Recovery of Information and Adjustment for Dependent Censoring Using Surrogate Markers

  • James M. Robins
  • Andrea Rotnitzky

Abstract

A class of tests and estimators for the parameters of the Cox proportional hazards model, the accelerated failure time model, and a model for the effect of treatment on the mean of a response variable of interest are proposed that use surrogate marker data to recover information lost due to independent censoring and to adjust for bias due to dependent censoring in randomized clinical trials. We construct an adaptive test that (i) is asymptotically distribution free under the null hypothesis of no treatment effect on survival, (ii) incorporates surrogate marker data, and (iii) is guaranteed to be locally more powerful than the ordinary log-rank test against proportional hazards alternatives when the baseline failure time distribution is Weibull. The proposed test is shown to outperform the log-rank test in a series of simulation experiments. We also prove the optimal estimator within our class is semiparametric efficient by first showing that our estimation problem is a special case of the general problem of parameter estimation in an arbitrary semiparametric model with data missing at random, and then deriving a representation for the efficient score in this more general problem.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Bibliography

  1. Andersen, P.K. and Gill, R.D. (1982). Cox’s regression model for counting processes: A large sample study. Ann. Statist. 10: 1100–1120.CrossRefGoogle Scholar
  2. Begun, J.M., Hall, W.J., Huang, W.M. and Wellner, J.A. (1983). Information and asymptotic efficiency in parametric-nonparametric models. Ann. Statist. 11: 432–452.CrossRefGoogle Scholar
  3. Bickel, P., Klassen, C., Ritov, Y., and Wellner, J. (1990). Efficient and adaptive inferences in semiparametric models. (forthcoming)Google Scholar
  4. Chamberlain, G. (1987). Asymptotic efficiency in estimation with conditional moment restrictions. J. Econometrics. 34: 305–324.CrossRefGoogle Scholar
  5. Gray, R.J., and Tsiatis, A.A. (1989). The linear rank test for use when the main interest is in differences and cure rates. Biometrics 45: 899–904.PubMedCrossRefGoogle Scholar
  6. Gill, R. (1980). Censoring and Stochastic Integrals, Mathematical Center Tract 124, Mathematische Centrum, Amsterdam.Google Scholar
  7. Kalbfleisch, J.D. and Prentice, R.L. (1980). The Statistical Analysis of Failure Time Data. Wiley, New York.Google Scholar
  8. Kress, R. (1990). Linear integral eauations. Springer-Verlag Berlin.Google Scholar
  9. Koul, H., Susarla, V. and Van Ryzin, J. (1981). Regression analysis with randomly right censored data. Ann. Statist. 9: 1276–88.CrossRefGoogle Scholar
  10. Lagakos, S., Lim, L., and Robins, J.M. (1990). Adjusting for early treatment termination in comparative clinical trials. Statist. Med. 9: 1417–1424.CrossRefGoogle Scholar
  11. Newey, W.K. (1990a). Semiparametric efficiency bounds., T Anol. Econom. 5: 99–135.Google Scholar
  12. Newey, W.K. (1990b). Efficient estimation of Tobit models under conditional symmetry in W. Barnett, J. Powell, and G. Tauchen, (eds), Semiparametric and Non-Parametric Methods in Econometrics and Statistics, Cambridge Univ. Press, Cambridge, pp. 291–336.Google Scholar
  13. Newey, W.K. (1992a). The variance of semiparametric estimators. Econometrics (forthcoming).Google Scholar
  14. Newey, W.K. (1992b). Efficient estimation of semiparametric models via moment restrictions, manuscript, MIT.Google Scholar
  15. Pakes, A. and Pollard, D. (1989). Simulation in the asymptotics of optimization estimators. Econometrica 57: 1027–1057.CrossRefGoogle Scholar
  16. Prentice, R.L. (1978). Linear rank tests with right censored data. Biometrika 65: 167–179.CrossRefGoogle Scholar
  17. Ritov, Y. and Wellner, J.A. (1988). Censoring, martingales, and the Cox model. Contemo. Math. Statist. Inf. Stoch. Pro. (N.U. Prabhu, ed.) 80:191–220. Amer. Math. Soc.Google Scholar
  18. Ritov, Y. (1990). Estimation in a linear regression model with censored data. Ann. Statist. 18: 303–328.CrossRefGoogle Scholar
  19. Robins, J.M. (1991). Estimating regression parameters in the presence of dependent censoring. (submitted)Google Scholar
  20. Robins, J.M. and Tsiatis, A.A. (1991). Correcting for non-compliance in randomized trials using rank preserving structural failure time models. Commun. Statist.-Theory Meth. 20: 2609–2631.CrossRefGoogle Scholar
  21. Robins, J.M., Rotnitzky, A., and Zhao, L.P. (1992). Analysis of semiparametric regression models for repeated outcomes under the presence of missing data. (submitted)Google Scholar
  22. Robins, J.M., and Rotnitzky, A. (1992). Semiparametric efficiency in multivariate regression models with missing data. (submitted)Google Scholar
  23. Robins, J.M., Zhao, L.P., Rotnitzky, A., and Lipsitz, S. (1991). Estimation of regression coefficients when a regressor is not always observed. (submitted).Google Scholar
  24. Rubin, D.B. (1976). Inference and missing data. Biometrika 63: 581–92.CrossRefGoogle Scholar
  25. Tsiatis, A.A. (1990). Estimating regression parameters using linear rank tests for censored data. Ann. Statist. 18: 354–372.CrossRefGoogle Scholar
  26. Wei, L.J., Ying, Z., and Lin, D.Y. (1990). Linear regression analysis of censored survival data based on rank tests. Biometrika 77: 845–852.CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 1992

Authors and Affiliations

  • James M. Robins
    • 1
  • Andrea Rotnitzky
    • 1
  1. 1.Departments of Epidemiology and BiostatisticsHarvard School of Public HealthBostonUSA

Personalised recommendations