Pith. sign in

REVIEW 1 major objections 6 minor 58 references

A Bayesian joint model with gamma-process priors ties multiple recurrent event types to death and beats a frequentist EM fit.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review

2026-08-04 17:53 UTC pith:FVZGOYZX

load-bearing objection Solid Bayesian machinery, but the simulation does not generate data from the proposed model, and the ALLHAT numbers contradict the published trial; needs major revision before the claims can be trusted. the 1 major comments →

arxiv 2509.10354 v1 pith:FVZGOYZX submitted 2025-09-12 stat.ME

Bayesian Semiparametric Joint Modeling of Gap-Time Distribution for Multitype Recurrent Events and a Terminal Event

Mithun Kumar Acharjee , AKM Fazlur Rahman This is my paper
classification stat.ME MSC 62N0162F1562P10
keywords recurrent eventsterminal eventjoint frailty modelgap timegamma process priorBayesian semiparametriccumulative hazardALLHAT
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper proposes a Bayesian semiparametric joint model for data in which patients experience several types of recurrent events—such as heart attacks and strokes—and eventually a terminal event, death. The model works on the gap time between events, links all event types through a single shared frailty, and places gamma-process priors on the baseline hazards, giving smooth nonparametric baselines. Its main technical claim is that the cumulative hazard and frailty estimators have closed forms and reduce to Breslow–Aalen-type estimators as the prior precision goes to zero. In simulations the Bayesian fit is reported to be more accurate, more robust to initialization and prior choices, and 2–4 times faster than a comparable frequentist EM fit. Applied to the ALLHAT trial, the model estimates treatment and race effects on acute and chronic cardiovascular events and death while quantifying unobserved dependence.

Core claim

On the author's terms, the paper establishes that a joint shared-frailty model on the gap-time scale with independent gamma-process priors on each baseline cumulative hazard yields tractable full conditional posteriors: the baseline hazards and frailties are updated in closed form, and the Bayes estimators in Eqs. (11)–(12) are explicit sums over the time grid. Taking the gamma-process precision parameters to zero recovers the Breslow–Aalen cumulative hazard estimator, so the Bayesian procedure contains the classical nonparametric estimator as a special case. The sampler avoids large matrix factorizations and scales nearly linearly with sample size. The authors argue that this combination—cl

What carries the argument

The central object is the shared gamma frailty W_i (unit mean, variance 1/nu) that multiplies every type-specific hazard, inducing dependence among recurrent event types and the terminal event. On the gap-time scale, each event type has its own baseline hazard assigned a gamma-process prior, which makes the increments Gamma-distributed and leads to conjugate conditional posteriors for the baseline cumulative hazards and for W_i. The key identities are the closed-form posterior mean estimators for the baseline cumulative hazards (Eqs. 11–12) and for the frailty (Eq. 10), which are sums over the grid of event-count increments divided by weighted at-risk sums; these are what connect the Bayesia

Load-bearing premise

The model assumes that, once the shared frailty and covariates are known, all gap times of every event type and the terminal event are independent, and censoring is noninformative—so if one event type directly changes another type's future hazard, or dropout depends on outcome, estimates will be biased.

What would settle it

Simulate data with two event types whose dependence is not captured by a single shared gamma frailty—for example, with type-specific frailties or with an acute event raising the subsequent chronic-event hazard—and check whether the model's posterior intervals cover the true regression and frailty parameters and whether the closed-form cumulative hazard estimates deviate from the truth. A direct comparison on such data would settle the scope of the claim.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • Clinicians and epidemiologists can fit a joint model of multiple recurrent event types and death without choosing arbitrary parametric baseline hazards, since gamma-process priors adapt to the hazard shape.
  • Because the cumulative hazard estimator has a closed form and reduces to the classical Breslow–Aalen estimator, results from this Bayesian model can be compared directly with standard frequentist nonparametric estimates.
  • The near-linear scaling and reported 2–4x speed gain over EM in the simulations make routine sensitivity analyses and larger datasets feasible.
  • The ALLHAT analysis illustrates the substantive payoff: treatment comparisons (Amlodipine and Lisinopril vs Chlorthalidone), race, and age effects on acute and chronic cardiovascular events and death, with a frailty estimate indicating strong within-patient dependence.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The precision-to-zero limit suggests a broader bridge: other nonparametric Bayesian priors with a precision parameter may similarly recover classical nonparametric estimators, giving a principled Bayesian justification for running the classical analysis.
  • The single shared frailty is a strong structural assumption; a natural testable extension is to allow type-specific frailties or a factor model with more than one latent dimension and compare predictive performance on data with heterogeneous dependence.
  • The gap-time formulation invites modeling the effect of past events on future hazards, e.g., a prior acute event changing the subsequent chronic-event hazard; the current model assumes that dynamics are captured entirely by the frailty and covariates.
  • The authors' stated limitation of noninformative censoring suggests a concrete extension: a joint model with dependent censoring, which would be needed in studies where dropout is related to disease burden.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

