Das Gesundheitswesen , Thieme Verlag Heft S 2-2021, Jahrgang 83) ISSN 1439-4421 Seite(n) S69 bis S76 DOI: 10.1055/a-1633-3827 CareLit-Dokument-Nr: 318600 |
|
Zusammenfassung In Studien mit Sekundärdaten wie Abrechnungsdaten von Krankenkassen wird man häufig vor methodische Herausforderungen gestellt, die v. a. durch die Zeitabhängigkeit, aber auch durch ungemessenes Confounding entstehen. In diesem Paper stellen wir Strategien vor, um verschiedene Biasquellen zu vermeiden und um den durch ungemessenes Confounding entstehenden Bias abzuschätzen. Wir illustrieren das Prinzip der Targets Trials, marginale Strukturmodelle und instrumentelle Variablen anhand von Studien mit der GePaRD Datenbank. Abschließend werden die Chancen und Limitationen von Record Linkage diskutiert, um fehlende Information in den Daten zu ergänzen. Abstract Studies using secondary data such as health care claims data are often faced with methodological challenges due to the time-dependence of key quantities or unmeasured confounding. In the present paper, we discuss approaches to avoid or suitably address various sources of potential bias. In particular, we illustrate the target trial principle, marginal structural models, and instrumental variables with examples from the “GePaRD” database. Finally, we discuss the strengths and limitations of record linkage which can sometimes be used to supply missing information. Schlüsselwörter Confounding - Datenlinkage - instrumentelle Variablen - marginale Strukturmodelle - Target Trials - Verzerrung Key words bias - confounding - instrumental variables - marginal structure models - record linkage - target trial 25 October 2021 © 2021. Thieme. All rights reserved. Georg Thieme Verlag Rüdigerstraße 14, 70469 Stuttgart, Germany Literatur 1 Wilkinson MD, Dumontier M, Aalbersberg IJ. et al. The FAIR Guiding Principles for scientific data management and stewardship. Sci Data 2016; 3: 160018 CrossrefPubMedGoogle Scholar 2 Nationale Forschungsdateninfrastruktur (NFDI) e. V. Nationale Forschungsdateninfrastruktur (2021). Im Internet https://www.nfdi.de PubMed 3 EOSC Association. European Open Science Cloud (2021). Im Internet https://eosc.eu/ PubMed 4 Jacobs S, Stallmann C, Pigeot I. Verknüpfung großer Sekundär- und Registerdatenquellen mit Daten aus Kohortenstudien. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 2015; 58: 822-828 CrossrefPubMedGoogle Scholar 5 Pigeot I, Ahrens W. Establishment of a pharmacoepidemiological database in Germany: methodological potential, scientific value and practical limitations. Pharmacoepidemiol Drug Saf 2008; 17: 215-223 CrossrefPubMedGoogle Scholar 6 March S, Andrich S, Drepper J. et al. Gute Praxis Datenlinkage (GPD). Gesundheitswesen 2019; 81: 636-650 Article in Thieme ConnectPubMedGoogle Scholar 7 Hernán MA, Sauer BC, Hernández-Díaz S. et al. Specifying a target trial prevents immortal time bias and other self-inflicted injuries in observational analyses. J Clin Epidemiol 2016; 79: 70-75 CrossrefPubMedGoogle Scholar 8 Suissa S. Immortal time bias in pharmacoepidemiology. Am J Epidemiol 2008; 167: 492-499 CrossrefPubMedGoogle Scholar 9 Haque R, Shi J, Schottinger JE. et al. Tamoxifen and antidepressant drug interaction in a cohort of 16,887 breast cancer survivors. J Natl Cancer Inst 2016; 108 PubMedGoogle Scholar 10 Institut für Epidemiologie und Sozialmedizin der Universität Münster. ZEBra-MSP Evaluation der Brustkrebsmortalität im deutschen Mammographie-Screening-Programm (2021). Im Internet: https://www.medizin.uni-muenster.de/epi/forschung/projekte/zebra-msp.html PubMed 11 Langner I, Riedel O, Czwikla J. et al. Linkage of routine data to other data sources in Germany: a practical example illustrating challenges and solutions. Gesundheitswesen 2020; 82: S117-S121 Article in Thieme ConnectPubMedGoogle Scholar 12 Ray WA. Evaluating medication effects outside of clinical trials: new-user designs. Am J Epidemiol 2003; 158: 915-920 CrossrefPubMedGoogle Scholar 13 García-Albéniz X, Hsu J, Hernán MA. The value of explicitly emulating a target trial when using real world evidence: an application to colorectal cancer screening. Eur J Epidemiol 2017; 32: 495-500 CrossrefPubMedGoogle Scholar 14 Didelez V. Commentary: Should the analysis of observational data always be preceded by specifying a target experimental trial?. Int J Epidemiol 2016; 45: 2049-2051 PubMedGoogle Scholar 15 Hernán MA, Robins JM. Using big data to emulate a target trial when a randomized trial is not available. Am J Epidemiol 2016; 183: 758-764 CrossrefPubMedGoogle Scholar 16 Dickerman BA, García-Albéniz X, Logan RW. et al. Emulating a target trial in case-control designs: an application to statins and colorectal cancer. Int J Epidemiol 2020; 49: 1637-1646 CrossrefPubMedGoogle Scholar 17 Hernán MA, Robins JM. Causal inference: what if. Boca Raton: