1. Home
  2. Issues
  3. Volume 20, Issue 2 (2022)
  4. Propensity Score Modeling in Electronic ...

Journal of Data Science


Propensity Score Modeling in Electronic Health Records with Time-to-Event Endpoints: Application to Kidney Transplantation
Volume 20, Issue 2 (2022), pp. 188–208
Pub. online: 20 April 2022      Type: Data Science In Action      Open accessOpen Access
Received
20 July 2021
Accepted
4 April 2022
Published
20 April 2022

Abstract

For large observational studies lacking a control group (unlike randomized controlled trials, RCT), propensity scores (PS) are often the method of choice to account for pre-treatment confounding in baseline characteristics, and thereby avoid substantial bias in treatment estimation. A vast majority of PS techniques focus on average treatment effect estimation, without any clear consensus on how to account for confounders, especially in a multiple treatment setting. Furthermore, for time-to event outcomes, the analytical framework is further complicated in presence of high censoring rates (sometimes, due to non-susceptibility of study units to a disease), imbalance between treatment groups, and clustered nature of the data (where, survival outcomes appear in groups). Motivated by a right-censored kidney transplantation dataset derived from the United Network of Organ Sharing (UNOS), we investigate and compare two recent promising PS procedures, (a) the generalized boosted model (GBM), and (b) the covariate-balancing propensity score (CBPS), in an attempt to decouple the causal effects of treatments (here, study subgroups, such as hepatitis C virus (HCV) positive/negative donors, and positive/negative recipients) on time to death of kidney recipients due to kidney failure, post transplantation. For estimation, we employ a 2-step procedure which addresses various complexities observed in the UNOS database within a unified paradigm. First, to adjust for the large number of confounders on the multiple sub-groups, we fit multinomial PS models via procedures (a) and (b). In the next stage, the estimated PS is incorporated into the likelihood of a semi-parametric cure rate Cox proportional hazard frailty model via inverse probability of treatment weighting, adjusted for multi-center clustering and excess censoring, Our data analysis reveals a more informative and superior performance of the full model in terms of treatment effect estimation, over sub-models that relaxes the various features of the event time dataset.

Supplementary material

 Supplementary Material
Figures and Tables pertaining to the motivating UNOS data analysis, as well as computing code (in R and SAS) and a representative UNOS dataset consisting of a random sample with 19000 observations are available as accompanying supplementary materials associated with this paper.

References

 
Ahmed A, Young JB, Love TE, Levesque R, Pitt B (2008). A propensity-matched study of the effects of chronic diuretic therapy on mortality and hospitalization in older adults with heart failure. International journal of Cardiology, 125: 246–253.
 
Ali MS, Groenwold RH, Pestman WR, Belitser SV, Roes KC, Hoes AW, et al. (2014). Propensity score balance measures in pharmacoepidemiology: A simulation study. Pharmacoepidemiology and Drug Safety, 23: 802–811.
 
Austin PC (2009). Some methods of propensity-score matching had superior performance to others: Results of an empirical investigation and monte carlo simulations. Biometrical Journal, 51: 171–184.
 
Austin PC, Stuart EA (2015). Moving towards best practice when using inverse probability of treatment weighting (iptw) using the propensity score to estimate causal treatment effects in observational studies. Statistics in Medicine, 34: 3661–3679.
 
Barsoum RS, William EA, Khalil SS (2017). Hepatitis C and Kidney disease: A narrative review. Journal of Advanced Research, 8: 113–130.
 
Bouthot BA, Murthy B, Schmid CH, Levey AS, Pereira BJ (1997). Long-term follow-up of hepatitis c virus infection among organ transplant recipients: Implications for policies on organ procurement1, 2. Transplantation, 63: 849–853.
 
Bucci JR, Lentine KL, Agodoa LY, Peters T, Schnitzler M, Abbott K (2004). Outcomes associated with recipient and donor hepatitis C serology status after kidney transplantation in the United States: Analysis of the USRDS/UNOS database. Clinical Transplants, 51–61.
 
Bucci JR, Matsumoto CS, Swanson SJ, Agodoa LY, Holtzmuller KC, Peters TG, et al. (2002). Donor hepatitis C seropositivity: Clinical correlates and effect on early graft and patient survival in adult cadaveric kidney transplantation. Journal of the American Society of Nephrology, 13: 2974–2982.
 
Burgette JM, Preisser JS, Rozier RG (2016). Propensity score weighting: An application to an Early Head Start dental study. Journal of Public Health Dentistry, 76: 17–29.
 
