Addressing missing covariates for the regression analysis of competing risks: Prognostic modelling for triaging patients diagnosed with prostate cancer

Abstract

Competing risks arise in medical research when subjects are exposed to various types or causes of death. Data from large cohort studies usually exhibit subsets of regressors that are missing for some study subjects. Furthermore, such studies often give rise to censored data. In this article, a carefully formulated likelihood-based technique for the regression analysis of right-censored competing risks data when two of the covariates are discrete and partially missing is developed. The approach envisaged here comprises two models: one describes the covariate effects on both long-term incidence and conditional latencies for each cause of death, whilst the other deals with the observation process by which the covariates are missing. The former is formulated with a well-established mixture model and the latter is characterised by copula-based bivariate probability functions for both the missing covariates and the missing data mechanism. The resulting formulation lends itself to the empirical assessment of non-ignorability by performing sensitivity analyses using models with and without a non-ignorable component. The methods are illustrated on a 20-year follow-up involving a prostate cancer cohort from the National Cancer Institutes Surveillance, Epidemiology, and End Results program.

Keywords

discrimination missing data prognostic score prostate cancer treatment risk factor-based triage SEER

Get full access to this article

View all access options for this article.

References

Stephenson

. Prostate cancer trends in the era of prostate-specific antigen. An update of incidence, mortality, and clinical factors from the SEER database. Urol Clin North Am 2002; 29: 173–181.

van Buuren

Boshuizen

Knook

. Multiple imputation of missing blood pressure covariates in survival analysis. Stat Med 1999; 18: 681–694.

Chen

Ibrahim

. Maximum likelihood methods for cure rate models with missing covariates. Biometrics 2001; 57: 43–52.

Unnebrink

Windeler

. Intention-to-treat: methods for dealing with missing values in clinical trials of progressively deteriorating diseases. Stat Med 2001; 20: 3931–3946.

Little

RJA

Rubin

. Statistical analysis with missing data., 2nd ed. New York: Wiley, 2002.

Clark

Altman

. Developing a prognostic model in the presence of missing data: an ovarian cancer case study. J Clin Epidemiol 2003; 56: 28–37.

Moons

KGM

Donders

Stijnen

. Using the outcome for imputation of missing predictor values was preferred. J Clin Epidemiol 2006; 59: 1092–1101.

Rubin

. Inference and missing data. Biometrika 1976; 63: 581–590.

Rubin

Stern

Vehovar

. Handling don’t know survey responses: the case of the Slovenian plebiscite. J Amer Statist Assoc 1995; 90: 822–828.

10.

Schafer

. Analysis of incomplete multivariate data, London: Chapman & Hall, 1997.

11.

Collins

Schafer

Kam

. A comparison of inclusive and restrictive strategies in modern missing data procedures. Psychol Methods 2001; 6: 330–351.

12.

Ibrahim

Chen

Lipsitz

. Missing-data methods for generalized linear models: a comparative review. J Amer Statist Assoc 2005; 100: 332–346.

13.

Ibrahim

. Incomplete data in generalized linear models. J Amer Statist Assoc 1990; 85: 765–769.

14.

Lipsitz

Ibrahim

. A conditional model for incomplete covariates in parametric regression models. Biometrika 1996; 83: 916–922.

15.

White

Royston

. Imputing missing covariate values for the Cox model. Stat Med 2009; 28: 1982–1998.

16.

. Multiple augmentation with partial missing regressors. Biom J 2006; 48: 83–92.

17.

Rotnitzky

Robins

Scharfstein

. Semiparametric regression for repeated outcomes with nonignorable nonresponse. J Am Stat Assoc 1998; 93: 1321–1339.

18.

Molenberghs

Verbeke

Thijs

. Influence analysis to assess sensitivity of the dropout process. Comput Stat Data Anal 2001; 37: 93–113.

19.

Verbeke

Molenberghs

Thijs

. Sensitivity analysis for nonrandom dropout: a local influence approach. Biometrics 2001; 57: 7–14.

20.

Bill-Axelson

Holmberg

Ruutu

. Radical prostatectomy versus watchful waiting in early prostate cancer. N Engl J Med 2005; 352: 1977–1984.

21.

Thompson

Thrasher

Aus

. AUA prostate cancer clinical guideline update panel. Guideline for the management of clinically localized Prostate cancer: 2007 update. J Urol 2007; 177: 2106–2131.

22.

Mohler J, Bahnson RR, Boston B, et al. NCCN clinical practice guidelines in oncology: prostate cancer. J Natl Compr Canc Netw 2010; 8: 162–200.

