Hostname: page-component-7b7567c8c-wxvwp Total loading time: 0 Render date: 2026-11-15T14:32:29.620Z Has data issue: false hasContentIssue false
  • English
  • Français

Reply to Wolkewitz: When to Use Cumulative Risk-Based Versus Rate-Based Approaches in the Analysis of Hospital-Acquired Infection Risk Factors? That Depends on the Question

Published online by Cambridge University Press:  26 July 2016

Kevin A. Brown,
Nick Daneman,
Vanessa Stevens,
Tom H. Greene and
Paul Arora
Kevin A. Brown*
Affiliation:
Public Health Ontario, Toronto, CanadaDivision of Epidemiology, Dalla Lana School of Public Health, University of Toronto, Canada
Nick Daneman
Affiliation:
Sunnybrook Health Sciences Center, University of Toronto, Canada
Vanessa Stevens
Affiliation:
VA Salt Lake City Health Care System, Salt Lake City, Utah, United StatesDepartment of Pharmacotherapy, University of Utah, Salt Lake City, United States
Tom H. Greene
Affiliation:
Division of Epidemiology, University of Utah, Salt Lake City, United States
Paul Arora
Affiliation:
Division of Epidemiology, Dalla Lana School of Public Health, University of Toronto, CanadaCenter for Global Child Health, The Hospital for Sick Children, Toronto, Canada.
*
Address correspondence to Kevin A. Brown, Public Health Ontario, Toronto, Canada (kevin.brown@oahpp.ca).
Rights & Permissions [Opens in a new window]

Abstract

An abstract is not available for this content. As you have access to this content, full HTML content is provided on this page. A PDF of this content is also available in through the ‘Save PDF’ action button.

Information

Type
Letters to the Editor
Information
Infection Control & Hospital Epidemiology , Volume 37 , Issue 9 , September 2016, pp. 1124 - 1125
Copyright
© 2016 by The Society for Healthcare Epidemiology of America. All rights reserved 

To the Editor—We thank Dr. Wolkewitz for his thoughtful comments and clear breakdown of cumulative risk-based and rate-based measures of association in hospital acquired infection (HAI) research. We agree that a thorough understanding of the distinction between rate-based and cumulative risk-based metrics is essential for researchers performing studies of HAI risk factors.

Another way of thinking about this distinction is through the lens of the study question, which is often either etiologic or prognostic in nature.Reference Noordzij, Leffondre, van Stralen, Zoccali, Dekker and Jager 1 The objective of an etiologic research question is to assess the causal association between a risk factor and a given outcome. That is, if a given exposure were introduced experimentally, would a given patient be more or less likely to experience the outcome.Reference Little and Rubin 2 On the other hand, prognostic research aims to predict the probability that a patient subgroup experiences an outcome on or before a given time point in a hospital stay, irrespective of whether a given risk factor caused an increased rate of disease.

In HAI research, patient subgroups with longer hospital stays may be more likely to develop an HAI during a given stay only because of the longer average duration of their stay. Whereas a risk-based approach would capture this as a difference in cumulative risk of HAI, a rate-based approach would find that the rate of HAI is no different. Epidemiologists interested in questions regarding etiology may be more likely to gravitate toward rate-based approaches, which have been subject to criticism by advocates of causal inference methods,Reference Hernán 3 while clinicians interested in prognostic questions may be more likely to veer toward cumulative risk-based approaches. Because of our primary interest in developing strategies for the prevention of HAI, which exclude modifying length of hospital stay, our research has tended to focus on rate-based approaches.Reference Brown, Daneman and Stevens 4 , Reference Brown, Jones and Daneman 5

In certain circumstances, a comparison of each approach may certainly be useful, while in others, it may not be worth the additional analytic burden. What is important is to understand and interpret the insights derived from cumulative risk-based and rate-based approaches correctly; to conflate the 2 approaches is to muddy the epidemiologic waters.

ACKNOWLEDGMENTS

Financial support: No financial support was provided relevant to this article.

Potential conflicts of interest: All authors report no conflicts of interest relevant to this article.

References

REFERENCES

1. Noordzij, M, Leffondre, K, van Stralen, KJ, Zoccali, C, Dekker, FW, Jager, KJ. When do we need competing risks methods for survival analysis in nephrology? Nephrol Dial Transpl 2013;28:2670–2677.Google Scholar
2. Little, RJ, Rubin, DB. Causal effects in clinical and epidemiological studies via potential outcomes: concepts and analytical approaches. Annu Rev Public Health 2000;21:121–145.Google Scholar
3. Hernán, MA. The hazards of hazard ratios. Epidemiology 2010;21:13–15.Google Scholar
4. Brown, KA, Daneman, N, Stevens, VW, et al. Integrating time-varying and ecological exposures into multivariate analyses of hospital-acquired infection risk factors: a review and demonstration. Infect Control Hosp Epidemiol 2016;37:411–419.Google Scholar
5. Brown, KA, Jones, M, Daneman, N, et al. Importation, antibiotics, and Clostridium difficile infection in veteran long-term care: a multilevel case–control study. Ann Intern Med 2016;164:787–794.Google Scholar