Burnham KP, Anderson DR (2004). Multimodel inference: Understanding AIC and BIC in model selection. Sociological Methods & Research, 33: 261–304.
 
Cacoub P, Comarmond C, Domont F, Savey L, Desbois AC, Saadoun D (2016). Extrahepatic manifestations of chronic hepatitis C virus infection. Therapeutic Advances in Infectious Disease, 3: 3–14.
 
Carbognin S, Solomon N, Yeo F, Swanson S, Bohen E, Koff J, et al. (2006). Acute renal allograft rejection following pegylated ifn-α treatment for chronic hcv in a repeat allograft recipient on hemodialysis: A case report. American Journal of Transplantation, 6: 1746–1751.
 
Carnegie NB, Harada M, Hill JL (2016). Assessing sensitivity to unmeasured confounding using a simulated potential confounder. Journal of Research on Educational Effectiveness, 9: 395–420.
 
Crump RK, Hotz VJ, Imbens GW, Mitnik OA (2006). Moving the goalposts: Addressing limited overlap in the estimation of average treatment effects by changing the estimand. Working Paper 330. National Bureau of Economic Research. http://www.nber.org/papers/t0330.
 
Crump RK, Hotz VJ, Imbens GW, Mitnik OA (2009). Dealing with limited overlap in estimation of average treatment effects. Biometrika, 96: 187–199.
 
Dharnidharka VR, Stevens G, Howard RJ (2005). En-bloc kidney transplantation in the United states: An analysis of united network of organ sharing (UNOS) data from 1987 to 2003. American Journal of Transplantation, 5: 1513–1517.
 
Dow JK, Endersby JW (2004). Multinomial probit and multinomial logit: A comparison of choice models for voting research. Electoral Studies, 23: 107–122.
 
D’Amour A, Ding P, Feller A, Lei L, Sekhon J (2021). Overlap in observational studies with high-dimensional covariates. Journal of Econometrics, 221: 644–654.
 
Feng S, Wolfe RA, Port FK (2005). Frailty survival model analysis of the national deceased donor kidney transplant dataset using poisson variance structures. Journal of the American Statistical Association, 100: 728–735.
 
Friedman J, Hastie T, Tibshirani R (2000). Additive logistic regression: A statistical view of boosting. The Annals of Statistics, 28: 337–407.
 
Gill J, Cho YW, Danovitch GM, Wilkinson A, Lipshutz G, Pham P-T, et al. (2008). Outcomes of dual adult kidney transplants in the United States: An analysis of the OPTN/UNOS database. Transplantation, 85: 62–68.
 
Gill J, Shah T, Hristea I, Chavalitdhamrong D, Anastasi B, Takemoto S, et al. (2009). Outcomes of simultaneous heart–kidney transplant in the us: A retrospective analysis using optn/unos data. American Journal of Transplantation, 9: 844–852.
 
Golub GH, Welsch JH (1969). Calculation of Gaussian quadrature rules. Mathematics of Computation, 23: 221–230.
 
Greifer N (2020). cobalt: Covariate balance tables and plots. https://CRAN.R-project.org/package=cobalt.
 
Gupta G, Kang L, Yu JW, Limkemann AJ, Garcia V, Bandyopadhyay D, et al. (2017). Long-term outcomes and transmission rates in hepatitis c virus-positive donor to hepatitis c virus-negative kidney transplant recipients: Analysis of united states national data. Clinical Transplantation, 31: e13055.
 
Hansen LP (1982). Large sample properties of generalized method of moments estimators. Econometrica: Journal of the Econometric Society, 50: 1029–1054.
 
Harder VS, Stuart EA, Anthony JC (2010). Propensity score techniques and the assessment of measured covariate balance to test causal associations in psychological research. Psychological methods, 15: 234–249.
 
He P, Kong G, Su Z (2013). Estimating the survival functions for right-censored and interval-censored data with piecewise constant hazard functions. Contemporary Clinical Trials, 35: 122–127.
 
Hernán MA, Robins JM (2006). Estimating causal effects from epidemiological data. Journal of Epidemiology & Community Health, 60: 578–586.
 
Hirano K, Imbens GW (2001). Estimation of causal effects using propensity score weighting: An application to data on right heart catheterization. Health Services and Outcomes Research Methodology, 2: 259–278.
 