23.

Larson

Dinse

. A mixture model for the regression analysis of competing risks data. Appl Statist 1985; 34: 201–211.

24.

Carroll

. On the use and utility of the Weibull model in the analysis of survival data. Controlled Clin Trials 2003; 24: 682–701.

25.

Collet

. Modelling survival data in medical research., 2nd ed. Boca Raton: Chapman and Hall/CRC, 2003.

26.

Fine

Gray

. A proportional hazards model for the subdistribution of a competing risk. J Amer Statist Assoc 1999; 94: 496–509.

27.

Maller

Zhou

. Survival analysis with long-term survivors, Chichester: Wiley, 1996.

28.

Ibrahim

Lipsitz

Chen

. Missing covariates in generalized linear models when the missing data mechanism is nonignorable. J R Statist Soc B 1999; 61: 173–190.

29.

Horton

Laird

. Maximum likelihood analysis of generalized linear models with missing covariates. Stat Methods Med Res 1998; 8: 37–50.

30.

Joe

. Multivariate models and dependence concepts, New York: Chapman & Hall, 1997.

31.

Nelsen

. An introduction to copulas., 2nd ed. New York: Springer-Verlag, 2006.

32.

Song

PX-K

. Multivariate dispersion models generated from Gaussian copula. Scand J Stat 2000; 27: 305–320.

33.

Nelsen

. Properties of a one-parameter family of bivariate distributions with specified marginals. Commun Stat-Theor M 1986; 15: 3277–3285.

34.

Genest

. Franks family of bivariate distributions. Biometrika 1987; 74: 549–555.

35.

Frank

. On the simultaneous associativity of F(x,y) and x+y-F(x,y). Aequat Math 1979; 19: 194–226.

36.

Cox

Snell

. Analysis of binary data., 2nd ed. London: Chapman and Hall, 1989, pp. 155–157.

37.

McCullagh

Nelder

. Generalized linear models., 2nd ed. London: Chapman & Hall, 1989.

38.

Raftery

. Bayesian model selection in social research. Sociol Methodol 1995; 25: 111–163.

39.

Ibrahim

. Comment: Incomplete data in clinical studies: analysis, sensitivity, and sensitivity analysis. Drug Inf J 2009; 43: 431–432.

40.

Kemp

Loukas

. The computer generation of bivariate discrete random variables. J R Statist Soc A 1978; 141: 513–519.

41.

Ries LAG, Eisner MP, Kosary CL, et al. (eds). SEER cancer statistics review, 1973–1999 Report. Bethesda, MD: National Cancer Institute, http://seer.cancer.gov/csr/1973_1999/ (2002, accessed 18 May, 2013).

42.

Golimbu

Provet

Al-Askari

. Radical prostatectomy for stage-D1 prostate-cancer: prognostic variables and results of treatment. Urology 1987; 30: 427–435.

43.

Russell

Laramore

Krall

. Eight years experience with neutron radiotherapy in the treatment of stages C and D prostate cancer: updated results of the RTOG 7704 randomized clinical trial. Prostate 1987; 11: 183–193.

44.

Myers

Zincke

Fleming

. Hormonal treatment at time of radical retropubic prostatectomy for stage D1 prostate cancer. J Urol 1983; 130: 99–101.

45.

Chambers JM and Hastie TJ. Statistical models. In: Chambers JM and Hastie TJ (eds) Statistical models in S. Pacific Grove, CA: Wadsworth & Brooks/Cole, 1992, pp.13–44.

46.

Draper

Smith

. Applied regression analysis., 2nd ed. New York: Wiley, 1981, pp. 305–307.

47.

Aalen

. Nonparametric estimation of partial transition probabilities in multiple decrement models. Ann Stat 1978; 6: 534–545.

48.

Aalen

. Nonparametric inference for a family of counting processes. Ann Stat 1978; 6: 701–726.

49.

Putter

Fiocco

Geskus

. Tutorial in biostatistics: competing risks and multi-state models. Stat Med 2007; 26: 2389–2430.

50.

Kattan

Heller

Brennan

. A competing-risks nomogram for sarcoma-specific death following local recurrence. Stat Med 2003; 30: 3515–3525.

51.

Hjort

. On inference in parametric survival data models. Int Stat Rev 1992; 60: 355–387.

52.

Thadhani

Tonelli

. Cohort studies: marching forward. Clin J Am Soc Nephrol 2006; 1: 1117–1123.