Abstract
Proportional hazard Cox regression models are frequently used to analyze the impact of different factors on time-to-event outcomes. Most practitioners are familiar with and interpret research results in terms of hazard ratios. Direct differences in survival curves are, however, easier to understand for the general population of users and to visualize graphically. Analyzing the difference among the survival curves for the population at risk allows easy interpretation of the impact of a therapy over the follow-up. When the available information is obtained from observational studies, the observed results are potentially subject to a plethora of measured and unmeasured confounders. Although there are procedures to adjust survival curves for measured covariates, the case of unmeasured confounders has not yet been considered in the literature. In this article we provide a semi-parametric procedure for adjusting survival curves for measured and unmeasured confounders. The method augments our novel instrumental variable estimation method for survival time data in the presence of unmeasured confounding with a procedure for mapping estimates onto the survival probability and the expected survival time scales.
Funding source: Patient-Centered Outcomes Research Institute
Award Identifier / Grant number: ME-1503-28261
Acknowledgments
All statements in this paper, including its findings and conclusions, are solely those of the authors and do not necessarily represent the views of the Patient-Centered Outcomes Research Institute (PCORI), its Board of Governors or Methodology Committee. The authors are sincerely grateful to our PCORI Patient Engagement and Governance Committee for reading a draft of the manuscript and for their efforts supporting the development of the research proposal and the research itself.
-
Author contribution: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.
-
Research funding: This research was supported by Patient-Centered Outcomes Research Institute (ME-1503-28261).
-
Conflict of Interest: The authors have no conflicts of interest to report.
References
1. Fried, T. Shared decision making–finding the sweet spot. N Engl J Med 2016;374:104–6. https://doi.org/10.1056/nejmp1510020.Search in Google Scholar PubMed
2. Cox, D. Regression models and life-tables. J R Stat Soc Series B Stat Methodol 1972;34:187–220. https://doi.org/10.1111/j.2517-6161.1972.tb00899.x.Search in Google Scholar
3. Royston, P, Parmar, M. Restricted mean survival time: an alternative to the hazard ratio for the design and analysis of randomized trials with a time-to-event outcome. BMC Med Res Methodol 2013;13:152. https://doi.org/10.1186/1471-2288-13-152.Search in Google Scholar PubMed PubMed Central
4. Martinussen, T, Vansteelandt, S, Andersen, P. Subtleties in the interpretation of hazard contrasts. Lifetime Data Anal 2020;75:1–23. https://doi.org/10.1007/s10985-020-09501-5.Search in Google Scholar PubMed
5. Rosenbaum, P, Rubin, D. The central role of the propensity score in observational studies for causal effects. Biometrika 1983;70:41–55. https://doi.org/10.1093/biomet/70.1.41.Search in Google Scholar
6. Díaz, I. Statistical inference for data-adaptive doubly robust estimators with survival outcomes. Stat Med 2019;38:2735–48. https://doi.org/10.1002/sim.8156.Search in Google Scholar PubMed
7. Dukes, O, Martinussen, T, Tchetgen Tchetgen, EJ, Vansteelandt, S. On doubly robust estimation of the hazard difference. Biometrics 2019;75:100–9. https://doi.org/10.1111/biom.12943.Search in Google Scholar PubMed PubMed Central
8. Angrist, J, Imbens, G, Rubin, D. Identification of causal effects using instrumental variables. J Am Stat Assoc 1996;91:444–55. https://doi.org/10.1080/01621459.1996.10476902.Search in Google Scholar
9. Tchetgen Tchetgen, EJ, Walter, S, Vansteelandt, S, Martinussen, T, Glymour, M. Instrumental variable estimation in a survival context. Epidemiology 2015;26:402–10. https://doi.org/10.1097/ede.0000000000000262.Search in Google Scholar
