Advertisement

Immortal Time Bias in Epidemiology

  • R. W. PlattEmail author
  • J. A. Hutcheon
  • S. Suissa
Epidemiologic Methods (R Maclehose, Section Editor)
Part of the following topical collections:
  1. Topical Collection on Epidemiologic Methods

Abstract

Purpose of Review

Immortal time occurs when study subjects’ person-time is misclassified. For example, if exposure is assigned over time, but treated as a binary “ever-exposed” variable, subjects in the exposed group are “immortal” prior to their exposure. We describe immortal time and the context in which it introduces bias and describe several approaches to avoid immortal time bias via design or mitigate it through analysis.

Recent Findings

Several authors have described examples of immortal time bias in clinical epidemiology, pharmacoepidemiology, and perinatal epidemiology. Solutions to immortal time bias include analyses that appropriately account for time-varying exposure, and design solutions that align exposure with the start of follow-up.

Summary

Immortal time bias is pervasive in epidemiology. It can cause substantial bias. It is, however, easily avoided and can be controlled using appropriate analytic and design strategies.

Keywords

Immortal time Time-varying exposure Target trial 

Notes

Compliance with Ethical Standards

Conflict of Interest

The authors declare that they have no conflict of interest.

Human and Animal Rights and Informed Consent

This article does not contain any studies with human or animal subjects performed by any of the authors.

References

Papers of particular interest, published recently, have been highlighted as: • Of importance

  1. 1.
    Hanley JA, Foster BJ. Avoiding blunders involving ‘immortal time’. Int J Epidemiol. 2014;43(3):949–61.CrossRefGoogle Scholar
  2. 2.
    Farr W. Vital statistics. London: The Sanitary Institute; 1885.Google Scholar
  3. 3.
    Gail MH. Does cardiac transplantation prolong life? A reassessment. Ann Intern Med. 1972;76(5):815–7.CrossRefGoogle Scholar
  4. 4.
    Suissa S. Immortal time bias in observational studies of drug effects. Pharmacoepidemiol Drug Saf. 2007;16(3):241–9.CrossRefGoogle Scholar
  5. 5.
    • Suissa S. Immortal time bias in pharmacoepidemiology. Am J Epidemiol. 2008;167(4):492–9 This gives a comprehensive description of immortal time bias, and of the potential effect it can have in a variety of study settings.CrossRefGoogle Scholar
  6. 6.
    Weberpals J, Jansen L, Herk-Sukel MPP, Kuiper JG, Aarts MJ, Vissers PAJ, et al. Immortal time bias in pharmacoepidemiological studies on cancer patient survival: empirical illustration for beta-blocker use in four cancers with different prognosis. Eur J Epidemiol. 2017;3(3):1–13.Google Scholar
  7. 7.
    Hernán MA, Alonso A, Logan R, Grodstein F, Michels KB, Willett WC, et al. Observational studies analyzed like randomized experiments: an application to postmenopausal hormone therapy and coronary heart disease. Epidemiology. 2008;19(6):766–79.CrossRefGoogle Scholar
  8. 8.
    Matok I, Azoulay L, Yin H, Suissa S. Immortal time bias in observational studies of drug effects in pregnancy. Birth Defects Research (Part A). 2014;100(9):658–62.CrossRefGoogle Scholar
  9. 9.
    Vazquez-Benitez G, Kharbanda EO, Naleway AL, Lipkind H, Sukumaran L, McCarthy NL, et al. Risk of preterm or small-for-gestational-age birth after influenza vaccination during pregnancy: caveats when conducting retrospective observational studies. Am J Epidemiol. 2016:kww043–11.Google Scholar
  10. 10.
    Daniel S, Koren G, Lunenfeld E, Levy A. Immortal time bias in drug safety cohort studies: spontaneous abortion following nonsteroidal antiinflammatory drug exposure. Am J Obstet Gynecol. 2015;212(3):307.e1–6.CrossRefGoogle Scholar
  11. 11.
    • Hutcheon JA, Kuret V, Joseph KS, Sabr Y, Lim K. Immortal time bias in the study of stillbirth risk factors. Epidemiology. 2013;24(6):787–90 This paper shows the occurrence of immortal time bias in pregnancy, and in other cases where occurrence of the outcome affects time at risk.CrossRefGoogle Scholar
  12. 12.
    Mumford SL, Schisterman EF, Cole SR, Westreich DJ, Platt RW. Time at risk and intention-to-treat analyses. Epidemiology. 2015;26(1):112–8.CrossRefGoogle Scholar
  13. 13.
    Suissa S. The quasi-cohort approach in pharmacoepidemiology. Epidemiology. 2015;26(2):242–6.CrossRefGoogle Scholar
  14. 14.
    Mi X, Hammill BG, Curtis LH, Lai EC-C, Setoguchi S. Use of the landmark method to address immortal person-time bias in comparative effectiveness research: a simulation study. Stat Med. 2016:1–13.Google Scholar
  15. 15.
    Hernán MA, Brumback BA, Robins JM. Marginal structural models to estimate the causal effect of zidovudine on the survival of HIV-positive men. Epidemiology. 2000;11(5):561–70.CrossRefGoogle Scholar
  16. 16.
    • Hernán MA, Sauer BC, Hernández-Diaz S, Platt RW, Shrier I. Specifying a target trial prevents immortal time bias and other self-inflicted injuries in observational analyses. J Clin Epidemiol. 2016;79(C):70–5 This paper describes how the use of a target randomized trial as a model for an observational design can help understand the causes of immortal time bias and prevent it.CrossRefGoogle Scholar
  17. 17.
    Lund JL, Richardson DB, Sturmer T. The active comparator, new user study design in pharmacoepidemiology: historical foundations and contemporary application. Curr Epidemiol Rep. 2015;2:221–8.CrossRefGoogle Scholar
  18. 18.
    Brookhart MA. Counterpoint: the treatment decision design. Am J Epidemiol. 2015:kwv214–6.Google Scholar
  19. 19.
    Suissa S, Moodie EEM, Dell’Aniello S. Prevalent new-user cohort designs for comparative drug effect studies by time-conditional propensity scores. Pharmacoepidemiol Drug Saf. 2016;26(4):459–68.CrossRefGoogle Scholar
  20. 20.
    Suissa S. Metformin to treat cancer. Epidemiology. 2017;28(3):455–8.CrossRefGoogle Scholar
  21. 21.
    von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Epidemiology. 2007;18:800–4.CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Department of Epidemiology, Biostatistics, and Occupational HealthMcGill UniversityMontrealCanada
  2. 2.Lady Davis Research Institute of the Jewish General HospitalMontrealCanada
  3. 3.Research Institute of the McGill University Health CentreMontrealCanada
  4. 4.Department of Obstetrics and GynaecologyUniversity of British ColumbiaVancouverCanada

Personalised recommendations