{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T01:18:56Z","timestamp":1790126336354,"version":"4.0.1"},"reference-count":93,"publisher":"Public Library of Science (PLoS)","issue":"2","license":[{"start":{"date-parts":[[2025,2,3]],"date-time":"2025-02-03T00:00:00Z","timestamp":1738540800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosdigitalhealth.org"],"crossmark-restriction":false},"short-container-title":["PLOS Digit Health"],"abstract":"<jats:p>Causal inference enables machine learning methods to estimate treatment effects of medical interventions from electronic health records (EHRs). The prevalence of such observational data and the difficulty for randomized controlled trials (RCT) to cover all population\/treatment relationships make these methods increasingly attractive for studying causal effects. However, researchers should be wary of many pitfalls. We propose and illustrate a framework for causal inference estimating the effect of albumin on mortality in sepsis using an Intensive Care database (MIMIC-IV) and comparing various sensitivity analyses to results from RCTs as gold-standard. The first step is study design, using the target trial concept and the PICOT framework: Population (patients with sepsis), Intervention (combination of crystalloids and albumin for fluid resuscitation), Control (crystalloids only), Outcome (28-day mortality), Time (intervention start within 24h of admission). We show that too large treatment-initiation times induce immortal time bias. The second step is selection of the confounding variables based on expert knowledge. Increasingly adding confounders enables to recover the RCT results from observational data. As the third step, we assess the influence of multiple models with varying assumptions, showing that a doubly robust estimator (AIPW) with random forests proved to be the most reliable estimator. Results show that these steps are all important for valid causal estimates. A valid causal model can then be used to individualize decision making: subgroup analyses showed that treatment efficacy of albumin was better for patients &gt;60 years old, males, and patients with septic shock. Without causal thinking, machine learning is not enough for optimal clinical decision on an individual patient level. Our step-by-step analytic framework helps avoiding many pitfalls of applying machine learning to EHR data, building models that avoid shortcuts and extract the best decision-making evidence.<\/jats:p>","DOI":"10.1371\/journal.pdig.0000721","type":"journal-article","created":{"date-parts":[[2025,2,3]],"date-time":"2025-02-03T13:53:23Z","timestamp":1738590803000},"page":"e0000721","update-policy":"https:\/\/doi.org\/10.1371\/journal.pdig.corrections_policy","source":"Crossref","is-referenced-by-count":10,"title":["Step-by-step causal analysis of EHRs to ground decision-making"],"prefix":"10.1371","volume":"4","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8072-9966","authenticated-orcid":true,"given":"Matthieu","family":"Doutreligne","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tristan","family":"Struja","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Judith","family":"Abecassis","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Claire","family":"Morgand","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Leo Anthony","family":"Celi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ga\u00ebl","family":"Varoquaux","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2025,2,3]]},"reference":[{"key":"pdig.0000721.ref001","doi-asserted-by":"crossref","unstructured":"Rajkomar A, Oren E, Chen K, Dai AM, Hajaj N, Hardt M, et al. Scalable and accurate deep learning with electronic health records. NPJ Digit Med. 2018;118. doi: 10.1038\/s41746-018-0029-1 