1 major / 6 minor

Summary. The paper proposes a Bayesian semiparametric joint frailty model for multitype recurrent events and a terminal event on the gap-time scale. A shared gamma frailty links all event types; each baseline cumulative hazard is assigned an independent gamma-process prior. The authors derive closed-form conditional posterior forms for the cumulative hazard and frailty components, propose a Metropolis-within-Gibbs sampler, and claim that the cumulative hazard estimators recover Breslow-Aalen-type estimators as the prior precision tends to zero. The empirical evaluation compares the Bayesian approach with a frequentist EM-based Breslow-type estimator in simulations and applies the model to ALLHAT data, with the abstract asserting 'superior performance in each criterion.'

Significance. If the model and computational claims are correct, the paper offers a useful and relatively simple Bayesian alternative for joint analysis of multitype recurrent events and a terminal event. The gamma-process prior yields closed-form posterior updates, and the near-linear scaling MCMC sampler is attractive for large biomedical datasets. The ALLHAT application addresses a clinically relevant question. However, the simulation design does not generate data from the model specified in Section 2.1, and several internal notation errors affect the likelihood and posterior derivations. The current evidence is therefore not sufficient to support the central claims; the paper would need corrected derivations and a redesigned simulation study.

major comments (1)
  1. [3.1.1] The simulation DGP resets the gap-time clock for all event types after every event: 'At each gap origin, ... draw independent candidate gaps ... The next event time is min(T1,T2,T0); ... reset the gap-time clock, and repeat.' This does not generate data from the type-specific renewal model of Section 2.1, where T^q_ij are IID gaps between successive type-q events and the likelihood uses B_qi(u)=u minus S_qi at the last type-q event, i.e., time since the last type-q event. Under the simulation, the hazard for a type-q event at time u depends on the time since the last event of any type, not on the time since the last type-q event. The terminal event is also re-drawn after each recurrent event, so its hazard depends on time since the last recurrent event rather than on time since study entry. Consequently, Tables 1-2 and Figures 2-5 evaluate the method under a misspecified DGP and cannot s
minor comments (6)
  1. [Abstract] 'We proposed' should be 'We propose'.
  2. [2.2, Eq. (2)] The product-integral notation is used heavily but not defined for readers unfamiliar with Jacod's formulation; a brief explanation or reference to Andersen et al. [5] would help.
  3. [3.1.1, 'Bayesian priors'] The 'misspecified' baseline-prior analysis (Figure 3) replaces the reference mean functions with exponential forms and also changes the precision c_q, c_0 from 0.1 to 0.01. It would be clearer to separate misspecification of the baseline mean from the change in precision.
  4. [Table 7] The columns labeled 'Lower CI' and 'Upper CI' appear to be on the log-hazard scale, while the HR column is exponentiated. Please label the columns explicitly (e.g., log-scale 95% CI) to avoid confusion.
  5. [References] Reference [21] and reference [45] are the same work (Ripatti and Palmgren, 2000) and should be merged or cited once.
  6. [Figure 6] The three panels are referred to as (a), (b), and (c) in the text; the figure caption should indicate the correspondence between the panels and the event types.

