×

zbMATH — the first resource for mathematics

Estimating propensity scores and causal survival functions using prevalent survival data. (English) Zbl 1271.62065
Summary: This article develops semiparametric approaches for estimation of propensity scores and causal survival functions from prevalent survival data. The analytical problem arises when the prevalent sampling is adopted for collecting failure times and, as a result, the covariates are incompletely observed due to their association with failure time. The proposed procedure for estimating propensity scores shares interesting features similar to the likelihood formulation in case-control studies, but in our case requires additional considerations in the intercept term. The result shows that the corrected propensity scores in a logistic regression setting can be obtained through standard estimation procedure with specific adjustments on the intercept term. For causal estimation, two different types of missing sources are encountered in our model: one can be explained by potential outcome framework; the other is caused by the prevalent sampling scheme. Statistical analysis without adjusting bias from both sources of missingness will lead to biased results in causal inference. The proposed methods were partly motivated by and applied to the Surveillance, Epidemiology, and End Results (SEER)-Medicare linked data for women diagnosed with breast cancer.

MSC:
62G05 Nonparametric estimation
62N02 Estimation in survival analysis and censored data
62J12 Generalized linear models (logistic models)
62P10 Applications of statistics to biology and medical sciences; meta analysis
92C50 Medical applications (general)
PDF BibTeX XML Cite
Full Text: DOI
References:
[1] Anstrom, Utilizing propensity scores to estimate causal treatment effects with censored time-lagged data, Biometrics 57 pp 1207– (2001) · Zbl 1209.62217 · doi:10.1111/j.0006-341X.2001.01207.x
[2] Asgharian, Length-biased sampling with right censoring: An unconditional approach, Journal of the American Statistical Association 97 pp 202– (2002) · Zbl 1073.62561 · doi:10.1198/016214502753479347
[3] Begg, Methodology for case-control studies with prevalent cases, Biometrika 74 pp 191– (1987) · doi:10.1093/biomet/74.1.191
[4] Bergeron, Covariate bias induced by length-bias sampling of failure times, Journal of the American Statistical Association 103 pp 737– (2008) · Zbl 05564527 · doi:10.1198/016214508000000382
[5] Breslow, Contribution to the discussion of the paper by D.R. Cox, Journal of the Royal Statistical Society Series B 34 pp 216– (1972)
[6] Brookmeyer, Biases in prevalent cohorts, Biometrics 43 pp 739– (1987) · Zbl 0715.62221 · doi:10.2307/2531529
[7] Buchholz, Radiation therapy for early-stage breast cancer after breast-conserving surgery, The New England Journal of Medicine 360 pp 63– (2009) · doi:10.1056/NEJMct0803525
[8] Chao, Influence curves for randomly truncated data, Biometrika 74 pp 426– (1987) · Zbl 0649.62032 · doi:10.1093/biomet/74.2.426
[9] Cox, Regression models and life-tables (with discussion), Journal of the Royal Statistical Society Series B 34 pp 187– (1972) · Zbl 0243.62041
[10] Dehejia, Causal effects in nonexperimental studies: Reevaluating the evaluation of training programs, Journal of the American Statistical Association 94 pp 1053– (1999) · doi:10.1080/01621459.1999.10473858
[11] Greenland, Model-based estimation of relative risks and other epidemiologic measures in studies of common outcomes and in case-control studies, American Journal of Epidemiology 160 pp 301– (2004) · doi:10.1093/aje/kwh221
[12] He, Estimation of the truncation probability in the random truncation model, The Annals of Statistics 26 pp 1011– (1998) · Zbl 0929.62036 · doi:10.1214/aos/1024691086
[13] Heckman, Matching as an econometric evaluation estimator: Evidence from evaluating a job training programme, The Review of Economic Studies 64 pp 605– (1997) · Zbl 0887.90039 · doi:10.2307/2971733
[14] Heckman, Econometric causality, International Statistical Review 76 pp 1– (2008) · Zbl 1379.62094 · doi:10.1111/j.1751-5823.2007.00024.x
