Competing risk regression models for epidemiologic data - PubMed (original) (raw)
. 2009 Jul 15;170(2):244-56.
doi: 10.1093/aje/kwp107. Epub 2009 Jun 3.
Affiliations
- PMID: 19494242
- PMCID: PMC2732996
- DOI: 10.1093/aje/kwp107
Competing risk regression models for epidemiologic data
Bryan Lau et al. Am J Epidemiol. 2009.
Abstract
Competing events can preclude the event of interest from occurring in epidemiologic data and can be analyzed by using extensions of survival analysis methods. In this paper, the authors outline 3 regression approaches for estimating 2 key quantities in competing risks analysis: the cause-specific relative hazard ((cs)RH) and the subdistribution relative hazard ((sd)RH). They compare and contrast the structure of the risk sets and the interpretation of parameters obtained with these methods. They also demonstrate the use of these methods with data from the Women's Interagency HIV Study established in 1993, treating time to initiation of highly active antiretroviral therapy or to clinical disease progression as competing events. In our example, women with an injection drug use history were less likely than those without a history of injection drug use to initiate therapy prior to progression to acquired immunodeficiency syndrome or death by both measures of association ((cs)RH = 0.67, 95% confidence interval: 0.57, 0.80 and (sd)RH = 0.60, 95% confidence interval: 0.50, 0.71). Moreover, the relative hazards for disease progression prior to treatment were elevated ((cs)RH = 1.71, 95% confidence interval: 1.37, 2.13 and (sd)RH = 2.01, 95% confidence interval: 1.62, 2.51). Methods for competing risks should be used by epidemiologists, with the choice of method guided by the scientific question.
Figures
Figure 1.
Cause-specific hazard schematic. The risk set starts with 30 individuals (solid circles). Over time, individuals have either event 1 (square) or event 2 (triangle). As individuals have either event, they are removed from the remaining risk sets. The calculation for the cause-specific hazard is given at the bottom of the figure.
Figure 2.
Subdistribution hazard schematic. The risk set starts with 30 individuals (solid circles). Over time, individuals have either event 1 (square) or event 2 (triangle). As individuals have the competing event (event 2, triangle), they are maintained in the risk set as triangles. Thus, over time, a greater proportion of the risk set becomes full of triangles that are individuals who have had the competing event prior to that time. The subdistribution hazard (SDH) for event 1 is given near the bottom of the figure along with the cause-specific hazard (CSH) for event 1 for comparison. Note that, because individuals are maintained in the risk set, the SDH tends to be lower than the CSH.
Figure 3.
Cumulative incidence of treatment initiation prior to acquired immunodeficiency syndrome (AIDS) or death (A and B) and the cumulative incidence of AIDS or death prior to treatment (C and D) by injection drug use status and type of cumulative incidence (cause-specific, A and C; subdistribution, B and D; csPH, from proportional cause-specific hazards model; sdPH, from proportional subdistribution hazards model). The mixture model comprised a lognormal distribution for initiation of treatment and a generalized-gamma distribution for the time to AIDS or death prior to treatment initiation. CI, confidence interval; csRH, cause-specific relative hazard; HAART, highly active antiretroviral therapy; IDU, injection drug use; sdRH, subdistribution relative hazard.
Similar articles
- Parametric mixture models to evaluate and summarize hazard ratios in the presence of competing risks with time-dependent hazards and delayed entry.
Lau B, Cole SR, Gange SJ. Lau B, et al. Stat Med. 2011 Mar 15;30(6):654-65. doi: 10.1002/sim.4123. Epub 2010 Nov 30. Stat Med. 2011. PMID: 21337360 Free PMC article. - The importance of censoring in competing risks analysis of the subdistribution hazard.
Donoghoe MW, Gebski V. Donoghoe MW, et al. BMC Med Res Methodol. 2017 Apr 4;17(1):52. doi: 10.1186/s12874-017-0327-3. BMC Med Res Methodol. 2017. PMID: 28376736 Free PMC article. - Semiparametric competing risks regression under interval censoring using the R package intccr.
Park J, Bakoyannis G, Yiannoutsos CT. Park J, et al. Comput Methods Programs Biomed. 2019 May;173:167-176. doi: 10.1016/j.cmpb.2019.03.002. Epub 2019 Mar 8. Comput Methods Programs Biomed. 2019. PMID: 31046992 Free PMC article. - Applying competing risks regression models: an overview.