Circularity Check

0 steps flagged

No significant circularity; the derivation is a standard Bayesian calculation from stated likelihood and priors.

full rationale

The derivation chain is self-contained and non-circular. Sections 2.2–2.3 start from a counting-process likelihood (Eq. 2), apply a gamma-process prior (Kalbfleisch 1978), and obtain gamma conditional posteriors (Eqs. 5–6). The estimators in Eqs. 7–8 are posterior means under squared-error loss; replacing W_i by its posterior mean (Eq. 10) yields Eqs. 11–12. The 'recovers Breslow–Aalen' statement is an algebraic limit as c->0, not an input. The only self-citation, Rahman et al. (2014) [11], is background for single-type gap-time estimation and is not used to justify the joint model or the closed-form hazards; the load-bearing citations (Kalbfleisch; Peña et al.) are independent. The simulation in §3.1.1 resets the gap-time clock for all event types after every event, so the DGP does not match the type-specific renewal assumption in §2.1; this is a substantive validity problem for the numerical comparisons, but it is not a circular reduction of the derivation to its inputs. The Discussion's limitation statement about independent frailties and noninformative censoring is an honest caveat, not a hidden circularity. No prediction is forced by a fitted parameter, and no uniqueness theorem from the authors is invoked.

Axiom & Free-Parameter Ledger

4 free parameters · 4 axioms · 1 invented entities

The model depends on standard counting-process likelihood theory, the gamma-process prior construction with a user-chosen partition, and assumptions of conditional independence given the frailty and noninformative censoring. The free parameters are the prior precision, reference baseline means, grid mesh, and prior hyperparameters; in the main simulation the reference means are chosen close to the true generator values, which mildly favors the Bayesian method. No new physical entities are introduced; the shared frailty is a standard latent variable.

free parameters (4)
  • c_q, c_0 (gamma-process precision) = 0.1 (0.01 in robustness section)
    Chosen by hand; controls how strongly the baseline hazards are pulled toward the reference means.
  • Working grid partition = {0, 0.03, ..., 3}
    Chosen by hand; finer meshes change the effective prior on the cumulative hazard.
  • Reference baseline means Λ*_0q, Λ*_00 = (t/1.1)^γ, (t/1.0)^γ, (t/3.1)^γ
    Prior means of the gamma processes; in the main simulation these are close to the true generator baselines (1.2, 1.1, 3.2), giving the Bayesian estimator a mild advantage over the EM comparator.
  • Prior hyperparameters for ν and β = ν ~ Gamma(2,2), β ~ N(0,1)
    Chosen by hand; sensitivity analyses in Section 3.1.2 show moderate robustness.
axioms (4)
  • standard math Joint likelihood for counting processes (Jacod 1975; Andersen et al. 2012) used to write Eq. (2).
    The full likelihood expression is justified by product-integral and martingale results for multivariate counting processes. Invoked in Section 2.2.
  • domain assumption Given shared frailty W_i and covariates, all gap times for each type and the terminal event time are independent; censoring is noninformative and independent of event times.
    Stated in Section 2.1 ('It is assumed that τ_i and {T^q_ij} are mutually independent') and acknowledged as a limitation in the Discussion.
  • domain assumption Baseline cumulative hazards are a priori independent gamma processes with independent increments over a fixed partition.
    Prior specification in Section 2.3; the choice of partition grid and precision are user inputs and are varied in sensitivity analyses.
  • domain assumption Frailty distribution is Gamma(ν, ν) with unit mean.
    Section 2.1 assumes this parametric form to make the frailty parameter ν identifiable; the log-normal prior is used only as a sensitivity check.
invented entities (1)
  • Shared frailty W_i no independent evidence
    purpose: A subject-level latent variable multiplying the baseline hazard for every recurrent event type and the terminal event, capturing within-subject dependence.
    Standard latent variable in frailty models; not directly observed or validated in this paper. It is not a new physical entity, but it is a postulated construct whose identifiability relies entirely on the observed event process.