31304302","DOI":"10.1038\/s41746-018-0029-1"},{"issue":"6","key":"pdig.0000721.ref002","doi-asserted-by":"crossref","first-page":"e271","DOI":"10.1016\/S2589-7500(19)30123-2","article-title":"A comparison of deep learning performance against health-care professionals in detecting diseases from medical imaging: a systematic review and meta-analysis","volume":"1","author":"X Liu","year":"2019","journal-title":"Lancet Digit Health"},{"issue":"1","key":"pdig.0000721.ref003","doi-asserted-by":"crossref","first-page":"7155","DOI":"10.1038\/s41598-020-62922-y","article-title":"BEHRT: Transformer for electronic health records","volume":"10","author":"Y Li","year":"2020","journal-title":"Sci Rep"},{"issue":"1","key":"pdig.0000721.ref004","doi-asserted-by":"crossref","first-page":"62","DOI":"10.1038\/s41746-021-00426-3","article-title":"Machine learning for patient risk stratification: standing on, or looking over, the shoulders of clinicians?","volume":"4","author":"BK Beaulieu-Jones","year":"2021","journal-title":"PJ Digit Med"},{"issue":"1","key":"pdig.0000721.ref005","doi-asserted-by":"crossref","first-page":"65","DOI":"10.1038\/s41746-021-00438-z","article-title":"Diagnostic accuracy of deep learning in medical imaging: a systematic review and meta-analysis","volume":"4","author":"R Aggarwal","year":"2021","journal-title":"NPJ Digit Med"},{"issue":"12","key":"pdig.0000721.ref006","doi-asserted-by":"crossref","first-page":"866","DOI":"10.7326\/M18-1990","article-title":"Ensuring fairness in machine learning to advance health equity","volume":"169","author":"A Rajkomar","year":"2018","journal-title":"Ann Intern Med"},{"issue":"4","key":"pdig.0000721.ref007","doi-asserted-by":"crossref","first-page":"e0000023","DOI":"10.1371\/journal.pdig.0000023","article-title":"Generalizability challenges of mortality risk prediction models: A retrospective analysis on a multi-center database","volume":"1","author":"H Singh","year":"2022","journal-title":"PLOS Digit Health"},{"issue":"6","key":"pdig.0000721.ref008","doi-asserted-by":"crossref","first-page":"e406","DOI":"10.1016\/S2589-7500(22)00063-2","article-title":"AI recognition of patient race in medical imaging: a modelling study","volume":"4","author":"JW Gichoya","year":"2022","journal-title":"Lancet Digit Health"},{"issue":"12","key":"pdig.0000721.ref009","doi-asserted-by":"crossref","first-page":"2176","DOI":"10.1038\/s41591-021-01595-0","article-title":"Underdiagnosis bias of artificial intelligence algorithms applied to chest radiographs in under-served patient populations","volume":"27","author":"L Seyyed-Kalantari","year":"2021","journal-title":"Nat Med"},{"issue":"11","key":"pdig.0000721.ref010","doi-asserted-by":"crossref","first-page":"665","DOI":"10.1038\/s42256-020-00257-z","article-title":"Shortcut learning in deep neural networks","volume":"2","author":"R Geirhos","year":"2020","journal-title":"Nat Mach Intell"},{"issue":"10","key":"pdig.0000721.ref011","doi-asserted-by":"crossref","first-page":"1135","DOI":"10.1001\/jamadermatol.2019.1735","article-title":"Association between surgical skin markings in dermoscopic images and diagnostic performance of a deep learning convolutional neural network for melanoma recognition","volume":"155","author":"JK Winkler","year":"2019","journal-title":"JAMA Dermatol"},{"issue":"7","key":"pdig.0000721.ref012","doi-asserted-by":"crossref","first-page":"610","DOI":"10.1038\/s42256-021-00338-7","article-title":"AI for radiographic COVID-19 detection selects shortcuts over signal","volume":"3","author":"AJ DeGrave","year":"2021","journal-title":"Nat Mach Intell"},{"key":"pdig.0000721.ref013","doi-asserted-by":"crossref","unstructured":"Badgeley MA, Zech JR, Oakden-Rayner L, Glicksberg BS, Liu M, Gale W, et al. Deep learning