Restricted mean survival time regression model with time-dependent covariates.

Chengfeng Zhang, Baoyi Huang, Hongji Wu, Hao Yuan, Yawen Hou, Zheng Chen
Author Information
  1. Chengfeng Zhang: Department of Biostatistics, School of Public Health (Guangdong Provincial Key Laboratory of Tropical Disease Research), Southern Medical University, Guangzhou, People's Republic of China.
  2. Baoyi Huang: Department of Biostatistics, School of Public Health (Guangdong Provincial Key Laboratory of Tropical Disease Research), Southern Medical University, Guangzhou, People's Republic of China.
  3. Hongji Wu: Department of Biostatistics, School of Public Health (Guangdong Provincial Key Laboratory of Tropical Disease Research), Southern Medical University, Guangzhou, People's Republic of China.
  4. Hao Yuan: Department of Biostatistics, School of Public Health (Guangdong Provincial Key Laboratory of Tropical Disease Research), Southern Medical University, Guangzhou, People's Republic of China.
  5. Yawen Hou: Department of Statistics, School of Economics, Jinan University, Guangzhou, People's Republic of China.
  6. Zheng Chen: Department of Biostatistics, School of Public Health (Guangdong Provincial Key Laboratory of Tropical Disease Research), Southern Medical University, Guangzhou, People's Republic of China. ORCID

Abstract

In clinical or epidemiological follow-up studies, methods based on time scale indicators such as the restricted mean survival time (RMST) have been developed to some extent. Compared with traditional hazard rate indicator system methods, the RMST is easier to interpret and does not require the proportional hazard assumption. To date, regression models based on the RMST are indirect or direct models of the RMST and baseline covariates. However, time-dependent covariates are becoming increasingly common in follow-up studies. Based on the inverse probability of censoring weighting (IPCW) method, we developed a regression model of the RMST and time-dependent covariates. Through Monte Carlo simulation, we verified the estimation performance of the regression parameters of the proposed model. Compared with the time-dependent Cox model and the fixed (baseline) covariate RMST model, the time-dependent RMST model has a better prediction ability. Finally, an example of heart transplantation was used to verify the above conclusions.

Keywords

References

  1. J Natl Cancer Inst. 2018 Mar 1;110(3):232-240 [PMID: 29933439]
  2. Biometrics. 2001 Dec;57(4):1030-8 [PMID: 11764241]
  3. Pharm Stat. 2020 Jul;19(4):436-453 [PMID: 32072769]
  4. Stat Med. 2011 Aug 30;30(19):2409-21 [PMID: 21611958]
  5. Stat Med. 2022 Sep 20;41(21):4081-4090 [PMID: 35746886]
  6. Am J Epidemiol. 2022 Jan 1;191(1):163-172 [PMID: 34550319]
  7. Stat Med. 2006 Dec 30;25(24):4279-92 [PMID: 16947139]
  8. Circ Cardiovasc Qual Outcomes. 2020 Apr;13(4):e005918 [PMID: 32228064]
  9. Biostatistics. 2014 Apr;15(2):222-33 [PMID: 24292992]
  10. Stat Methods Med Res. 2010 Feb;19(1):71-99 [PMID: 19654170]
  11. J Biopharm Stat. 2021 Mar;31(2):141-155 [PMID: 32892707]
  12. Stat Med. 2019 Sep 10;38(20):3832-3860 [PMID: 31119770]
  13. Stat Med. 2019 May 20;38(11):2074-2102 [PMID: 30652356]
  14. Biometrics. 2011 Sep;67(3):740-9 [PMID: 21039400]
  15. Biometrics. 2022 Mar;78(1):192-201 [PMID: 33616953]
  16. Stat Med. 1996 Feb 28;15(4):361-87 [PMID: 8668867]
  17. J Hyg (Lond). 1949 Jun;47(2):188 [PMID: 15406758]
  18. Lifetime Data Anal. 2004 Dec;10(4):335-50 [PMID: 15690989]
  19. J Heart Lung Transplant. 2018 Oct;37(10):1155-1168 [PMID: 30293612]
  20. Stat Med. 2012 Dec 20;31(29):3946-58 [PMID: 22763916]
  21. Lifetime Data Anal. 2018 Jan;24(1):176-199 [PMID: 28224260]
  22. Am J Kidney Dis. 2017 Apr;69(4):492-494 [PMID: 28340850]
  23. Comput Methods Programs Biomed. 2021 Aug;207:106155 [PMID: 34038865]

MeSH Term

Follow-Up Studies
Humans
Probability
Proportional Hazards Models
Survival Analysis
Survival Rate

Word Cloud

Created with Highcharts 10.0.0RMSTtime-dependentmodelcovariatestimesurvivalregressionmeanfollow-upstudiesmethodsbasedrestricteddevelopedComparedhazardmodelsbaselineinverseprobabilitycensoringweightingclinicalepidemiologicalscaleindicatorsextenttraditionalrateindicatorsystemeasierinterpretrequireproportionalassumptiondateindirectdirectHoweverbecomingincreasinglycommonBasedIPCWmethodMonteCarlosimulationverifiedestimationperformanceparametersproposedCoxfixedcovariatebetterpredictionabilityFinallyexamplehearttransplantationusedverifyconclusionsRestrictedanalysis

Similar Articles

Cited By (3)