reviewed 2026-08-04 · how reviews work

0 comments
read the original abstract

In biomedical settings, multitype recurrent events such as stroke and heart failure occur frequently, often concluding with a terminal event such as death. Understanding the links between these recurring and terminal events is fundamental to developing interventions that delay detrimental outcomes. Joint modeling is needed to quantify the dependence between event types and between recurrent events and mortality. We propose a Bayesian semiparametric joint model on the gap-time scale for multitype recurrent events and a terminal event. The model includes a shared frailty that links all recurrent types and the terminal event. Each baseline hazard is assigned a gamma-process prior, while regression and frailty parameters receive standard parametric priors. This ensures flexible baselines and familiar effect measures. The construction gives closed-form expressions for the cumulative hazard and frailty component and connects to Breslow-Aalen type estimators as a special case of our estimator, linking the Bayesian procedure to the classical approach. Computationally, we develop a simple MCMC sampler that avoids large matrix factorizations and scales nearly linearly in sample size. A comprehensive simulation evaluates four criteria: accuracy, prediction, robustness, and computation. There is no exact frequentist version of our specification; for comparison, we fit the same model with an EM algorithm in a frequentist framework. Our model and MCMC algorithm demonstrate superior performance on each criterion. We illustrate the approach with data from the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT), jointly analyzing acute and chronic cardiovascular recurrences and death.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

58 extracted references · 32 canonical work pages

  1. [1]

    W., Quinn, R

    Ravani, P., Gillespie, B. W., Quinn, R. R., MacRae, J., Manns, B., Mendelssohn, D., Tonelli, M., Hemmelgarn, B., James, M., Pannu, N.,et al.(2013). Tem- poral risk profile for infectious and noninfectious complications of hemodialysis access.Journal of the American Society of Nephrology, 24(10), 1668–1677. https://doi.org/10.1681/ASN.2012121234

    work page doi:10.1681/asn.2012121234 2013

  2. [2]

    Tuli, S., Drake, J., Lawless, J., Wigg, M., & Lamberti-Pasculli, M. (2000). Risk factors for repeated cerebrospinal shunt failures in pedi- atric patients with hydrocephalus.Journal of Neurosurgery, 92(1), 31–38. https://doi.org/10.3171/jns.2000.92.1.0031

    work page doi:10.3171/jns.2000.92.1.0031 2000

  3. [3]

    Byar, D. P. (1980). The Veterans Administration study of chemoprophylaxis for recurrent stage I bladder tumours: Comparisons of placebo, pyridoxine and topical thiotepa. InBladder Tumors and Other Topics in Urological Oncology(pp. 363– 370). Springer

    1980

  4. [4]

    K., & Gill, R

    Andersen, P. K., & Gill, R. D. (1982). Cox’s regression model for counting processes: A large sample study.The Annals of Statistics, 10(4), 1100–1120. https://doi.org/10.1214/aos/1176345976

    arXiv 1982

  5. [5]

    K., Borgan, Ø., Gill, R

    Andersen, P. K., Borgan, Ø., Gill, R. D., & Keiding, N. (2012).Statistical Models Based on Counting Processes. Springer. https://doi.org/10.1007/978-1- 4612-4348-9

    doi:10.1007/978-1- 2012

  6. [6]

    S., & Cai, J

    Pepe, M. S., & Cai, J. (1993). Some graphical displays and marginal regression analyses for recurrent failure times and time-dependent covari- ates.Journal of the American Statistical Association, 88(423), 811–820. https://doi.org/10.1080/01621459.1993.10476346

    arXiv 1993

  7. [7]

    F., & Nadeau, C

    Lawless, J. F., & Nadeau, C. (1995). Some simple robust methods for the analysis of recurrent events.Technometrics, 37(2), 158–168. https://doi.org/10.1080/00401706.1995.10484302

    arXiv 1995

  8. [8]

    Y., Wei, L.-J., Yang, I., & Ying, Z

    Lin, D. Y., Wei, L.-J., Yang, I., & Ying, Z. (2000). Semiparametric regression for the mean and rate functions of recurrent events.Journal of the Royal Statistical Society: Series B, 62(4), 711–730. https://doi.org/10.1111/1467-9868.00264

    arXiv 2000

  9. [9]

    A., Strawderman, R

    Pe˜ na, E. A., Strawderman, R. L., & Hollander, M. (2001). Nonparametric estima- tion with recurrent event data.Journal of the American Statistical Association, 96(456), 1299–1315. https://doi.org/10.1198/016214501753381922

    work page doi:10.1198/016214501753381922 2001

  10. [10]