[15] Joffe, Invited commentary:propensity scores, American Journal of Epidemiology 150 pp 327– (1999) · doi:10.1093/oxfordjournals.aje.a010011
[16] Kinne, Staging and follow-up of breast cancer patients, Cancer 67 (4) pp 1196– (1991) · doi:10.1002/1097-0142(19910215)67:4+<1196::AID-CNCR2820671515>3.0.CO;2-4
[17] Lane, Analysis of covariance and standardization as instances of prediction, Biometrics 38 pp 613– (1982) · doi:10.2307/2530043
[18] Liang, Regression analysis under non-standard situations: A pairwise pseudolikelihood approach, Journal of the Royal Statistical Society: Series B 62 pp 773– (2000) · Zbl 0963.62068 · doi:10.1111/1467-9868.00263
[19] Luo, Nonparametric estimation for right-censored length-biased data: A pseudo partial likelihood approach, Biometrika 96 pp 873– (2009) · Zbl 1179.62142 · doi:10.1093/biomet/asp064
[20] Mansson, On the estimation and use of propensity scores in case-control and case-cohort studies, American Journal of Epidemiology 166 pp 332– (2007) · doi:10.1093/aje/kwm069
[21] Perez, Strong representation of a generalized product-limit estimator for truncated and censored data with some applications, Journal of Nonparametric Statistics 10 pp 213– (1999) · Zbl 1007.62511 · doi:10.1080/10485259908832761
[22] Prentice, Logistic disease incidence models and case-control studies, Biometrika 66 pp 403– (1979) · Zbl 0428.62078 · doi:10.1093/biomet/66.3.403
[23] Qin, Inferences for case-control and semiparametric two-sample density ratio models, Biometrika 85 pp 619– (1998) · Zbl 0954.62053 · doi:10.1093/biomet/85.3.619
[24] Qin, Statistical methods for analyzing right-censored length-biased data under cox model, Biometrics 66 pp 382– (2010) · Zbl 1192.62209 · doi:10.1111/j.1541-0420.2009.01287.x
[25] Rose, Why match? Investigating matched case-control study designs with causal effect estimation, The International Journal of Biostatistics 5 (1) (2009) · doi:10.2202/1557-4679.1127
[26] Rosenbaum, Model-based direct adjustment, Journal of the American Statistical Association 82 pp 387– (1987) · Zbl 0622.62010 · doi:10.1080/01621459.1987.10478441
[27] Rosenbaum, The central role of the propensity score in observational studies for causal effects, Biometrika 70 pp 41– (1983) · Zbl 0522.62091 · doi:10.1093/biomet/70.1.41
[28] Rubin, Estimating causal effects from large data sets using propensity scores, Annals of Internal Medicine 127 pp 757– (1997) · doi:10.7326/0003-4819-127-8_Part_2-199710151-00064
[29] Shen, Estimation of the truncation probability with left-truncated and right censored data, Journal of Nonparametric Statistics 17 pp 957– (2005) · Zbl 1079.62096 · doi:10.1080/10485250500393222
[30] Shen, Analyzing length-biased data with semiparametric transformation and accelerated failure time models, Journal of the American Statistical Association 104 pp 1192– (2009) · Zbl 1388.62294 · doi:10.1198/jasa.2009.tm08614
[31] Tan, A distributional approach for causal inference using propensity scores, Journal of the American Statistical Association 101 pp 1619– (2006) · Zbl 1171.62320 · doi:10.1198/016214506000000023
[32] Vardi, Nonparametric estimation in the presence of length bias, Annals of Statistics 10 pp 616– (1982) · Zbl 0491.62034 · doi:10.1214/aos/1176345802
[33] Wang, Nonparametric estimation from cross-sectional survival data, Journal of the American Statistical Association 86 pp 130– (1991) · Zbl 0739.62026 · doi:10.1080/01621459.1991.10475011
[34] Wang, Hazards regression analysis for length-biased data, Biometrika 83 pp 343– (1996) · Zbl 0864.62080 · doi:10.1093/biomet/83.2.343
[35] Warren, Overview of the seer-medicare data: Content, research applications, and generalizability to the united states elderly population, Med Care 40 pp 3– (2002) · doi:10.1097/00005650-200208001-00002
[36] Woodroofe, Estimating a distribution function with truncated data, The Annals of Statistics 13 pp 163– (1985) · Zbl 0574.62040 · doi:10.1214/aos/1176346584
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.