predicts hip fracture using confounding patient and healthcare variables. NPJ Digit Med. 2019;231. doi: 10.1038\/s41746-019-0105-1 31304378","DOI":"10.1038\/s41746-019-0105-1"},{"issue":"6464","key":"pdig.0000721.ref014","doi-asserted-by":"crossref","first-page":"447","DOI":"10.1126\/science.aax2342","article-title":"Dissecting racial bias in an algorithm used to manage the health of populations","volume":"366","author":"Z Obermeyer","year":"2019","journal-title":"Science"},{"issue":"1","key":"pdig.0000721.ref015","doi-asserted-by":"crossref","first-page":"1107","DOI":"10.1038\/s41467-021-21390-2","article-title":"Temporal bias in case-control design: preventing reliable predictions of the future","volume":"12","author":"W Yuan","year":"2021","journal-title":"Nat Commun"},{"issue":"8","key":"pdig.0000721.ref016","doi-asserted-by":"crossref","first-page":"1065","DOI":"10.1001\/jamainternmed.2021.2626","article-title":"External validation of a widely implemented proprietary sepsis prediction model in hospitalized patients","volume":"181","author":"A Wong","year":"2021","journal-title":"JAMA Intern Med"},{"issue":"7","key":"pdig.0000721.ref017","doi-asserted-by":"crossref","first-page":"369","DOI":"10.1038\/s42256-020-0197-y","article-title":"Causal inference and counterfactual prediction in machine learning for actionable healthcare","volume":"2","author":"M Prosperi","year":"2020","journal-title":"Nat Mach Intell"},{"key":"pdig.0000721.ref018","unstructured":"Plecko D, Bareinboim E. Causal fairness analysis. arXiv preprint arXiv:220711385. 2022."},{"issue":"3","key":"pdig.0000721.ref019","doi-asserted-by":"crossref","first-page":"219","DOI":"10.1136\/thx.2006.066837","article-title":"External validity of randomised controlled trials in asthma: to whom do the results of the trials apply?","volume":"62","author":"J Travers","year":"2007","journal-title":"Thorax"},{"key":"pdig.0000721.ref020","doi-asserted-by":"crossref","unstructured":"Averitt AJ, Weng C, Ryan P, Perotte A. Translating evidence into practice: eligibility criteria fail to eliminate clinically significant differences between real-world and study populations. NPJ Digit Med. 2020;367. doi: 10.1038\/s41746-020-0277-8 32411828","DOI":"10.1038\/s41746-020-0277-8"},{"issue":"1","key":"pdig.0000721.ref021","doi-asserted-by":"crossref","first-page":"170","DOI":"10.1038\/s41746-021-00542-0","article-title":"Broadening the reach of the FDA Sentinel system: A roadmap for integrating electronic health record data in a causal analysis framework","volume":"4","author":"RJ Desai","year":"2021","journal-title":"NPJ Digit Med"},{"issue":"1","key":"pdig.0000721.ref022","doi-asserted-by":"crossref","first-page":"58","DOI":"10.1038\/s41746-023-00794-y","article-title":"A standardized framework for risk-based assessment of treatment effect heterogeneity in observational healthcare databases","volume":"6","author":"A Rekkas","year":"2023","journal-title":"NPJ Digit Med"},{"key":"pdig.0000721.ref023","doi-asserted-by":"crossref","first-page":"70","DOI":"10.1016\/j.jclinepi.2016.04.014","article-title":"Specifying a target trial prevents immortal time bias and other self-inflicted injuries in observational analyses","volume":"79","author":"MA Hern\ufffdn","year":"2016","journal-title":"J Clin Epidemiol"},{"issue":"9596","key":"pdig.0000721.ref024","doi-asserted-by":"crossref","first-page":"1453","DOI":"10.1016\/S0140-6736(07)61602-X","article-title":"The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies","volume":"370","author":"E von Elm","year":"2007","journal-title":"Lancet"},{"issue":"10","key":"pdig.0000721.ref025","doi-asserted-by":"crossref","first-page":"e1001885","DOI":"10.1371\/journal.pmed.1001885","article-title":"The REporting of studies Conducted using Observational Routinely-collected health Data (RECORD) statement","volume":"12","author":"EI Benchimol","year":"2015","journal-title":"PLoS Med"},{"key":"pdig.0000721.ref026","unstructured":"Hern\u00e1n MA, Robins JM. Causal inference: What if? 2020."