    A., Slate, E

    Pe˜ na, E. A., Slate, E. H., & Gonz´ alez, J. R. (2007). Semiparametric inference for a general class of models for recurrent events.Journal of Statistical Planning and Inference, 137(6), 1727–1747. https://doi.org/10.1016/j.jspi.2006.05.014 29

    doi:10.1016/j.jspi.2006.05.014 2007

  11. [11]

    Rahman, A. K. M. F., Lynch, J. D., & Pe˜ na, E. A. (2014). Non- parametric Bayes estimation of gap-time distribution with recurrent event data.Journal of Nonparametric Statistics, 26(3), 575–598. https://doi.org/10.1080/10485252.2014.910698

    arXiv 2014

  12. [12]

    Yu, Z., & Liu, L. (2011). A joint model of recurrent events and a terminal event with a nonparametric covariate function.Statistics in Medicine, 30(22), 2683–

    2011

  13. [13]

    M., & Williams, L

    Yu, Z., Liu, L., Bravata, D. M., & Williams, L. S. (2014). Joint model of recurrent events and a terminal event with time-varying coefficients.Biometrical Journal, 56(2), 183–197. https://doi.org/10.1002/bimj.201200240

    work page doi:10.1002/bimj.201200240 2014

  14. [14]

    Ghosh, D., & Lin, D. Y. (2000). Nonparametric analysis of recurrent events and death.Biometrics, 56(2), 554–562. https://doi.org/10.1111/j.0006- 341X.2000.00554.x

    arXiv 2000

  15. [15]

    M., & Wang, M

    Li, Z., Chinchilli, V. M., & Wang, M. (2019). A Bayesian joint model of recurrent events and a terminal event.Biometrical Journal, 61(1), 187–202. https://doi.org/10.1002/bimj.201700302

    work page doi:10.1002/bimj.201700302 2019

  16. [16]

    J., & Lawless, J

    Cook, R. J., & Lawless, J. F. (2007).The Statistical Analysis of Recurrent Events. Springer. https://doi.org/10.1007/978-0-387-69810-6

    work page doi:10.1007/978-0-387-69810-6 2007

  17. [17]

    Huang, C.-Y., & Wang, M.-C. (2004). Joint modeling and estimation for recur- rent event processes and failure time data.Journal of the American Statistical Association, 99(468), 1153–1165. https://doi.org/10.1198/016214504000001116

    work page doi:10.1198/016214504000001116 2004

  18. [18]

    Huang, X., & Liu, L. (2007). A joint frailty model for survival and gap times between recurrent events.Biometrics, 63(2), 389–397. https://doi.org/10.1111/j.1541-0420.2006.00719.x

    arXiv 2007

  19. [19]

    Rondeau, V., Mathoulin-Pelissier, S., Jacqmin-Gadda, H., Brouste, V., & Soubeyran, P. (2007). Joint frailty models for recurring events and death using maximum penalized likelihood estimation: Application on cancer events. Biostatistics, 8(4), 708–721. https://doi.org/10.1093/biostatistics/kxl043

    work page doi:10.1093/biostatistics/kxl043 2007

  20. [20]

    (2000).Analysis of Multivariate Survival Data

    Hougaard, P. (2000).Analysis of Multivariate Survival Data. Springer. https://doi.org/10.1007/978-1-4612-1304-8

    doi:10.1007/978-1-4612-1304-8 2000

  21. [22]

    (2008).The Frailty Model

    Duchateau, L., & Janssen, P. (2008).The Frailty Model. Springer. (ebook DOI: 10.1007/978-0-387-72835-3) 30

    work page doi:10.1007/978-0-387-72835-3 2008

  22. [23]

    J., & Krishnan, T

    McLachlan, G. J., & Krishnan, T. (2008).The EM Algorithm and Extensions (2nd ed.). Wiley. (ebook DOI: 10.1002/9780470191613)

    doi:10.1002/9780470191613 2008

  23. [24]

