Confounding in Observational Studies Based on Large Health Care Databases: Problems and Potential Solutions - a Primer for the Clinician

Overview

Journal Clin Epidemiol

Publisher Dove Medical Press

Specialty Public Health

Date 2017 Apr 14

PMID 28405173

Citations 62

Authors

Mette Norgaard

Vera Ehrenstein

Jan P Vandenbroucke

Affiliations

Soon will be listed here.

Abstract

Population-based health care databases are a valuable tool for observational studies as they reflect daily medical practice for large and representative populations. A constant challenge in observational designs is, however, to rule out confounding, and the value of these databases for a given study question accordingly depends on completeness and validity of the information on confounding factors. In this article, we describe the types of potential confounding factors typically lacking in large health care databases and suggest strategies for confounding control when data on important confounders are unavailable. Using Danish health care databases as examples, we present the use of proxy measures for important confounders and the use of external adjustment. We also briefly discuss the potential value of active comparators, high-dimensional propensity scores, self-controlled designs, pseudorandomization, and the use of positive or negative controls.

Citing Articles

Multi-Center Outcome Analysis of 16 Face Transplantations - A Retrospective OPTN Study.

Knoedler L, Schaschinger T, Niederegger T, Hundeshagen G, Panayi A, Cetrulo Jr C Transpl Int. 2025; 38:14107.

PMID: 39944217 PMC: 11813647. DOI: 10.3389/ti.2025.14107.

Suhexiang pill for acute ischemic stroke in real-world practice setting (SUNRISE): protocol of a multicenter registry.

Lai X, Xiong X, Jia Q, Liu T, Yang Z, Zhang C BMC Complement Med Ther. 2025; 25(1):30.

PMID: 39875901 PMC: 11773706. DOI: 10.1186/s12906-025-04762-9.

The association of medical cannabis use with quality of life in Illinois' opioid alternative pilot program.

Dubois C, Bobitt J, Ding L, Eurich D, Knapp A, Jordan N J Epidemiol Popul Health. 2025; 73(1):202803.

PMID: 39864303 PMC: 11785479. DOI: 10.1016/j.jeph.2024.202803.

Erythropoietin treatment and osteoporotic fracture risk in hemodialysis patients: A nationwide population-based study.

Lee C, Sung F, Hung P, Muo C, Wu M, Huang T Osteoporos Sarcopenia. 2025; 10(4):157-164.

PMID: 39835325 PMC: 11742347. DOI: 10.1016/j.afos.2024.11.002.

Causal associations of ambient particulate matter 10 and Alzheimer's disease: result from a two-sample multivariable Mendelian randomization study.

Huang Z, He G, Sun S, Feng Y, Huang Y Arch Med Sci. 2024; 20(5):1604-1618.

PMID: 39649256 PMC: 11623180. DOI: 10.5114/aoms/185360.

References

Valley T, Walkey A, Lindenauer P, Wiener R, Cooke C . Association Between Noninvasive Ventilation and Mortality Among Older Patients With Pneumonia. Crit Care Med. 2016; 45(3):e246-e254. PMC: 5315597. DOI: 10.1097/CCM.0000000000002076. View

Carroll R, Metcalfe C, Steeg S, Davies N, Cooper J, Kapur N . Psychosocial Assessment of Self-Harm Patients and Risk of Repeat Presentation: An Instrumental Variable Analysis Using Time of Hospital Presentation. PLoS One. 2016; 11(2):e0149713. PMC: 4769277. DOI: 10.1371/journal.pone.0149713. View

Toh S, Garcia Rodriguez L, Hernan M . Confounding adjustment via a semi-automated high-dimensional propensity score algorithm: an application to electronic medical records. Pharmacoepidemiol Drug Saf. 2011; 20(8):849-57. PMC: 3222935. DOI: 10.1002/pds.2152. View

Buckley D, Fu R, Freeman M, Rogers K, Helfand M . C-reactive protein as a risk factor for coronary heart disease: a systematic review and meta-analyses for the U.S. Preventive Services Task Force. Ann Intern Med. 2009; 151(7):483-95. DOI: 10.7326/0003-4819-151-7-200910060-00009. View

Schmidt M, Schmidt S, Sandegaard J, Ehrenstein V, Pedersen L, Sorensen H . The Danish National Patient Registry: a review of content, data quality, and research potential. Clin Epidemiol. 2015; 7:449-90. PMC: 4655913. DOI: 10.2147/CLEP.S91125. View