Chapman & Hall/CRC; 2020 Google Scholar 18 Witte J, Didelez V. Covariate selection strategies for causal inference: classification and comparison. Biom J 2019; 61: 1270-1289 CrossrefPubMedGoogle Scholar 19 Jackson JW, Schmid I, Stuart EA. Propensity scores in pharmacoepidemiology: beyond the horizon. Curr Epidemiol Rep 2017; 4: 271-280 CrossrefPubMedGoogle Scholar 20 Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med 2009; 28: 3083-3107 CrossrefPubMedGoogle Scholar 21 Enders D, Ohlmeier C, Garbe E. The potential of high-dimensional propensity scores in health services research: an exemplary study on the quality of care for elective percutaneous coronary interventions. Health Serv Res 2018; 53: 197-213 CrossrefPubMedGoogle Scholar 22 Daniel RM, Cousens SN, De Stavola BL. et al. Methods for dealing with time-dependent confounding. Stat Med 2013; 32: 1584-1618 CrossrefPubMedGoogle Scholar 23 Enders D, Engel S, Linder R. et al. Robust versus consistent variance estimators in marginal structural Cox models. Stat Med 2018; 37: 3455-3470 CrossrefPubMedGoogle Scholar 24 Robins JM, Hernán MA, Brumback B. Marginal structural models and causal inference in epidemiology. Epidemiology 2000; 11: 550-560 CrossrefPubMedGoogle Scholar 25 Young JG, Cain LE, Robins JM. et al. Comparative effectiveness of dynamic treatment regimes: an application of the parametric g-formula. Stat Biosci 2011; 3: 119-143 CrossrefPubMedGoogle Scholar 26 Li X, Young JG, Toh S. Estimating effects of dynamic treatment strategies in pharmacoepidemiologic studies with time-varying confounding: a primer. Curr Epidemiol Rep 2017; 4: 288-297 CrossrefPubMedGoogle Scholar 27 Gran JM, Røysland K, Wolbers M. et al. A sequential Cox approach for estimating the causal effect of treatment in the presence of time-dependent confounding applied to data from the Swiss HIV Cohort Study. Stat Med 2010; 29: 2757-2768 CrossrefPubMedGoogle Scholar 28 Petersen M, Schwab J, Gruber S. et al. Targeted maximum likelihood estimation for dynamic and static longitudinal marginal structural working models. J Causal Inference 2014; 2: 147-185 CrossrefPubMedGoogle Scholar 29 Greenland S. An introduction to instrumental variables for epidemiologists. Int J Epidemiol 2018; 47: 358 CrossrefPubMedGoogle Scholar 30 Didelez V, Sheehan N. Mendelian randomization as an instrumental variable approach to causal inference. Stat Methods Med Res 2007; 16: 309-330 CrossrefPubMedGoogle Scholar 31 Martens EP, Pestman WR, de Boer A. et al. Instrumental variables: application and limitations. Epidemiology 2006; 17: 260-267 CrossrefPubMedGoogle Scholar 32 Kollhorst B, Abrahamowicz M, Pigeot I. The proportion of all previous patients was a potential instrument for patientsʼ actual prescriptions of nonsteroidal anti-inflammatory drugs. J Clin Epidemiol 2016; 69: 96-106 CrossrefPubMedGoogle Scholar 33 Schneeweiss S. Sensitivity analysis and external adjustment for unmeasured confounders in epidemiologic database studies of therapeutics. Pharmacoepidemiol Drug Saf 2006; 15: 291-303 CrossrefPubMedGoogle Scholar 34 VanderWeele TJ, Ding P. Sensitivity analysis in observational research: introducing the E-value. Ann Intern Med 2017; 167: 268-274 CrossrefPubMedGoogle Scholar 35 Greenland S. Multiple-bias modelling for analysis of observational data. J R Stat Soc Ser A Stat Soc 2005; 168: 267-306 CrossrefPubMedGoogle Scholar 36 Dorie V, Harada M, Carnegie NB. et al. A flexible, interpretable framework for assessing sensitivity to unmeasured confounding. Stat Med 2016; 35: 3453-3470 CrossrefPubMedGoogle Scholar 37 Lash TL, Fox MP, Fink AK. Applying quantitative bias analysis to epidemiologic data. 1. Aufl. New York: Springer; 2009. DOI: 10.1007/978-0-387-87959-8 CrossrefGoogle Scholar 38 Ahrens W, Greiser KH, Linseisen J. et al. Erforschung von Erkrankungen in der NAKO Gesundheitsstudie. Die wichtigsten gesundheitlichen Endpunkte und ihre Erfassung. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 2020; 63: 376-384 CrossrefPubMedGoogle Scholar 39 Ohlmeier C, Hoffmann F, Giersiepen K. et al. Verknüpfung von Routinedaten der Gesetzlichen Krankenversicherung mit Daten eines Krankenhausinformationssystems: machbar, aber auch „nützlich“?. Gesundheitswesen 2015; 77: e8-e14 Article in Thieme ConnectPubMedGoogle Scholar 40 Cain KC, Breslow NE. Logistic regression analysis and efficient design for two-stage studies. Am J Epidemiol 1988; 128: 1198-1206 CrossrefPubMedGoogle Scholar 41 Behr S, Schill W, Pigeot I. Does additional confounder information alter the estimated risk of bleeding associated with phenprocoumon use – results of a two-phase study. Pharmacoepidemiol Drug Saf 2012; 21: 535-545 CrossrefPubMedGoogle Scholar
{{detailinfo.data.api.data.document[0].apa}}
{{detailinfo.data.api.data.document[0].vancouver}}
{{detailinfo.data.api.data.document[0].harvard}}