    E., Bagger, H., Eliasen, P., Lyngborg, K., Videbæk, J., Cole, D

    Køber, L., Torp-Pedersen, C., Carlsen, J. E., Bagger, H., Eliasen, P., Lyngborg, K., Videbæk, J., Cole, D. S., Auclert, L., Pauly, N. C.,et al.(1995). A clinical trial of the angiotensin-converting–enzyme inhibitor trandolapril in patients with left ventricular dysfunction after myocardial infarction.New England Journal of Medicine, 333(25), 1670–1676. ht...

    work page doi:10.1056/nejm199512213332503 1995

  24. [25]

    B., & McGee, D

    Kannel, W. B., & McGee, D. L. (1979). Diabetes and cardiovas- cular disease: The Framingham study.JAMA, 241(19), 2035–2038. https://doi.org/10.1001/jama.241.19.2035

    work page doi:10.1001/jama.241.19.2035 1979

  25. [26]

    Writing Group for the Women’s Health Initiative Investigators. (2002). Risks and benefits of estrogen plus progestin in healthy postmenopausal women: Principal results from the Women’s Health Initiative randomized controlled trial.JAMA, 288(3), 321–333. https://doi.org/10.1001/jama.288.3.321

    work page doi:10.1001/jama.288.3.321 2002

  26. [27]

    The ARIC Investigators. (1989). The Atherosclerosis Risk in Communities (ARIC) study: Design and objectives.American Journal of Epidemiology, 129(4), 687–702. https://doi.org/10.1093/oxfordjournals.aje.a115184

    work page doi:10.1093/oxfordjournals.aje.a115184 1989

  27. [28]

    Ghosh, A., Chan, W., Younes, N., & Davis, B. R. (2023). A dynamic risk model for multitype recurrent events.American Journal of Epidemiology, 192(4), 621–631. https://doi.org/10.1093/aje/kwac229

    work page doi:10.1093/aje/kwac229 2023

  28. [29]

    E., & Davis, B

    Lin, L.-A., Luo, S., Chen, B. E., & Davis, B. R. (2017). Bayesian analysis of multi-type recurrent events and dependent termination with nonparametric covariate functions.Statistical Methods in Medical Research, 26(6), 2869–2884. https://doi.org/10.1177/0962280215597584

    work page doi:10.1177/0962280215597584 2017

  29. [30]

    Liu, P., & Pe˜ na, E. A. (2015). Dynamic modeling & analysis of recurrent com- peting risks and a terminal event. InProceedings of the 16th Applied Stochastic Models and Data Analysis (ASMDA 2015) International Conference, University of Piraeus, Greece, June 30–July 4, 2015, Book 1, p. 111

    2015

  30. [31]

    G., Chen, M.-H., & Sinha, D

    Ibrahim, J. G., Chen, M.-H., & Sinha, D. (2001).Bayesian Survival Analysis. Springer. https://doi.org/10.1007/978-1-4757-3447-8

    doi:10.1007/978-1-4757-3447-8 2001

  31. [32]

    Jachno, K., Heritier, S., & Wolfe, R. (2021). Impact of a non-constant baseline hazard on detection of time-dependent treatment effects: A simulation study. BMC Medical Research Methodology, 21(1), 177. https://doi.org/10.1186/s12874- 021-01365-y

    doi:10.1186/s12874- 2021

  32. [33]

    A., & Rohatgi, V

    Pe˜ na, E. A., & Rohatgi, V. K. (1993). Small sample and efficiency results for the Nelson–Aalen estimator.Journal of Statistical Planning and Inference, 37(2), 193–202. https://doi.org/10.1016/0378-3758(93)90123-4 31

    work page doi:10.1016/0378-3758(93)90123-4 1993

  33. [34]

    L., Williams, B

    Prentice, R. L., Williams, B. J., & Peterson, A. V. (1981). On the regres- sion analysis of multivariate failure time data.Biometrika, 68(2), 373–379. https://doi.org/10.1093/biomet/68.2.373

    work page doi:10.1093/biomet/68.2.373 1981

  34. [35]

    Amorim, L. D. A. F., & Cai, J. (2015). Modelling recurrent events: A tutorial for analysis in epidemiology.International Journal of Epidemiology, 44(1), 324–333. https://doi.org/10.1093/ije/dyu222

    work page doi:10.1093/ije/dyu222 2015

  35. [36]