Booth C, Tannock I . Randomised controlled trials and population-based observational research: partners in the evolution of medical evidence. Br J Cancer. 2014; 110(3):551-5. PMC: 3915111. DOI: 10.1038/bjc.2013.725. View

Hernan M, Takkouche B, Caamano-Isorna F, Gestal-Otero J . A meta-analysis of coffee drinking, cigarette smoking, and the risk of Parkinson's disease. Ann Neurol. 2002; 52(3):276-84. DOI: 10.1002/ana.10277. View

Boef A, Souverein P, Vandenbroucke J, van Hylckama Vlieg A, de Boer A, le Cessie S . Instrumental variable analysis as a complementary analysis in studies of adverse effects: venous thromboembolism and second-generation versus third-generation oral contraceptives. Pharmacoepidemiol Drug Saf. 2016; 25(3):317-24. DOI: 10.1002/pds.3956. View

Schneeweiss S, Patrick A, Sturmer T, Brookhart M, Avorn J, MacLure M . Increasing levels of restriction in pharmacoepidemiologic database studies of elderly and comparison with randomized trial results. Med Care. 2007; 45(10 Supl 2):S131-42. PMC: 2905666. DOI: 10.1097/MLR.0b013e318070c08e. View

10.

Isong I, Richmond T, Kawachi I, Avendano M . Childcare Attendance and Obesity Risk. Pediatrics. 2016; 138(5). PMC: 5079080. DOI: 10.1542/peds.2016-1539. View

11.

VanderWeele T, Hernan M, Robins J . Causal directed acyclic graphs and the direction of unmeasured confounding bias. Epidemiology. 2008; 19(5):720-8. PMC: 4242711. DOI: 10.1097/EDE.0b013e3181810e29. View

12.

Britton A, McKee M, Black N, McPherson K, Sanderson C, Bain C . Threats to applicability of randomised trials: exclusions and selective participation. J Health Serv Res Policy. 1999; 4(2):112-21. DOI: 10.1177/135581969900400210. View

13.

Garbe E, Kloss S, Suling M, Pigeot I, Schneeweiss S . High-dimensional versus conventional propensity scores in a comparative effectiveness study of coxibs and reduced upper gastrointestinal complications. Eur J Clin Pharmacol. 2012; 69(3):549-57. DOI: 10.1007/s00228-012-1334-2. View

14.

Brenner H, Rothenbacher D, Bode G, Adler G . Relation of smoking and alcohol and coffee consumption to active Helicobacter pylori infection: cross sectional study. BMJ. 1998; 315(7121):1489-92. PMC: 2127930. DOI: 10.1136/bmj.315.7121.1489. View

15.

Fleischer N, Diez Roux A . Using directed acyclic graphs to guide analyses of neighbourhood health effects: an introduction. J Epidemiol Community Health. 2008; 62(9):842-6. DOI: 10.1136/jech.2007.067371. View

16.

Jackson L, Jackson M, Nelson J, Neuzil K, Weiss N . Evidence of bias in estimates of influenza vaccine effectiveness in seniors. Int J Epidemiol. 2005; 35(2):337-44. DOI: 10.1093/ije/dyi274. View

17.

Powers K, Kay D, Factor S, Zabetian C, Higgins D, Samii A . Combined effects of smoking, coffee, and NSAIDs on Parkinson's disease risk. Mov Disord. 2007; 23(1):88-95. DOI: 10.1002/mds.21782. View

18.

Lund J, Ostgard L, Prandoni P, Sorensen H, de Nully Brown P . Incidence, determinants and the transient impact of cancer treatments on venous thromboembolism risk among lymphoma patients in Denmark. Thromb Res. 2015; 136(5):917-23. DOI: 10.1016/j.thromres.2015.09.001. View

19.

Maclure M . The case-crossover design: a method for studying transient effects on the risk of acute events. Am J Epidemiol. 1991; 133(2):144-53. DOI: 10.1093/oxfordjournals.aje.a115853. View

20.

Gaist D, Andersen L, Hallas J, Sorensen H, Schroder H, Friis S . Use of statins and risk of glioma: a nationwide case-control study in Denmark. Br J Cancer. 2013; 108(3):715-20. PMC: 3593536. DOI: 10.1038/bjc.2012.536. View