causal epidemiology
publisher Elsevier
publishDate 2021
url https://doaj.org/article/bb2207714bbc48ba8dcec8671c215be9
work_keys_str_mv AT louisanthonycoxjr towardpracticalcausalepidemiology
_version_ 1718429919515508736