Huang R, Xu R, Dulai PS (2020). Sensitivity analysis of treatment effect to unmeasured confounding in observational studies with survival and competing risks outcomes. Statistics in Medicine, 39: 3397–3411.
 
Imai K, King G, Stuart EA (2008). Misunderstandings between experimentalists and observationalists about causal inference. Journal of the Royal Statistical Society: Series A (Statistics in Society), 171: 481–502.
 
Imai K, Ratkovic M (2014). Covariate balancing propensity score. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 76: 243–263.
 
Imbens GW (2000). The role of the propensity score in estimating dose-response functions. Biometrika, 87: 706–710.
 
Kamal L, Jonathan WY, Reichman TW, Kang L, Bandyopadhyay D, Kumar D, et al. (2020). Impact of induction immunosuppression strategies in simultaneous liver/kidney transplantation. Transplantation, 104: 395–403.
 
Kang JD, Schafer JL, et al. (2007). Demystifying double robustness: A comparison of alternative strategies for estimating a population mean from incomplete data. Statistical Science, 22: 523–539.
 
Kucirka L, Singer A, Ros R, Montgomery R, Dagher N, Segev D (2010). Underutilization of hepatitis C-positive kidneys for hepatitis C-positive recipients. American Journal of Transplantation, 10: 1238–1246.
 
Lawless J, Zhan M (1998). Analysis of interval-grouped recurrent-event data using piecewise constant rate functions. Canadian Journal of Statistics, 26: 549–565.
 
Lee BK, Lessler J, Stuart EA (2010). Improving propensity score weighting using machine learning. Statistics in Medicine, 29: 337–346.
 
Li F, Morgan KL, Zaslavsky AM (2018). Balancing covariates via propensity score weighting. Journal of the American Statistical Association, 113: 390–400.
 
Li L, Greene T (2013). A weighting analogue to pair matching in propensity score analysis. The International Journal of Biostatistics, 9: 215–234.
 
Lin DY, Wei L-J (1989). The robust inference for the Cox proportional hazards model. Journal of the American statistical Association, 84: 1074–1078.
 
Liu L, Huang X (2008). The use of Gaussian quadrature for estimation in frailty proportional hazards models. Statistics in Medicine, 27: 2665–2683.
 
Loh W-Y (2011). Classification and regression trees. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 1: 14–23.
 
Lunceford JK, Davidian M (2004). Stratification and weighting via the propensity score in estimation of causal treatment effects: A comparative study. Statistics in Medicine, 23: 2937–2960.
 
Ma X, Wang J (2020). Robust inference using inverse probability weighting. Journal of the American Statistical Association, 115: 1851–1860.
 
Maller RA, Zhou S (1992). Estimating the proportion of immunes in a censored sample. Biometrika, 79: 731–739.
 
Maluf DG, Fisher RA, King AL, Gibney EM, Mas VR, Cotterell AH, et al. (2007). Hepatitis C virus infection and kidney transplantation: Predictors of patient and graft survival. Transplantation, 83: 853–857.
 
Mao H, Li L, Yang W, Shen Y (2018). On the propensity score weighting analysis with survival outcome: Estimands, estimation, and inference. Statistics in Medicine, 37: 3745–3763.
 
McCaffrey DF, Griffin BA, Almirall D, Slaughter ME, Ramchand R, Burgette LF (2013). A tutorial on propensity score estimation for multiple treatments using generalized boosted models. Statistics in Medicine, 32: 3388–3414.
 
McCaffrey DF, Ridgeway G, Morral AR (2004). Propensity score estimation with boosted regression for evaluating causal effects in observational studies. Psychological Methods, 9: 403.
 
Mirzaee M, Azmandian J, Zeraati H, Mahmoodi M, Mohammad K, Etminan A, et al. (2014). Short-term and long-term survival of kidney allograft: Cure model analysis. Iranian Journal of Kidney Diseases, 8: 225–230.
 
Morgan C (2018). Reducing bias using propensity score matching. Journal of Nuclear Cardiology, 25: 404–406.
 
Neuhäuser M, Thielmann M, Ruxton GD (2018). The number of strata in propensity score stratification for a binary outcome. Archives of Medical Science, 14: 695–700.
 
Nocedal J, Wright S (2006). Numerical Optimization. Springer Science & Business Media.
 
Ojo AO, Meier-Kriesche H-U, Hanson JA, Leichtman A, Magee JC, Cibrik D, et al. (2001). The impact of simultaneous pancreas-kidney transplantation on long-term patient survival1. Transplantation, 71: 82–89.
 
