Skip to main content

Loss to Follow-Up in Cohort Studies: How Much is Too Much?

Abstract

Loss to follow-up is problematic in most cohort studies and often leads to bias. Although guidelines suggest acceptable follow-up rates, the authors are unaware of studies that test the validity of these recommendations. The objective of this study was to determine whether the recommended follow-up thresholds of 60–80% are associated with biased effects in cohort studies. A simulation study was conducted using 1000 computer replications of a cohort of 500 observations. The logistic regression model included a binary exposure and three confounders. Varied correlation structures of the data represented various levels of confounding. Differing levels of loss to follow-up were generated through three mechanisms: missing completely at random (MCAR), missing at random (MAR) and missing not at random (MNAR). The authors found no important bias with levels of loss that varied from 5 to 60% when loss to follow-up was related to MCAR or MAR mechanisms. However, when observations were lost to follow-up based on a MNAR mechanism, the authors found seriously biased estimates of the odds ratios with low levels of loss to follow-up. Loss to follow-up in cohort studies rarely occurs randomly. Therefore, when planning a cohort study, one should assume that loss to follow-up is MNAR and attempt to achieve the maximum follow-up rate possible.

This is a preview of subscription content, access via your institution.

References

  1. Kleinbaum DG, Morgenstern H, Kupper LL. Selection bias in epidemiologic studies. Am J Epidemiol 1981; 113: 452–463.

    Google Scholar 

  2. Greenland S. Response and follow-up bias in cohort studies. Am J Epidemiol 1977; 106: 184–187.

    Google Scholar 

  3. Schafer JL, Graham JW. Missing data: our view of the state of the art. Psychol Meth 2002; 7: 147–177.

    Google Scholar 

  4. Butler CW, Snyder M, Wood DE, Curtis JR, Albert RK, Benditt JO. Underestimation of mortality following lung volume reduction surgery resulting from incomplete follow-up. Chest 2001; 119: 1056–1060.

    Google Scholar 

  5. Rothman KJ, Greenland S. Modern Epidemiology. Philadelphia, PA: Lippincott-Raven, 1998.

    Google Scholar 

  6. Hollen PJ, Gralla RJ, Cox C, Eberly SW, Kris MG. A dilemma in analysis: issues in the serial measurement of quality of life in patients with advanced lung cancer. Lung Cancer 1997; 18: 119–136.

    Google Scholar 

  7. Deeg DJH. Attrition in longitudinal population studies: does it affect the generalizability of the findings? An introduction to the series. J Clin Epidemiol 2002; 55: 213–215.

    Google Scholar 

  8. Lohr SL. Nonresponse. In: Lohr SL (ed.), Sampling: design and analysis. Pacific Grove: Duxbury Press, 1999: 255–287.

    Google Scholar 

  9. Altman DG. Statistics in medical journals: some recent trends. Stat Med 2000; 19: 3275–3289.

    Google Scholar 

  10. Babbie ER. Survey research methods. Belmont, CA: Wadsworth, 1973.

    Google Scholar 

  11. Little RJA, Rubin DB. Statistical analysis with missing data. New York: John Wiley & Sons, 1987.

    Google Scholar 

  12. Twisk J, de Vente W. Attrition in longitudinal studies: how to deal with missing data. J Clin Epidemiol 2002; 55: 329–337.

    Google Scholar 

  13. Siddiqui O, Flay BR, Hu FB. Factors affecting attrition in a longitudinal smoking prevention study. Prev Med 1996; 25: 554–560.

    Google Scholar 

  14. Collins LM, Schafer JL, Kam C. A comparison of inclusive and restrictive strategies in modern missing data procedures. Psychol Meth 2001; 6: 330–351.

    Google Scholar 

  15. McLean RR, Hannan MT, Epstein BE, Bouxsein ML, Cupples LA, Murabito J, et al. Elderly cohort study subjects unable to return for follow-up have lower bone mass than those who can return. Am J Epidemiol 2000; 151: 689–692.

    Google Scholar 

  16. Bisgard KM, Folsom AR, Hong CP, Sellers TA. Mortality and cancer rates in nonrespondents to a prospective study of older women: 5-year follow-up. Am J Epidemiol 1994; 139: 990–1000.

    Google Scholar 

  17. Diggle PJ. Testing for random dropouts in repeated measurement data. Biometrics 1989; 45: 1255–1258.

    Google Scholar 

  18. Elashoff JD, Elashoff RM. Two-samples problems for a dichotomous variable with missing data. Appl Stats 1974; 23: 26–34.

    Google Scholar 

  19. Maldonado G, Greenland S. The importance of critically interpreting simulation studies. Epidemiology 1997;8: 453–456.

    Google Scholar 

  20. Kristman V, Manno M, Côté P. The potential impact of attrition bias in cohort studies: A simulation study. Working Paper No. 180. Toronto: Institute for Work & Health, 2002.

    Google Scholar 

  21. Crawford SL, Tennstedt SL, McKinlay JB. A comparison of analytic methods for non-random missingness of outcome data. J Clin Epidemiol 1995; 48: 209–219.

    Google Scholar 

  22. Bootsma-van der Wiel A, van Exel E, de Craen AJM, Gussekloo J, Lagaay AM, Knook DL, et al. A high response is not essential to prevent selection bias: results from the Leiden 85-plus study. J Clin Epidemiol 2002; 55: 1119–1125.

    Google Scholar 

  23. Kempen GIJM, van Sonderen E. Psychological attributes and changes in disability among low-functioning older persons: does attrition affect the outcomes? J Clin Epidemiol 2002; 55: 224–229.

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Kristman, V., Manno, M. & Côté, P. Loss to Follow-Up in Cohort Studies: How Much is Too Much?. Eur J Epidemiol 19, 751–760 (2004). https://doi.org/10.1023/B:EJEP.0000036568.02655.f8

Download citation

  • Issue Date:

  • DOI: https://doi.org/10.1023/B:EJEP.0000036568.02655.f8

  • Bias (epidemiology)
  • Cohort studies
  • Computer simulation
  • Epidemiologic methods
  • Follow-up studies
  • Logistic models