},{"issue":"5","key":"pdig.0000721.ref027","doi-asserted-by":"crossref","first-page":"658","DOI":"10.1210\/endrev\/bnab007","article-title":"Conducting real-world evidence studies on the clinical outcomes of diabetes treatments","volume":"42","author":"S Schneeweiss","year":"2021","journal-title":"Endocr Rev"},{"issue":"1","key":"pdig.0000721.ref028","doi-asserted-by":"crossref","first-page":"1014","DOI":"10.1038\/s41467-022-28546-8","article-title":"Uncovering interpretable potential confounders in electronic medical records","volume":"13","author":"J Zeng","year":"2022","journal-title":"Nat Commun"},{"issue":"4","key":"pdig.0000721.ref029","doi-asserted-by":"crossref","first-page":"492","DOI":"10.1093\/aje\/kwm324","article-title":"Immortal time bias in pharmaco-epidemiology","volume":"167","author":"S Suissa","year":"2008","journal-title":"Am J Epidemiol"},{"key":"pdig.0000721.ref030","unstructured":"Oke J, Fanshawe T, Nunan D. Lead time bias, catalogue of bias collaboration. 2021. Available from: https:\/\/catalogofbias.org\/biases\/lead-time-bias\/."},{"key":"pdig.0000721.ref031","doi-asserted-by":"crossref","unstructured":"Fu EL, Evans M, Carrero JJ, Putter H, Clase CM, Caskey FJ. Timing of dialysis initiation to reduce mortality and cardiovascular events in advanced chronic kidney disease: nationwide cohort study. BMJ. 2021;375.","DOI":"10.1136\/bmj-2021-066306"},{"key":"pdig.0000721.ref032","doi-asserted-by":"crossref","unstructured":"Wang SV, Sreedhara SK, Bessette LG, Schneeweiss S. Understanding variation in the results of real-world evidence studies that seem to address the same question. J Clin Epidemiol. 2022;151161\u201370. doi: 10.1016\/j.jclinepi.2022.08.012 36075314","DOI":"10.1016\/j.jclinepi.2022.08.012"},{"key":"pdig.0000721.ref033","unstructured":"Bankhead C ND Aronson JK. Attrition bias, Catalogue of Bias Collaboration. 2017. Available from: https:\/\/catalogofbias.org\/biases\/attrition-bias\/."},{"key":"pdig.0000721.ref034","unstructured":"Greenland S, Pearl J, Robins J. Causal diagrams for epidemiologic research. Epidemiology. 199937\u201348."},{"issue":"3","key":"pdig.0000721.ref035","doi-asserted-by":"crossref","first-page":"211","DOI":"10.1007\/s10654-019-00494-6","article-title":"Principles of confounder selection","volume":"34","author":"TJ VanderWeele","year":"2019","journal-title":"Eur J Epidemiol"},{"issue":"3","key":"pdig.0000721.ref036","doi-asserted-by":"crossref","first-page":"607","DOI":"10.1002\/sim.8792","article-title":"Confounder selection strategies targeting stable treatment effect estimators","volume":"40","author":"WW Loh","year":"2021","journal-title":"Stat Med"},{"issue":"16","key":"pdig.0000721.ref037","doi-asserted-by":"crossref","first-page":"1376","DOI":"10.1001\/jama.2023.4221","article-title":"Emulation of randomized clinical trials with nonrandomized database analyses: results of 32 clinical trials","volume":"329","author":"SV Wang","year":"2023","journal-title":"JAMA"},{"issue":"2","key":"pdig.0000721.ref038","doi-asserted-by":"crossref","first-page":"29","DOI":"10.1257\/jep.28.2.29","article-title":"High-dimensional methods and inference on structural and treatment effects","volume":"28","author":"A Belloni","year":"2014","journal-title":"J Econ Perspect"},{"key":"pdig.0000721.ref039","doi-asserted-by":"crossref","unstructured":"Chernozhukov V, Chetverikov D, Demirer M, Duflo E, Hansen C, Newey W. Double\/debiasing machine learning for treatment and structural parameters. 