10. MacKenzie, T, Tosteson, T, Morden, N, Stukel, T, O’Malley, A. Using instrumental variables to estimate a Cox’s proportional hazards regression subject to additive confounding. Health Serv Outcome Res Methodol 2014;14:54–68. https://doi.org/10.1007/s10742-014-0117-x.Search in Google Scholar PubMed PubMed Central
11. Li, J, Fine, J, Brookhart, A. Instrumental variable additive hazards models. Biometrics 2015;71:122–30. https://doi.org/10.1111/biom.12244.Search in Google Scholar PubMed
12. Brueckner, M, Titman, A, Jaki, T. Instrumental variable estimation in semi-parametric additive hazards models. Biometrics 2019;75:110–20. https://doi.org/10.1111/biom.12952.Search in Google Scholar PubMed PubMed Central
13. Wang, L, Tchetgen Tchetgen, EJ. Bounded, efficient and multiply robust estimation of average treatment effects using instrumental variables. J R Stat Soc Series B Stat Methodol 2018;80:531–50. https://doi.org/10.1111/rssb.12262.Search in Google Scholar PubMed PubMed Central
14. Wang, L, Tchetgen Tchetgen, EJ, Martinussen, T, Vansteelandt, S. Learning causal hazard ratio with endogeneity; 2018. arXiv e-prints 2018: arXiv:1807.05313.Search in Google Scholar
15. Lee, Y, Kennedy, E, Mitra, N. Doubly robust nonparametric instrumental variable estimators for survival outcomes; 2020. arXiv e-prints 2020: arXiv:2007.12973.10.1093/biostatistics/kxab036Search in Google Scholar PubMed
16. Anderson, T. Origins of the limited information maximum likelihood and two-stage least squares estimators. J Econom 2005;127:1–16. https://doi.org/10.1016/j.jeconom.2004.09.012.Search in Google Scholar
17. Greene, W, Zhang, G. Econometric analysis. New Jersey, USA: Prentice Hall; 2003.Search in Google Scholar
18. Martens, E, Pestman, W, de Boer, A, Belitser, S, Klungel, O. Instrumental variables: application and limitations. Epidemiology 2006;17:261–7. https://doi.org/10.1097/01.ede.0000215160.88317.cb.Search in Google Scholar PubMed
19. Martínez-Camblor, P, Mackenzie, T, Staiger, D, Goodney, P, O’Malley, A. Adjusting for bias introduced by instrumental variable estimation in the Cox proportional hazards model. Biostatistics 2019;20:80–96. https://doi.org/10.1093/biostatistics/kxx062.Search in Google Scholar PubMed
20. Martínez-Camblor, P, MacKenzie, T, Staiger, D, Goodney, P, O’Malley, A. An instrumental variable procedure for estimating Cox models with non-proportional hazards in the presence of unmeasured confounding. J R Stat Soc Ser C Appl Stat 2019;68:985–1005. https://doi.org/10.1111/rssc.12341.Search in Google Scholar
21. Irwin, J. The standard error of an estimate of expectation of life, with special reference to expectation of tumourless life in experiments with mice. J Hyg 1949;47:188–9. https://doi.org/10.1017/s0022172400014443.Search in Google Scholar PubMed PubMed Central
22. Monnickendam, G, Zhu, M, McKendrick, J, Su, Y. Measuring survival benefit in health technology assessment in the presence of nonproportional hazards. Value Health 2019;22:431–8. https://doi.org/10.1016/j.jval.2019.01.005.Search in Google Scholar PubMed
23. Wey, A, Vock, D, Connett, J, Rudser, K. Estimating restricted mean treatment effects with stacked survival models. Stat Med 2016;35:3319–32. https://doi.org/10.1002/sim.6929.Search in Google Scholar PubMed PubMed Central
24. Zhao, L, Claggett, B, Tian, L, Uno, H, Pfeffer, MA, Solomon, SD, et al. On the restricted mean survival time curve in survival analysis. Biometrics 2016;72:215–21. https://doi.org/10.1111/biom.12384.Search in Google Scholar PubMed PubMed Central
25. Efron, B, Tibshirani, R. An introduction to the bootstrap. No. 57 in monographs on statistics and applied probability. Boca Raton, Florida, USA: Chapman & Hall/CRC; 1993.Search in Google Scholar
26. R Core Team. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2017.Search in Google Scholar
27. Hernán, M. The hazards of hazard ratios. Epidemiology 2010;21:13–5. https://doi.org/10.1097/ede.0b013e3181c1ea43.Search in Google Scholar PubMed PubMed Central
