Negative controls: a tool for detecting confounding and bias in observational studies

Marc Lipsitch, Eric Tchetgen Tchetgen, Ted Cohen, Marc Lipsitch, Eric Tchetgen Tchetgen, Ted Cohen

Abstract

Noncausal associations between exposures and outcomes are a threat to validity of causal inference in observational studies. Many techniques have been developed for study design and analysis to identify and eliminate such errors. Such problems are not expected to compromise experimental studies, where careful standardization of conditions (for laboratory work) and randomization (for population studies) should, if applied properly, eliminate most such noncausal associations. We argue, however, that a routine precaution taken in the design of biologic laboratory experiments--the use of "negative controls"--is designed to detect both suspected and unsuspected sources of spurious causal inference. In epidemiology, analogous negative controls help to identify and resolve confounding as well as other sources of error, including recall bias or analytic flaws. We distinguish 2 types of negative controls (exposure controls and outcome controls), describe examples of each type from the epidemiologic literature, and identify the conditions for the use of such negative controls to detect confounding. We conclude that negative controls should be more commonly employed in observational studies, and that additional work is needed to specify the conditions under which negative controls will be sensitive detectors of other sources of error in observational studies.

Figures

FIG 1
FIG 1
Causal diagram for the effect of an exposure of interest (A) on an outcome of interest (Y), with confounders L (assumed measured) and U (assumed uncontrolled) that cause both A and Y. The dashed line between L and U indicates that either may cause the other, and they may share common causes.
FIG 2
FIG 2
Causal diagram showing an ideal negative control outcome N for use in evaluating studies of the causal relationship between exposure A and outcome Y. N should ideally have the same incoming arrows as Y, except that A does not cause N; to the extent this criterion is met, N is called U-comparable to Y.
FIG 3
FIG 3
Causal diagram showing an ideal negative control exposure B for use in evaluating studies of the causal relationship between exposure A and outcome Y. B should ideally have the same incoming arrows as A; to the extent this criterion is met, B is called U-comparable to A. Z is an instrumental variable of the A-Y relationship and is depicted to illustrate the difference between an instrumental variable and a negative control variable.

References

    1. Hernan MA, Hernandez-Diaz S, Robins JM. A structural approach to selection bias. Epidemiology. 2004;15(5):615–25.
    1. Lu YJ, Gross J, Bogaert D, Finn A, Bagrade L, Zhang Q, Kolls JK, Srivastava A, Lundgren A, Forte S, Thompson CM, Harney KF, Anderson PW, Lipsitch M, Malley R. Interleukin-17A mediates acquired immunity to pneumococcal colonization. PLoS Pathog. 2008;4(9):e1000159.
    1. Fisher RA. The Design of Experiments. Edinburgh: Oliver and Boyd; 1935.
    1. Greenland S. Introduction to Regression Modeling. In: Rothman KJ, Greenland S, Lash TL, editors. Modern Epidemiology. 3. New York: Lippincott, Williams and Wilkins; 2008. pp. 418–455.
    1. Rivetti D, Jefferson T, Thomas R, Rudin M, Rivetti A, Di Pietrantonj C, Demicheli V. Vaccines for preventing influenza in the elderly. Cochrane Database Syst Rev. 2006;3:CD004876.
    1. Simonsen L, Taylor RJ, Viboud C, Miller MA, Jackson LA. Mortality benefits of influenza vaccination in elderly people: an ongoing controversy. Lancet Infect Dis. 2007;7(10):658–66.
    1. Jackson LA, Jackson ML, Nelson JC, Neuzil KM, Weiss NS. Evidence of bias in estimates of influenza vaccine effectiveness in seniors. Int J Epidemiol. 2006;35(2):337–44.
    1. Simonsen L, Reichert TA, Viboud C, Blackwelder WC, Taylor RJ, Miller MA. Impact of Influenza Vaccination on Seasonal Mortality in the US Elderly Population. Arch Intern Med. 2005;165(3):265–72.
    1. Zaadstra BM, Chorus AM, van Buuren S, Kalsbeek H, van Noort JM. Selective association of multiple sclerosis with infectious mononucleosis. Mult Scler. 2008;14(3):307–13.
    1. Suissa S, Ernst P. Bias in observational study of the effectiveness of nasal corticosteroids in asthma. J Allergy Clin Immunol. 2005;115(4):714–9.
    1. Bell CM, Bajcar J, Bierman AS, Li P, Mamdani MM, Urbach DR. Potentially unintended discontinuation of long-term medication use after elective surgical procedures. Arch Intern Med. 2006;166(22):2525–31.
    1. Harris AD, Samore MH, Lipsitch M, Kaye KS, Perencevich E, Carmeli Y. Control-Group Selection Importance in Studies of Antimicrobial Resistance: Examples Applied to Pseudomonas aeruginosa, Enterococci, and Escherichia coli. Clin Infect Dis. 2002;34(12):1558–63.
    1. Brookhart MA, Wang PS, Solomon DH, Schneeweiss S. Evaluating short-term drug effects using a physician-specific prescribing preference as an instrumental variable. Epidemiology. 2006;17(3):268–75.
    1. Hernan MA, Robins JM. Instruments for causal inference: an epidemiologist’s dream? Epidemiology. 2006;17(4):360–72.
    1. Martens EP, Pestman WR, de Boer A, Belitser SV, Klungel OH. Instrumental variables: application and limitations. Epidemiology. 2006;17(3):260–7.
    1. Glymour MM, Greenland S. Causal diagrams. In: Rothman KJ, Greenland S, Lash TL, editors. Modern Epidemiology. 3. Philadelphia: Lippincott, Williams & Wilkins; 2008.
    1. Greenland S. An introduction to instrumental variables for epidemiologists. Int J Epidemiol. 2000;29(4):722–9.
    1. Cook TD, Campbell DT. Quasi-Experimentation: Design & Analysis Issues for Field Settings. Boston: Houghton-Mifflin; 1979.
    1. Hill AB. The Environment and Disease: Association or Causation? Proc R Soc Med. 1965;58:295–300.
    1. Aschengrau A, Seage GR., III . Essentials of Epidemiology in Public Health. Boston: Jones and Bartlett; 2008.
    1. Rothman KJ, Greenland S, Poole C, Lash TL. Causation and Causal Inference. In: Rothman KJ, Greenland S, Lash TL, editors. Modern Epidemiology. 3. Philadelphia: Lippincott, Williams and Wilkins; 2008. pp. 5–31.
    1. Weiss NS. Can the “specificity” of an association be rehabilitated as a basis for supporting a causal hypothesis? Epidemiology. 2002;13(1):6–8.

Source: PubMed

Sponsors and Collaborators

Medical Conditions

Drug Interventions

3
Subscribe