2018.","DOI":"10.3386\/w23564"},{"key":"pdig.0000721.ref040","article-title":"Causal Inference for Observational studies: Tutorial","author":"U Shalit","year":"2016"},{"key":"pdig.0000721.ref041","unstructured":"Sharma A. Tutorial on causal inference and counterfactual reasoning. 2018. Available from: https:\/\/causalinference.gitlab.io\/kdd-tutorial\/."},{"issue":"12","key":"pdig.0000721.ref042","doi-asserted-by":"crossref","first-page":"3041","DOI":"10.1007\/s10115-021-01621-0","article-title":"Causal inference for time series analysis: problems, methods and evaluation","volume":"63","author":"R Moraffah","year":"2021","journal-title":"Knowl Inf Syst"},{"issue":"1","key":"pdig.0000721.ref043","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1214\/09-STS313","article-title":"Matching methods for causal inference: A review and a look forward","volume":"25","author":"EA Stuart","year":"2010","journal-title":"Stat Sci"},{"issue":"28","key":"pdig.0000721.ref044","doi-asserted-by":"crossref","first-page":"3661","DOI":"10.1002\/sim.6607","article-title":"Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studies","volume":"34","author":"PC Austin","year":"2015","journal-title":"Stat Med"},{"issue":"3","key":"pdig.0000721.ref045","doi-asserted-by":"crossref","first-page":"392","DOI":"10.1093\/oxfordjournals.aje.a114254","article-title":"The role of model selection in causal inference from nonexperimental data","volume":"123","author":"JM Robins","year":"1986","journal-title":"Am J Epidemiol"},{"issue":"1","key":"pdig.0000721.ref046","first-page":"7489","article-title":"Generalization bounds and representation learning for estimation of potential outcomes and causal effects","volume":"23","author":"FD Johansson","year":"2022","journal-title":"J Machine Learn Res."},{"issue":"1","key":"pdig.0000721.ref047","doi-asserted-by":"crossref","first-page":"65","DOI":"10.1093\/aje\/kww165","article-title":"Targeted maximum likelihood estimation for causal inference in observational studies","volume":"185","author":"MS Schuler","year":"2017","journal-title":"Am J Epidemiol"},{"issue":"1","key":"pdig.0000721.ref048","doi-asserted-by":"crossref","first-page":"43","DOI":"10.1214\/18-STS667","article-title":"Automated versus do-it-yourself methods for causal inference","volume":"34","author":"V Dorie","year":"2019","journal-title":"Statistical Science."},{"key":"pdig.0000721.ref049","article-title":"Validating causal inference models via influence functions. In: International Conference on Machine Learning. PMLR","author":"A Alaa","year":"2019"},{"key":"pdig.0000721.ref050","unstructured":"Curth A, Svensson D, Weatherall J, van der Schaar M. Really doing great at estimating CATE? A critical look at ML benchmarking practices in treatment effect estimation. Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2). 2021."},{"key":"pdig.0000721.ref051","unstructured":"Johnson A, Bulgarelli L, Pollard T, Horng S, Celi L, Mark R. Mimic-iv. PhysioNet. 2020."},{"issue":"15","key":"pdig.0000721.ref052","doi-asserted-by":"crossref","first-page":"1345","DOI":"10.1056\/NEJMp2113319","article-title":"Methods of public health research - strengthening causal inference from observational data","volume":"385","author":"MA Hern\ufffdn","year":"2021","journal-title":"N Engl J Med"},{"key":"pdig.0000721.ref053","doi-asserted-by":"crossref","unstructured":"Richardson WS, Wilson MC, Nishikawa J, Hayward RS. The well-built clinical question: a key to evidence-based decisions. ACP J Club. 1995;123(3):A12-3. doi: 10.7326\/acpjc-1995-123-3-a12 7582737","DOI":"10.7326\/ACPJC-1995-123-3-A12"},{"key":"pdig.0000721.ref054","unstructured":"Riva JJ, Malik KMP, Burnie SJ, Endicott AR, Busse JW. What is your research question? An introduction to the PICOT format for clinicians. J Can Chiropr Assoc. 