Owen A (1991). Empirical likelihood for linear models. The Annals of Statistics, 19: 1725–1747.
 
Pereira BG, Wright T, Schmid C, Levey A, Group NEOBHCS, et al. (1995). A controlled study of hepatitis c transmission by organ transplantation. The Lancet, 345: 484–487.
 
Pinheiro JC, Bates DM (1995). Approximations to the log-likelihood function in the nonlinear mixed-effects model. Journal of computational and Graphical Statistics, 4: 12–35.
 
Ridgeway G, McCaffrey D, Morral A, Griffin BA, Burgette L, Cefalu M (2020). twang: Toolkit for Weighting and Analysis of Nonequivalent Groups. URLhttps://CRAN.R-project.org/package=twang. R package version 1.6.
 
Rosenbaum PR (2010). Design of Observational Studies. Springer, New York, NY.
 
Rosenbaum PR, Rubin DB (1983). The central role of the propensity score in observational studies for causal effects. Biometrika, 70: 41–55.
 
Rubin DB (2005). Causal inference using potential outcomes: Design, modeling. decisions. Journal of the American Statistical Association, 100: 322–331.
 
Scott DR, Wong JK, Spicer TS, Dent H, Mensah FK, McDonald S, et al. (2010). Adverse impact of hepatitis c virus infection on renal replacement therapy and renal transplant patients in Australia and New Zealand. Transplantation, 90: 1165–1171.
 
Setodji CM, McCaffrey DF, Burgette LF, Almirall D, Griffin BA (2017). The right tool for the job: Choosing between covariate balancing and generalized boosted model propensity scores. Epidemiology, 28: 802–811.
 
Singh N, Neidlinger N, Djamali A, Leverson G, Voss B, Sollinger HW, et al. (2012). The impact of hepatitis c virus donor and recipient status on long-term kidney transplant outcomes: University of wisconsin experience. Clinical Transplantation, 26: 684–693.
 
Stuart EA (2010). Matching methods for causal inference: A review and a look forward. Statistical Science, 25: 1–21.
 
Stuart EA, Lee BK, Leacy FP (2013). Prognostic score–based balance measures can be a useful diagnostic for propensity score methods in comparative effectiveness research. Journal of Clinical Epidemiology, 66: S84–S90.
 
Sy JP, Taylor JM (2000). Estimation in a cox proportional hazards cure model. Biometrics, 56: 227–236.
 
Testa G, Siegler M (2014). Increasing the supply of kidneys for transplantation by making living donors the preferred source of donor kidneys. Medicine, 93: e318.
 
Vaughan AS, Kelley CF, Luisi N, del Rio C, Sullivan PS, Rosenberg ES (2015). An application of propensity score weighting to quantify the causal effect of rectal sexually transmitted infections on incident HIV among men who have sex with men. BMC Medical Research Methodology, 15.
 
Wei F, Liu J, Liu F, Hu H, Ren H, Hu P (2014). Interferon-based anti-viral therapy for hepatitis c virus infection after renal transplantation: An updated meta-analysis. PLoS One, 9: e90611.
 
Yang S (2018). Propensity score weighting for causal inference with clustered data. Journal of Causal Inference. doi.org/10.1515/jci–2017–0027.
 
Yang S, Ding P (2018). Asymptotic inference of causal effects with observational studies trimmed by the estimated propensity scores. Biometrika, 105: 487–493.
 
Yang S, Imbens GW, Cui Z, Faries DE, Kadziola Z (2016). Propensity score matching and subclassification in observational studies with multi-level treatments. Biometrics, 72: 1055–1065.
 
Yang S, Pieper K, Cools F (2020). Semiparametric estimation of structural failure time model in continuous-time processes. Biometrika, 107: 123–136.
 
Yang S, Tsiatis AA, Blazing M (2018). Modeling survival distribution as a function of time to treatment discontinuation: A dynamic treatment regime approach. Biometrics, 74: 900–909.

Related articles PDF XML
Related articles PDF XML

Copyright
2022 The Author(s). Published by the School of Statistics and the Center for Applied Statistics, Renmin University of China.
Open access article under the CC BY license.

Keywords
cure rate frailty heavy censoring inverse probability weighting kidney transplantation propensity scores

Funding
Bandyopadhyay’s research was supported by grant R01DE024984 from the US National Institutes of Health.

Metrics
since February 2021
1078

Article info
views

492

PDF
downloads


Share


RSS