Stochastic multiple imputation for latent heterogeneity in interval-censored data: A prior-stratified approach (PS-SMI)

Abstract

Standard parametric regression models for interval-censored time-to-event data frequently depend on a restrictive assumption of population homogeneity. When applied to complex, heterogeneous populations containing unobserved latent subgroups, these traditional models invariably estimate an averaged hazard function. Consequently, this homogenization systematically biases survival projections, overestimating survival times for high-risk participants while yielding overly pessimistic prognoses for low-risk individuals. To address this fundamental limitation and the known instability of traditional iterative mixture models, we introduce the prior-stratified stochastic multiple imputation algorithm, designed to disentangle latent heterogeneity and robustly impute continuous event times. Because estimating expectation-maximization mixtures from sparse interval-censored data often results in component collapse, prior-stratified stochastic multiple imputation structurally bypasses the fragile iterative expectation-maximization loop. Instead, it utilizes a one-time static Bayesian risk stratification, seeded by clinical priors, to anchor the likelihood space. This explicitly decouples the mixture into independent, strictly convex Weibull regressions, guaranteeing stable convergence even under extreme missingness. To rigorously quantify estimation uncertainty, the algorithm subsequently executes multiple repeated stochastic draws from the inferred participant-specific truncated distributions. Simulation studies demonstrate that our algorithm successfully prevents component collapse under extreme censoring (more than 40%) and structural misspecification, maintaining high phenotypic identifiability where standard unanchored mixtures fail entirely. Empirical validation utilizing a semi-synthetic primary biliary cholangitis cohort, the signal tandmobiel dental emergence dataset, and the highly sparse Finkelstein breast cancer dataset confirms the algorithm’s robust capacity to autonomously recover latent risk architectures and track non-parametric (Turnbull) ground truths without supervised labeling. Our algorithm offers a rigorously quantified methodological bridge, converting complex, heterogeneous interval-censored observations into complete datasets, thereby unlocking conventional survival analysis toolkits while safely preserving biological dimorphism.

Keywords

• Expectation-maximization
• interval censoring
• latent class analysis
• multiple imputation
• stochastic imputation

Supporting Institution

The University of BUrdwan

Ethical Statement

This submission was no where sent for publication or review.

References

1. [1] D. G. Kleinbaum and M. Klein, Survival Analysis: A Self-Learning Text, Springer, New York, 1996.
2. [2] J. Sun, Interval censoring, in Encyclopedia of Biostatistics, Wiley, New York, 3, 2090–2095, 1998.
3. [3] K. M. Leung, R. M. Elashoff and A. A. Afifi, Censoring issues in survival analysis, Annu. Rev. Public Health 18 (1), 83–104, 1997.
4. [4] B. W. Turnbull, Nonparametric estimation of a survivorship function with doubly censored data, J. Am. Stat. Assoc. 69 (345), 169–173, 1974.
5. [5] D. Sinha, M. H. Chen and S. K. Ghosh, Bayesian analysis and model selection for interval-censored survival data, Biometrics 55(2), 585–590, 1999.
6. [6] C. B. Guure, N. A. Ibrahim and M. B. Adam, Bayesian inference of the Weibull model based on interval-censored survival data, Comput. Math. Methods Med. 2013, 849520, 2013.
7. [7] X. Lin, B. Cai, L. Wang and Z. Zhang, A Bayesian proportional hazards model for general interval-censored data, Lifetime Data Anal. 21 (3), 470–490, 2015.
8. [8] C. Pan, B. Cai and L. Wang, A Bayesian approach for analyzing partly intervalcensored data under the proportional hazards model, Stat. Methods Med. Res. 29 (11), 3192–3204, 2020.

9. [9] M. C. M. Wong, K. F. Lam and E. C. M. Lo, Bayesian analysis of clustered intervalcensored data, J. Dent. Res. 84 (9), 817–821, 2005.
10. [10] K. Zuma, A Bayesian analysis of correlated interval-censored data, Commun. Stat. Theory Methods 36 (4), 725–730, 2007.
11. [11] D. B. Rubin, An overview of multiple imputation, in Proc. Survey Research Methods Section, American Statistical Association, pp. 79–84, 1988.
12. [12] J. F. Lawless and D. Babineau, Models for interval censoring and simulation-based inference for lifetime distributions, Biometrika 93 (3), 671–686, 2006.
13. [13] S. S. Vallender, Calculation of the Wasserstein distance between probability distributions on the line, Theory Probab. Appl. 18 (4), 784–786, 1974.
14. [14] P. A. Murtaugh, E. R. Dickson, G. M. Van Dam, M. Malinchoc, P. M. Grambsch, A. L. Langworthy and C. H. Gips, Primary biliary cirrhosis: prediction of short-term survival based on repeated participant visits, Hepatology 20 (1), 126–134, 1994.
15. [15] E. Lesaffre, A. Komárek and D. Declerck, An overview of methods for intervalcensored data with an emphasis on applications in dentistry, Stat. Methods Med. Res. 14 (6), 539–552, 2005.
16. [16] D. M. Finkelstein, A proportional hazards model for interval-censored failure time data, Biometrics, 845–854, 1986.
17. [17] J. C. Lindsey and L. M. Ryan, Methods for interval-censored data, Stat. Med. 17 (2), 219–238, 1998.