{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,5]],"date-time":"2026-10-05T02:01:34Z","timestamp":1791165694670,"version":"4.1.0"},"reference-count":55,"publisher":"Wiley","issue":"24","license":[{"start":{"date-parts":[[2019,8,22]],"date-time":"2019-08-22T00:00:00Z","timestamp":1566432000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"funder":[{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"publisher","award":["U01AI069924"],"award-info":[{"award-number":["U01AI069924"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"publisher","award":["U01AI069919"],"award-info":[{"award-number":["U01AI069919"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Statistics in Medicine"],"published-print":{"date-parts":[[2019,10,30]]},"abstract":"<jats:p>Longitudinal targeted maximum likelihood estimation (LTMLE) has very rarely been used to estimate dynamic treatment effects in the context of time\u2010dependent confounding affected by prior treatment when faced with long follow\u2010up times, multiple time\u2010varying confounders, and complex associational relationships simultaneously. Reasons for this include the potential computational burden, technical challenges, restricted modeling options for long follow\u2010up times, and limited practical guidance in the literature. However, LTMLE has desirable asymptotic properties, ie, it is doubly robust, and can yield valid inference when used in conjunction with machine learning. It also has the advantage of easy\u2010to\u2010calculate analytic standard errors in contrast to the g\u2010formula, which requires bootstrapping. We use a topical and sophisticated question from HIV treatment research to show that LTMLE can be used successfully in complex realistic settings, and we compare results to competing estimators. Our example illustrates the following practical challenges common to many epidemiological studies: (1) long follow\u2010up time (30 months); (2) gradually declining sample size; (3) limited support for some intervention rules of interest; (4) a high\u2010dimensional set of potential adjustment variables, increasing both the need and the challenge of integrating appropriate machine learning methods; and (5) consideration of collider bias. Our analyses, as well as simulations, shed new light on the application of LTMLE in complex and realistic settings: We show that (1) LTMLE can yield stable and good estimates, even when confronted with small samples and limited modeling options; (2) machine learning utilized with a small set of simple learners (if more complex ones cannot be fitted) can outperform a single, complex model, which is tailored to incorporate prior clinical knowledge; and (3) performance can vary considerably depending on interventions and their support in the data, and therefore critical quality checks should accompany every LTMLE analysis. We provide guidance for the practical application of LTMLE.<\/jats:p>","DOI":"10.1002\/sim.8340","type":"journal-article","created":{"date-parts":[[2019,8,22]],"date-time":"2019-08-22T13:03:12Z","timestamp":1566478992000},"page":"4888-4911","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":54,"title":["Using longitudinal targeted maximum likelihood estimation in complex settings with dynamic interventions"],"prefix":"10.1002","volume":"38","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8475-0591","authenticated-orcid":false,"given":"M.","family":"Schomaker","sequence":"first","affiliation":[{"name":"Centre for Infectious Disease Epidemiology &amp; Research University of Cape Town Cape Town South Africa"},{"name":"Institute of Public Health, Medical Decision Making and Health Technology Assessment, Department of Public Health, Health Services Research and Health Technology Assessment UMIT \u2010 University for Health Sciences, Medical Informatics and Technology Hall in Tirol Austria"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6683-5164","authenticated-orcid":false,"given":"M.A.","family":"Luque\u2010Fernandez","sequence":"additional","affiliation":[{"name":"Biomedical Research Institute of