2012;56(3):167\u201371. 22997465"},{"issue":"8","key":"pdig.0000721.ref055","doi-asserted-by":"crossref","first-page":"758","DOI":"10.1093\/aje\/kwv254","article-title":"Using big data to emulate a target trial when a randomized trial is not available","volume":"183","author":"MA Hern\ufffdn","year":"2016","journal-title":"Am J Epidemiol"},{"key":"pdig.0000721.ref056","unstructured":"Catalogue of Bias Collaboration; 2023. Available from: https:\/\/catalogofbias.org\/biases\/."},{"key":"pdig.0000721.ref057","doi-asserted-by":"crossref","unstructured":"Weiskopf N, Dorr D, Jackson C, Lehmann H, Thompson C. Healthcare utilization is a collider: an introduction to collider bias in EHR data reuse. Journal of the American Medical Informatics Association. 2023ocad013.","DOI":"10.1093\/jamia\/ocad013"},{"key":"pdig.0000721.ref058","unstructured":"The Book of OHDSI: Observational Health Data Sciences and Informatics. OHDSI. 2021."},{"key":"pdig.0000721.ref059","unstructured":"Pearl J, Mackenzie D. The book of why: the new science of cause and effect. 2018."},{"issue":"469","key":"pdig.0000721.ref060","doi-asserted-by":"crossref","first-page":"322","DOI":"10.1198\/016214504000001880","article-title":"Causal inference using potential outcomes","volume":"100","author":"DB Rubin","year":"2005","journal-title":"J Am Stat Assoc"},{"key":"pdig.0000721.ref061","doi-asserted-by":"crossref","unstructured":"Attia J, Holliday E, Oldmeadow C. A proposal for capturing interaction and effect modification using DAGs. 2022.","DOI":"10.1093\/ije\/dyac126"},{"issue":"5","key":"pdig.0000721.ref062","doi-asserted-by":"crossref","first-page":"745","DOI":"10.1097\/EDE.0b013e318225c2be","article-title":"DAGitty: a graphical tool for analyzing causal diagrams","volume":"22","author":"J Textor","year":"2011","journal-title":"Epidemiology"},{"issue":"1","key":"pdig.0000721.ref063","doi-asserted-by":"crossref","first-page":"39","DOI":"10.1111\/rssb.12348","article-title":"Making sense of sensitivity: extending omitted variable bias","volume":"82","author":"C Cinelli","year":"2019","journal-title":"J R Stat Soc B"},{"issue":"3","key":"pdig.0000721.ref064","doi-asserted-by":"crossref","DOI":"10.1214\/23-STS911","article-title":"An Introduction to proximal causal inference","volume":"39","author":"EJ Tchetgen Tchetgen","year":"2024","journal-title":"Statist Sci"},{"key":"pdig.0000721.ref065","unstructured":"Wager S. Stats 361: Causal inference; 2020."},{"issue":"1","key":"pdig.0000721.ref066","doi-asserted-by":"crossref","first-page":"4","DOI":"10.1162\/003465304323023651","article-title":"Nonparametric estimation of average treatment effects under exogeneity: a review","volume":"86","author":"GW Imbens","year":"2004","journal-title":"Rev Econ Stat"},{"key":"pdig.0000721.ref067","unstructured":"Colnet B, Josse J, Varoquaux G, Scornet E. Risk ratio, odds ratio, risk difference... Which causal measure is easier to generalize? arXiv preprint arXiv:230316008. 2023"},{"issue":"2","key":"pdig.0000721.ref068","doi-asserted-by":"crossref","first-page":"644","DOI":"10.1016\/j.jeconom.2019.10.014","article-title":"Overlap in observational studies with high-dimensional covariates","volume":"221","author":"A D\u2019Amour","year":"2021","journal-title":"J Econ"},{"issue":"4","key":"pdig.0000721.ref069","doi-asserted-by":"crossref","first-page":"e0174708","DOI":"10.1371\/journal.pone.0174708","article-title":"Creating an automated trigger for sepsis clinical decision support at emergency department triage using machine learning","volume":"12","author":"S Horng","year":"2017","journal-title":"PLoS One"},{"key":"pdig.0000721.ref070","unstructured":"Jiang L, Liu X, Nejatian N, Nasir-Moin M, Wang D, Abidin A. Health system-scale language models are all-purpose prediction engines.. Nature. 