28. Xu, R, O’Quigley, J. Estimating average regression effect under non-proportional hazards. Biostatistics 2000;1:423–39. https://doi.org/10.1093/biostatistics/1.4.423.Search in Google Scholar PubMed
29. Nieto, F, Coresh, J. Adjusting survival curves for confounders: a review and a new method. Am J Epidemiol 1996;143:1059–68. https://doi.org/10.1093/oxfordjournals.aje.a008670.Search in Google Scholar PubMed
30. MacKenzie, T, Brown, J, Likosky, D, Wu, Y, Grunkemeier, G. Review of case-mix corrected survival curves. Ann Thorac Surg 2012;93:1416–25. https://doi.org/10.1016/j.athoracsur.2011.12.094.Search in Google Scholar PubMed
31. Martinussen, T, Vansteelandt, S. On collapsibility and confounding bias in Cox and Aalen regression models. Lifetime Data Anal 2013;19:279–96. https://doi.org/10.1007/s10985-013-9242-z.Search in Google Scholar PubMed
32. Aalen, O, Cook, RJ, Røysland, K. Does Cox analysis of a randomized survival study yield a causal treatment effect? Lifetime Data Anal 2015;21:579–93. https://doi.org/10.1007/s10985-015-9335-y.Search in Google Scholar PubMed
33. Hernán, M, Robins, J. Instruments for causal inference: an epidemioligist’s dream? Epidemiology 2006;17:360–72. https://doi.org/10.1097/01.ede.0000222409.00878.37.Search in Google Scholar PubMed
34. Thanasassoulis, P, O’Donnell, T. Mendelian randomization. J Am Med Assoc 2009;301:2386–8. https://doi.org/10.1001/jama.2009.812.Search in Google Scholar PubMed PubMed Central
35. Pearl, J. Causal diagrams for empirical research. Biometrika 1995;82:669–88. https://doi.org/10.1093/biomet/82.4.669.Search in Google Scholar
36. Hougaard, P. Frailty models for survival data. Lifetime Data Anal 1995;1:255–73. https://doi.org/10.1007/bf00985760.Search in Google Scholar
37. Cheng, SC, Fine, JP, Wei, LJ. Prediction of cumulative incidence function under the proportional hazards model. Biometrics 1998;54:219–28. https://doi.org/10.2307/2534009.Search in Google Scholar
38. van der Vaart, A. Asymptotic statistics. Cambridge, UK: Cambridge University Press; 1998.10.1017/CBO9780511802256Search in Google Scholar
39. Martínez-Camblor, P, Pérez-Fernández, S, Corral, N. Efficient nonparametric confidence bands for receiver operating-characteristic curves. Stat Methods Med Res 2018;27:1892–908. https://doi.org/10.1177/0962280216672490.Search in Google Scholar
40. Kosorok, M, Lee, B, Fine, J. Robust inference for univariate proportional hazards frailty regression models. Ann Stat 2004;32:1448–91. https://doi.org/10.1214/009053604000000535.Search in Google Scholar
41. Greenland, S. An introduction to instrumental variables for epidemiologists. Int J Epidemiol 2000;29:722–9. https://doi.org/10.1093/ije/29.4.722.Search in Google Scholar
42. Kang, H, Peck, L, Keele, L. Inference for instrumental variables: a randomization inference approach. J R Stat Soc Ser A Stat Soc 2018;181:1231–54. https://doi.org/10.1111/rssa.12353.Search in Google Scholar
43. Staiger, D, Stock, J. Instrumental variables regression with weak instruments. Econometrica 1997;65:557–86. https://doi.org/10.2307/2171753.Search in Google Scholar
44. Youden, WJ. Index for rating diagnostic tests. Cancer 1950;3:32–5. https://doi.org/10.1002/1097-0142(1950)3:1<32::aid-cncr2820030106%3e3.0.co;2-3.10.1002/1097-0142(1950)3:1<32::AID-CNCR2820030106>3.0.CO;2-3Search in Google Scholar
45. O’Malley, AJ, Zou, KH, Fielding, JR, Tempany, CM. Bayesian regression methodology for estimating a receiver operating characteristic curve with two radiologic applications. Acad Radiol 2001;8:713–25. https://doi.org/10.1016/s1076-6332(03)80578-0.Search in Google Scholar
Supplementary material
As supplementary material, we provide the file supplementary. R which contains the R routines used in the Monte Carlo simulations and in the real data examples. The online version of this article offers supplementary material (https://doi.org/10.1515/ijb-2019-0146).
© 2020 Walter de Gruyter GmbH, Berlin/Boston