    J., & Lim, L

    Kelly, P. J., & Lim, L. L.-Y. (2000). Survival analysis for recurrent event data: An application to childhood infectious diseases.Statistics in Medicine, 19(1), 13–33. https://doi.org/10.1002/(SICI)1097-0258(20000115)19:1¡13::AID- SIM275¿3.0.CO;2-G

    work page doi:10.1002/(sici)1097-0258(20000115)19:1 2000

  36. [37]

    Wang, M.-C., & Chang, S.-H. (1999). Nonparametric estimation of a recurrent survival function.Journal of the American Statistical Association, 94(445), 146–

    1999

  37. [38]

    M., Chin, A

    Bailey, L., Weaver, F. M., Chin, A. S., & Carbone, L. D. (2015). Estimation of a recurrent event gap-time distribution: An application to morbidity outcomes fol- lowing lower extremity fracture in veterans with spinal cord injury.Health Services and Outcomes Research Methodology, 15, 1–22. https://doi.org/10.1007/s10742- 014-0122-8

    work page doi:10.1007/s10742- 2015

  38. [39]

    Li, Y., Seo, S., & Lee, K. H. (2021). Bayesian survival analysis using gamma processes with adaptive time partition.Journal of Statistical Computation and Simulation, 91(14), 2937–2952. https://doi.org/10.1080/00949655.2021.1896749

    arXiv 2021

  39. [40]

    R., Cutler, J

    Davis, B. R., Cutler, J. A., Gordon, D. J., Furberg, C. D., Wright, J. T., Jr., Cush- man, W. C., Grimm, R. H., LaRosa, J., Whelton, P. K., Perry, H. M.,et al.(1996). Rationale and design for the Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT).American Journal of Hypertension, 9(4), 342–360. https://doi.org/10.1016/089...

    work page doi:10.1016/0895-7061(96)00037-4 1996

  40. [41]

    Jacod, J. (1975). Multivariate point processes: predictable projection, Radon– Nikodym derivatives, representation of martingales.Zeitschrift f¨ ur Wahrschein- lichkeitstheorie und verwandte Gebiete,31(3), 235–253

    1975

  41. [42]

    A., Strawderman, R

    Pe˜ na, E. A., Strawderman, R. L., & Hollander, M. (2000). A weak convergence result relevant in recurrent and renewal models. InRecent Advances in Reliability Theory: Methodology, Practice, and Inference(pp. 493–514). Springer

    2000

  42. [43]

    Kalbfleisch, J. D. (1978). Non-parametric Bayesian analysis of survival time data. Journal of the Royal Statistical Society: Series B (Methodological),40(2), 214– 221

    1978

  43. [44]

    Hougaard, P. (1995). Frailty models for survival data.Lifetime Data Analysis, 1(3), 255–273. https://doi.org/10.1007/BF00985760 32

    doi:10.1007/bf00985760 1995

  44. [45]

    Ripatti, S., & Palmgren, J. (2000). Estimation of multivariate frailty models using penalized partial likelihood.Biometrics,56(4), 1016–1022. https://doi.org/10.1111/j.0006-341X.2000.01016.x

    arXiv 2000

  45. [46]

    M., Grambsch, P

    Therneau, T. M., Grambsch, P. M., & Pankratz, V. S. (2003). Penalized survival models and frailty.Journal of Computational and Graphical Statistics,12(1), 156–175. https://doi.org/10.1198/1061860031365

    work page doi:10.1198/1061860031365 2003

  46. [47]

    ALLHAT Collaborative Research Group. (2002). Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic: The ALLHAT trial.JAMA,288(23), 2981–

    2002

  47. [48]

    T., Jr., Dunn, J

    Wright, J. T., Jr., Dunn, J. K., Cutler, J. A., Davis, B. R., Cushman, W. C., Ford, C. E., Haywood, L. J., Leenen, F. H. H., Margolis, K. L., Papademetriou, V.,et al.(2005). Outcomes in hypertensive Black and non-Black patients treated with chlorthalidone, amlodipine, and lisinopril.JAMA,293(13), 1595–1608. https://doi.org/10.1001/jama.293.13.1595

    work page doi:10.1001/jama.293.13.1595 2005

  48. [49]