2023:1\u20136."},{"issue":"23","key":"pdig.0000721.ref071","doi-asserted-by":"crossref","first-page":"3309","DOI":"10.1002\/sim.7820","article-title":"Comparing methods for estimation of heterogeneous treatment effects using observational data from health care databases","volume":"37","author":"T Wendling","year":"2018","journal-title":"Stat Med"},{"key":"pdig.0000721.ref072","doi-asserted-by":"crossref","unstructured":"van der Laan MJ, Polley EC, Hubbard AE. Super learner. Stat Appl Genet Mol Biol. 2007;6:Article25. doi: 10.2202\/1544-6115.1309 17910531","DOI":"10.2202\/1544-6115.1309"},{"key":"pdig.0000721.ref073","doi-asserted-by":"crossref","unstructured":"Doutreligne M, Varoquaux G. How to select predictive models for causal inference?. arXiv preprint. 2023.","DOI":"10.2139\/ssrn.4467871"},{"issue":"5","key":"pdig.0000721.ref074","doi-asserted-by":"crossref","first-page":"291","DOI":"10.1002\/pds.1200","article-title":"Sensitivity analysis and external adjustment for unmeasured confounders in epidemiologic database studies of therapeutics","volume":"15","author":"S Schneeweiss","year":"2006","journal-title":"Pharmacoepidemiol Drug Saf"},{"key":"pdig.0000721.ref075","doi-asserted-by":"crossref","unstructured":"Thabane L, Mbuagbaw L, Zhang S, Samaan Z, Marcucci M, Ye C, et al. A tutorial on sensitivity analyses in clinical trials: the what, why, when and how. BMC Med Res Methodol. 2013;1392. doi: 10.1186\/1471-2288-13-92 23855337","DOI":"10.1186\/1471-2288-13-92"},{"key":"pdig.0000721.ref076","unstructured":". Statistical Principles for Clinical Trials: Addendum: Estimands and Sensitivity Analysis in Clinical Trials. FDA. 2021."},{"issue":"9","key":"pdig.0000721.ref077","doi-asserted-by":"crossref","first-page":"1046","DOI":"10.1016\/j.jclinepi.2015.05.029","article-title":"Assessment of vibration of effects due to model specification can demonstrate the instability of observational associations","volume":"68","author":"CJ Patel","year":"2015","journal-title":"J Clin Epidemiol"},{"issue":"6","key":"pdig.0000721.ref078","doi-asserted-by":"crossref","first-page":"1088","DOI":"10.1093\/aje\/kwaa235","article-title":"Assessing Heterogeneity of Treatment Effects in Observational Studies","volume":"190","author":"SE Robertson","year":"2021","journal-title":"Am J Epidemiol"},{"issue":"17","key":"pdig.0000721.ref079","doi-asserted-by":"crossref","first-page":"1809","DOI":"10.1001\/jama.2013.280502","article-title":"Effects of fluid resuscitation with colloids vs crystalloids on mortality in critically ill patients presenting with hypovolemic shock: the CRISTAL randomized trial","volume":"310","author":"D Annane","year":"2013","journal-title":"JAMA"},{"issue":"6","key":"pdig.0000721.ref080","doi-asserted-by":"crossref","first-page":"702","DOI":"10.1186\/s13054-014-0702-y","article-title":"Comparison of the effects of albumin and crystalloid on mortality in adult patients with severe sepsis and septic shock: a meta-analysis of randomized clinical trials","volume":"18","author":"J-Y Xu","year":"2014","journal-title":"Crit Care"},{"issue":"6","key":"pdig.0000721.ref081","doi-asserted-by":"crossref","first-page":"679","DOI":"10.1097\/SHK.0000000000001468","article-title":"Resuscitation fluids in septic shock: a network meta-analysis of randomized controlled trials","volume":"53","author":"B Li","year":"2020","journal-title":"Shock"},{"issue":"15","key":"pdig.0000721.ref082","doi-asserted-by":"crossref","first-page":"1412","DOI":"10.1056\/NEJMoa1305727","article-title":"Albumin replacement in patients with severe sepsis or septic shock","volume":"370","author":"P Caironi","year":"2014","journal-title":"N Engl J Med"},{"issue":"1","key":"pdig.0000721.ref083","doi-asserted-by":"crossref","first-page":"42","DOI":"10.1186\/s13613-021-00830-8","article-title":"Early combination of albumin with crystalloids administration might be beneficial for the survival of septic patients: a retrospective analysis from MIMIC-IV database","volume":"11","author":"S Zhou","year":"2021","journal-title":"Ann Intensive Care"},{"key":"pdig.0000721.ref084","unstructured":"Battocchi K, Dillon E, Hei M, Lewis G, Oka P, Oprescu M. EconML: A python package for ML-based heterogeneous treatment effects estimation. 