Granada\u00a0\u2013\u00a0Noncommunicable and Cancer Epidemiology Group, Andalusian School of Public Health University of Granada Granada Spain"},{"name":"Department of Epidemiology and Population Health London School of Hygiene and Tropical Medicine London UK"},{"name":"Department of Epidemiology Harvard T.H. Chan School of Public Health Boston Massachusetts"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"V.","family":"Leroy","sequence":"additional","affiliation":[{"name":"INSERM Toulouse France"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"M.A.","family":"Davies","sequence":"additional","affiliation":[{"name":"Centre for Infectious Disease Epidemiology &amp; Research University of Cape Town Cape Town South Africa"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2019,8,22]]},"reference":[{"key":"e_1_2_10_2_1","doi-asserted-by":"publisher","DOI":"10.2202\/1557-4679.1182"},{"key":"e_1_2_10_3_1","doi-asserted-by":"publisher","DOI":"10.1097\/EDE.0000000000000476"},{"key":"e_1_2_10_4_1","doi-asserted-by":"publisher","DOI":"10.1093\/aje\/kwx317"},{"key":"e_1_2_10_5_1","doi-asserted-by":"publisher","DOI":"10.1002\/sim.7628"},{"key":"e_1_2_10_6_1","doi-asserted-by":"publisher","DOI":"10.1097\/EDE.0000000000000905"},{"key":"e_1_2_10_7_1","doi-asserted-by":"publisher","DOI":"10.1515\/jci-2013-0025"},{"key":"e_1_2_10_8_1","doi-asserted-by":"publisher","DOI":"10.1038\/gene.2016.33"},{"key":"e_1_2_10_9_1","doi-asserted-by":"publisher","DOI":"10.2202\/1557-4679.1376"},{"key":"e_1_2_10_10_1","doi-asserted-by":"publisher","DOI":"10.1093\/aje\/kwx213"},{"key":"e_1_2_10_11_1","doi-asserted-by":"publisher","DOI":"10.18637\/jss.v081.i01"},{"key":"e_1_2_10_12_1","doi-asserted-by":"publisher","DOI":"10.1002\/sim.6099"},{"key":"e_1_2_10_13_1","doi-asserted-by":"publisher","DOI":"10.1515\/jci-2013-0007"},{"key":"e_1_2_10_14_1","doi-asserted-by":"publisher","DOI":"10.1097\/EDE.0000000000000178"},{"key":"e_1_2_10_15_1","doi-asserted-by":"publisher","DOI":"10.1111\/biom.12105"},{"key":"e_1_2_10_16_1","doi-asserted-by":"publisher","DOI":"10.1214\/14-AOAS727"},{"issue":"1","key":"e_1_2_10_17_1","first-page":"165","article-title":"Double robust and efficient estimation of a prognostic model for events in the presence of dependent censoring","volume":"17","author":"Schnitzer ME","year":"2016","journal-title":"Biostatistics"},{"key":"e_1_2_10_18_1","doi-asserted-by":"publisher","DOI":"10.1515\/1557-4679.1334"},{"key":"e_1_2_10_19_1","doi-asserted-by":"publisher","DOI":"10.1515\/em-2016-0004"},{"key":"e_1_2_10_20_1","doi-asserted-by":"publisher","DOI":"10.1515\/ijb-2017-0054"},{"key":"e_1_2_10_21_1","doi-asserted-by":"publisher","DOI":"10.1097\/00001648-200009000-00011"},{"key":"e_1_2_10_22_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1541-0420.2005.00377.x"},{"key":"e_1_2_10_23_1","doi-asserted-by":"publisher","DOI":"10.1016\/0270-0255(86)90088-6"},{"key":"e_1_2_10_24_1","doi-asserted-by":"publisher","DOI":"10.1515\/1557-4679.1370"},{"key":"e_1_2_10_25_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4419-9782-1"},{"key":"e_1_2_10_26_1","doi-asserted-by":"publisher","DOI":"10.1198\/000313002753631330"},{"key":"e_1_2_10_27_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pmed.1001044"},{"key":"e_1_2_10_28_1","doi-asserted-by":"publisher","DOI":"10.1056\/NEJMoa0800971"},{"key":"e_1_2_10_29_1","doi-asserted-by":"publisher","DOI":"10.1002\/sim.5686"},{"key":"e_1_2_10_30_1","doi-asserted-by":"publisher","DOI":"10.1016\/S1473-3099(12)70242-6"},{"key":"e_1_2_10_31_1","first-page":"553","volume-title":"Longitudinal Data Analysis","author":"Robins J","year":"2009"},{"key":"e_1_2_10_32_1","doi-asserted-by":"publisher","DOI":"10.1177\/1536867X1201100401"},{"key":"e_1_2_10_33_1","doi-asserted-by":"publisher","DOI":"10.1007\/s12561-011-9040-7"},{"key":"e_1_2_10_34_1","first-page":"2191","volume-title":"Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors","author":"Robins J","year":"2004"},{"key":"e_1_2_10_35_1","first-page":"453","article-title":"Optimal timing of antiretroviral treatment initiation in HIV\u2010positive children and adolescents: a multiregional analysis from Southern Africa, West Africa and Europe","volume":"46","author":"Schomaker M","year":"2017","journal-title":"Int J Epidemiol"},{"key":"e_1_2_10_36_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-65304-4"},{"key":"e_1_2_10_37_1","doi-asserted-by":"publisher","DOI":"10.1002\/sim.7527"},{"key":"e_1_2_10_38_1","doi-asserted-by":"publisher","DOI":"10.1007\/BF00117832"},{"key":"e_1_2_10_39_1","unstructured":"PolleyE LeDellE KennedyC van derLaanM.Superlearner: super learner prediction. R package version 2.0\u201022.https:\/\/CRAN.R-project.org\/package=SuperLearner.2017."