Haller B, Schmidt G, Ulm K. Haller B, et al. Lifetime Data Anal. 2013 Jan;19(1):33-58. doi: 10.1007/s10985-012-9230-8. Epub 2012 Sep 26. Lifetime Data Anal. 2013. PMID: 23010807 Review. - Practical methods for competing risks data: a review.
Bakoyannis G, Touloumi G. Bakoyannis G, et al. Stat Methods Med Res. 2012 Jun;21(3):257-72. doi: 10.1177/0962280210394479. Epub 2011 Jan 7. Stat Methods Med Res. 2012. PMID: 21216803 Review.
Cited by
- Elevated Serum Magnesium Levels May Delay the Loss of Residual Renal Function among Patients Receiving Peritoneal Dialysis: A Prospective Study.
Zhao J, Lin X, Wang J, Guo X, Peng F, Zuo X, Tian C, Ying C. Zhao J, et al. Biol Trace Elem Res. 2024 Oct 30. doi: 10.1007/s12011-024-04432-w. Online ahead of print. Biol Trace Elem Res. 2024. PMID: 39477852 - Weighing risks and benefits in the presence of competing risks.
Lesko CR, Zalla LC, Heyward J, Joseph C, Edwards JK. Lesko CR, et al. Curr Epidemiol Rep. 2023 Dec;10(4):221-239. doi: 10.1007/s40471-023-00331-1. Epub 2023 Sep 22. Curr Epidemiol Rep. 2023. PMID: 39473700 Free PMC article. - Heart failure and major haemorrhage in people with atrial fibrillation.
Jones NR, Smith M, Lay-Flurrie S, Yang Y, Hobbs FR, Taylor CJ. Jones NR, et al. Open Heart. 2024 Oct 14;11(2):e002975. doi: 10.1136/openhrt-2024-002975. Open Heart. 2024. PMID: 39401957 Free PMC article. - Incidence and predictors of ventilator-associated pneumonia using a competing risk analysis: a single-center prospective cohort study in Egypt.
Elsheikh M, Kuriyama A, Goto Y, Takahashi Y, Toyama M, Nishikawa Y, El Heniedy MA, Abdelraouf YM, Okada H, Nakayama T. Elsheikh M, et al. BMC Infect Dis. 2024 Sep 19;24(1):1007. doi: 10.1186/s12879-024-09909-6. BMC Infect Dis. 2024. PMID: 39300386 Free PMC article. - Secondary solid malignancies in long-term survivors after total body irradiation.
Gruber I, Wolff D, Koelbl O. Gruber I, et al. Radiat Oncol. 2024 Sep 17;19(1):122. doi: 10.1186/s13014-024-02520-8. Radiat Oncol. 2024. PMID: 39289692 Free PMC article.
References
- Crowder MJ. Classical Competing Risks. Boca Raton, FL: Chapman & Hall/CRC; 2001.
- Kalbfleisch JD, Prentice RL. The Statistical Analysis of Failure Time Data. New York, NY: John Wiley & Sons, Inc; 1980.
- Pintilie M. Competing Risks: A Practical Perspective. Chichester, England: John Wiley & Sons, Ltd; 2006.
- Lau B, Gange SJ, Moore RD. Risk of non-AIDS-related mortality may exceed risk of AIDS-related mortality among individuals enrolling into care with CD4+ counts greater than 200 cells/mm3. J Acquir Immune Defic Syndr. 2007;44(2):179–187. - PubMed
Publication types
MeSH terms
Grants and funding
- U01-AI-42590/AI/NIAID NIH HHS/United States
- K01-AI071754/AI/NIAID NIH HHS/United States
- U01 DA036935/DA/NIDA NIH HHS/United States
- U01-AI069918/AI/NIAID NIH HHS/United States
- K01 AI071754/AI/NIAID NIH HHS/United States
- U01 AI042590-12/AI/NIAID NIH HHS/United States
- U01 AI069918/AI/NIAID NIH HHS/United States
- U01 AI042590/AI/NIAID NIH HHS/United States
LinkOut - more resources
Full Text Sources
Medical