2019."},{"key":"pdig.0000721.ref085","first-page":"2825","article-title":"Scikit-learn: Machine learning in python","volume":"12","author":"F Pedregosa","year":"2011","journal-title":"J Machine Learn Res."},{"issue":"9","key":"pdig.0000721.ref086","doi-asserted-by":"crossref","first-page":"kwab201","DOI":"10.1093\/aje\/kwab201","article-title":"Challenges in obtaining valid causal effect estimates with machine learning algorithms","volume":"192","author":"AI Naimi","year":"2023","journal-title":"Am J Epidemiol"},{"key":"pdig.0000721.ref087","doi-asserted-by":"crossref","first-page":"495","DOI":"10.1186\/s13063-015-1023-4","article-title":"A literature review on the representativeness of randomized controlled trial samples and implications for the external validity of trial results","volume":"16","author":"T Kennedy-Martin","year":"2015","journal-title":"Trials"},{"key":"pdig.0000721.ref088","doi-asserted-by":"crossref","unstructured":"Kent DM, Steyerberg E, van Klaveren D. Personalized evidence based medicine: predictive approaches to heterogeneous treatment effects. BMJ. 2018;363.","DOI":"10.1136\/bmj.k4245"},{"issue":"5","key":"pdig.0000721.ref089","doi-asserted-by":"crossref","first-page":"612","DOI":"10.1109\/JPROC.2021.3058954","article-title":"Toward Causal representation learning","volume":"109","author":"B Scholkopf","year":"2021","journal-title":"Proc IEEE"},{"issue":"1","key":"pdig.0000721.ref090","doi-asserted-by":"crossref","first-page":"3923","DOI":"10.1038\/s41467-020-17419-7","article-title":"Improving the accuracy of medical diagnosis with causal machine learning","volume":"11","author":"JG Richens","year":"2020","journal-title":"Nat Commun"},{"issue":"2","key":"pdig.0000721.ref091","doi-asserted-by":"crossref","first-page":"348","DOI":"10.1161\/01.CIR.53.2.348","article-title":"Multivariate prediction of coronary heart disease in the Western Collaborative Group Study compared to the findings of the Framingham study","volume":"53","author":"RJ Brand","year":"1976","journal-title":"Circulation"},{"issue":"10","key":"pdig.0000721.ref092","doi-asserted-by":"crossref","first-page":"1671","DOI":"10.1093\/aje\/kwac108","article-title":"The Future of Causal Inference","volume":"191","author":"N Mitra","year":"2022","journal-title":"Am J Epidemiol"},{"issue":"1","key":"pdig.0000721.ref093","doi-asserted-by":"crossref","first-page":"7","DOI":"10.1038\/s41746-023-00753-7","article-title":"Making machine learning matter to clinicians: model actionability in medical decision-making","volume":"6","author":"DE Ehrmann","year":"2023","journal-title":"NPJ Digit Med"}],"container-title":["PLOS Digital Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0000721","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,2,3]],"date-time":"2025-02-03T13:54:04Z","timestamp":1738590844000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0000721"}},"subtitle":[],"editor":[{"given":"Vesela","family":"Kovacheva","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2025,2,3]]},"references-count":93,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2025,2,3]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pdig.0000721","relation":{},"ISSN":["2767-3170"],"issn-type":[{"value":"2767-3170","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,2,3]]}}}