    J., Muntner, P., Alonso, A., Bittencourt, M

    Benjamin, E. J., Muntner, P., Alonso, A., Bittencourt, M. S., Callaway, C. W., Carson, A. P., Chamberlain, A. M., Chang, A. R., Cheng, S., Das, S. R.,et al.(2019). Heart disease and stroke statistics—2019 update: A report from the American Heart Association.Circulation,139(10), e56–e528. https://doi.org/10.1161/CIR.0000000000000659

    doi:10.1161/cir.0000000000000659 2019

  49. [50]

    Fuster, V., & Kovacic, J. C. (2014). Acute coronary syndromes: Pathology, diagnosis, genetics, prevention, and treatment.Circulation Research,114(12), 1847–1851. https://doi.org/10.1161/CIRCRESAHA.114.302562

    work page doi:10.1161/circresaha.114.302562 2014

  50. [51]

    W., Jessup, M., Bozkurt, B., Butler, J., Casey, D

    Yancy, C. W., Jessup, M., Bozkurt, B., Butler, J., Casey, D. E., Jr., Colvin, M. M., Drazner, M. H., Filippatos, G. S., Fonarow, G. C., Givertz, M. M., et al.(2017). 2017 ACC/AHA/HFSA focused update of the 2013 ACCF/AHA guideline for the management of heart failure.Circulation,136, e137–e161. https://doi.org/10.1161/CIR.0000000000000509

    work page doi:10.1161/cir.0000000000000509 2017

  51. [52]

    Fowkes, F. G. R., Rudan, D., Rudan, I., Aboyans, V., Denenberg, J. O., McDer- mott, M. M., Norman, P. E., Sampson, U. K. A., Williams, L. J., Mensah, G. A., et al.(2013). Comparison of global estimates of prevalence and risk factors for peripheral artery disease in 2000 and 2010: A systematic review and analysis.The Lancet,382(9901), 1329–1340. https://do...

    work page doi:10.1016/s0140-6736(13)61249-0 2013

  52. [53]

    2015 ESC Guidelines for the management of acute coronary syndromes in patients presenting with- out persistent ST-segment elevation.European Heart Journal,37(3), 267–315

    Roffi, M., Patrono, C., Collet, J.-P.,et al.(2016). 2015 ESC Guidelines for the management of acute coronary syndromes in patients presenting with- out persistent ST-segment elevation.European Heart Journal,37(3), 267–315. https://doi.org/10.1093/eurheartj/ehv320 33

    work page doi:10.1093/eurheartj/ehv320 2016

  53. [54]

    D., Gornik, H

    Gerhard-Herman, M. D., Gornik, H. L., Barrett, C.,et al.(2017). 2016 AHA/ACC guideline on the management of patients with lower extremity peripheral artery disease.Journal of the American College of Cardiology,69(11), e71–e126. https://doi.org/10.1016/j.jacc.2016.11.007

    work page doi:10.1016/j.jacc.2016.11.007 2017

  54. [55]

    S., Mozaffarian, D., Roger, V

    Go, A. S., Mozaffarian, D., Roger, V. L.,et al.(2013). Heart disease and stroke statistics—2013 update: A report from the American Heart Association. Circulation,127(1), e6–e245. https://doi.org/10.1161/CIR.0b013e31828124ad

    work page doi:10.1161/cir.0b013e31828124ad 2013

  55. [56]

    R., Wilkinson, I

    Townsend, R. R., Wilkinson, I. B., Schiffrin, E. L.,et al.(2015). Recommendations for improving and standardizing vascular research on arterial stiffness: A scientific statement from the American Heart Association.Hypertension,66(3), 698–722. https://doi.org/10.1161/HYP.0000000000000033 34

    work page doi:10.1161/hyp.0000000000000033 2015

  56. [153]

    https://doi.org/10.1080/01621459.1999.10474132

    arXiv 1999

  57. [2695]

    https://doi.org/10.1002/sim.4263

    work page doi:10.1002/sim.4263

  58. [2997]

    https://doi.org/10.1001/jama.288.23.2981

    work page doi:10.1001/jama.288.23.2981

This paper was first reviewed by deepseek-v4-flash on August 4, 2026.