},{"key":"e_1_2_10_40_1","article-title":"Super learner","volume":"6","author":"van\u00a0der Laan M","year":"2008","journal-title":"Stat Appl Genet Mol Biol"},{"key":"e_1_2_10_41_1","doi-asserted-by":"publisher","DOI":"10.1093\/ije\/dyr080"},{"key":"e_1_2_10_42_1","doi-asserted-by":"publisher","DOI":"10.1186\/1471-2458-11-519"},{"key":"e_1_2_10_43_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pmed.1001555"},{"issue":"2","key":"e_1_2_10_44_1","first-page":"237","article-title":"Growth and mortality outcomes for different antiretroviral therapy initiation criteria in children aged 1\u20105 years: a causal modelling analysis","volume":"27","author":"Schomaker M","year":"2016","journal-title":"Epidemiology"},{"key":"e_1_2_10_45_1","doi-asserted-by":"publisher","DOI":"10.1093\/biostatistics\/kxl027"},{"key":"e_1_2_10_46_1","doi-asserted-by":"publisher","DOI":"10.1093\/aje\/kwq418"},{"key":"e_1_2_10_47_1","unstructured":"SchwabJ LendleS PetersenM van derLaanM.ltmle: longitudinal targeted maximum likelihood estimation. R package version 0.9\u20109.https:\/\/CRAN.R-project.org\/package=ltmle.2016."},{"key":"e_1_2_10_48_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4899-3242-6"},{"issue":"1","key":"e_1_2_10_49_1","article-title":"A targeted maximum likelihood estimator of a causal effect on a bounded continuous outcome","volume":"6","author":"Gruber S","year":"2010","journal-title":"Int J Biostat"},{"key":"e_1_2_10_50_1","doi-asserted-by":"publisher","DOI":"10.1214\/07-STS227D"},{"key":"e_1_2_10_51_1","doi-asserted-by":"publisher","DOI":"10.1201\/9781420010404"},{"key":"e_1_2_10_52_1","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.1985.10478157"},{"key":"e_1_2_10_53_1","doi-asserted-by":"publisher","DOI":"10.1002\/sim.5316"},{"key":"e_1_2_10_54_1","doi-asserted-by":"publisher","DOI":"10.1177\/0962280210386207"},{"key":"e_1_2_10_55_1","doi-asserted-by":"crossref","unstructured":"SofryginO van derLaanMJ NeugebauerR.simcausal: simulating longitudinal data with causal inference applications. R package version 0.5.3.https:\/\/CRAN.R-project.org\/package=simcausal.2016.","DOI":"10.32614\/CRAN.package.simcausal"},{"key":"e_1_2_10_56_1","article-title":"When and when not to use optimal model averaging","author":"Schomaker M","year":"2019","journal-title":"Statistical papers"}],"container-title":["Statistics in Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.wiley.com\/onlinelibrary\/tdm\/v1\/articles\/10.1002%2Fsim.8340","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/sim.8340","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/sim.8340","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/sim.8340","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,7,22]],"date-time":"2024-07-22T11:43:03Z","timestamp":1721648583000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/sim.8340"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,8,22]]},"references-count":55,"journal-issue":{"issue":"24","published-print":{"date-parts":[[2019,10,30]]}},"alternative-id":["10.1002\/sim.8340"],"URL":"https:\/\/doi.org\/10.1002\/sim.8340","archive":["Portico"],"relation":{},"ISSN":["0277-6715","1097-0258"],"issn-type":[{"value":"0277-6715","type":"print"},{"value":"1097-0258","type":"electronic"}],"subject":[],"published":{"date-parts":[[2019,8,22]]},"assertion":[{"value":"2018-02-14","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2019-07-19","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2019-08-22","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}