Journal of Statistical Theory and Practice

, Volume 7, Issue 2, pp 421–441 | Cite as

Correcting the Results of the Wrong Model: Treatment Effects Under Early Detection of Cancer

  • Shih-Yuan Lee
  • Alex TsodikovEmail author


Early detection of cancer leads to variability of the point of diagnosis advanced by the amount of the so-called lead time, a random variable. Estimated treatment effects by the proportional hazards (PH) model may be biased if this variability is ignored. We study how true and PH-estimated treatment effects differ in screened versus unscreened populations and offer an approximate correction for the reported PH-based estimate that does not require raw data, targeting a meta-analysis-type application. We rely on a joint cancer incidence and survival model of prostate cancer to furnish key information for the correction. The procedure is applied to a series of prostate cancer data analyses using the PH models reported in the literature. Simulations are used for assessing the quality of the method and sensitivity analyses.


Bias Early detection Lead time Misspecified model Proportional hazards 

AMS Subject Classification

62A01 62P10 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Albertsen, P., J. Hanley, D. Penson, G. Barrows, and J. Fine. 2007. 13-year outcomes following treatment for clinically localized prostate cancer in a population based cohort. J. Urol., 177(3), 932–936.CrossRefGoogle Scholar
  2. Anderson, G., and T. Fleming. 1995. Model misspecification in proportional hazards regression. Biometrika, 82, 527–541.MathSciNetCrossRefGoogle Scholar
  3. Bill-Axelson, A., L. Holmberg, M. Ruutu, M. Haggman, S. Andersson, S. Bratell, A. Spangberg, C. Busch, S. Nordling, H. Garmo, et al. 2005. Radical prostatectomy versus watchful waiting in early prostate cancer. N. Engl. J. Med., 352(19), 1977–1984.CrossRefGoogle Scholar
  4. Cox, D. 1972. Regression models and life-tables. J. R. Stat. Soc. Ser. B (Methodological), 34, 187–220.MathSciNetzbMATHGoogle Scholar
  5. Etzioni, R., A. Tsodikov, A. Mariotto, A. Szabo, S. Falcon, J. Wegelin, D. diTommaso, K. Karnofski, R. Gulati, D. Penson, and E. Feuer. 2008. Quantifying the role of PSA screening in the US prostate cancer mortality decline. Cancer Causes Control, 19, 175–181.CrossRefGoogle Scholar
  6. Fienberg, S. 1980. The analysis of cross-classified categorical data. Cambridge, MA., MIT Press.zbMATHGoogle Scholar
  7. Gail, M., S. Wieand, and S. Piantadosi. 1984. Biased estimates of treatment effect in randomized experiments with nonlinear regressions and omitted covariates. Biometrika, 71(3), 431–444.MathSciNetCrossRefGoogle Scholar
  8. Kempthorne, O. 1977. Why randomize. J. Stat. Plan. Inference, 1, 1–25.MathSciNetCrossRefGoogle Scholar
  9. Lagakos, S., and D. Schoenfeld. 1984. Properties of proportional-hazards score tests under misspecified regression models. Biometrics, 40(4), 1037–1048.MathSciNetCrossRefGoogle Scholar
  10. Lin, D. Y., and L. J. Wei. 1989. The robust inference for the Cox proportional hazards model. J. Am. Stat. Associ., 84, 1074–1078.MathSciNetCrossRefGoogle Scholar
  11. Mariotto, A., R. Etzioni, M. Krapcho, and E. Feuer. 2007. Reconstructing prostate-specific antigen (PSA) testing patterns among black and white men in the US from Medicare claims and the National Health Interview Survey. Cancer, 109, 1877–1886.CrossRefGoogle Scholar
  12. Schumacher, M., M. Olschewski, and C. Schmoor. 1987. The impact of heterogeneity on the comparison of survival times. Stat. Medi., 6(7), 773–784.CrossRefGoogle Scholar
  13. Shea, G. 1979. Monotone regression and covariance structure. Ann. Stat., 7, 1121–1126.MathSciNetCrossRefGoogle Scholar
  14. Struthers, C., and J. Kalbfleisch. 1986. Misspecified proportional hazard models. Biometrika, 73(2), 363–369.MathSciNetCrossRefGoogle Scholar
  15. Tewari, A., J. Raman, P. Chang, S. Rao, G. Divine, and M. Menon. 2006. Long-term survival probability in men with clinically localized prostate cancer treated either conservatively or with definitive treatment (radiotherapy or radical prostatectomy). Urology, 68(6), 1268–1274.CrossRefGoogle Scholar
  16. Tsodikov, A. 2003. Semiparametric models: A generalized self-consistency approach. J. R. Stat. Soci. Seri. B Stat. Methodol., 65, 759–774.MathSciNetCrossRefGoogle Scholar
  17. Tsodikov, A., A. Szabo, and J. Wegelin. 2006. A population model of prostate cancer incidence. Stat. Medi., 25(16), 2846–2866.MathSciNetCrossRefGoogle Scholar
  18. Van der Vaart, A. (2000). Asymptotic statistics. Cambridge, UK, Cambridge University Press.Google Scholar

Copyright information

© Grace Scientific Publishing 2013

Authors and Affiliations

  1. 1.Millennium PharmaceuticalsCambridgeUSA
  2. 2.Department of BiostatisticsUniversity of Michigan School of Public HealthAnn ArborUSA

Personalised recommendations