Skip to main content
SpringerLink
Log in

Regression Analysis of Mixed Panel-Count Data with Application to Cancer Studies

  • Published:
Statistics in Biosciences Aims and scope Submit manuscript

Abstract

Both panel-count data and panel-binary data are common data types in recurrent event studies. Because of inconsistent questionnaires or missing data during the follow-ups, mixed data types need to be addressed frequently. A recently proposed semiparametric approach uses a proportional means model to facilitate regression analyses of mixed panel-count and panel-binary data. This method can use all available information regardless of the record type and provide unbiased estimates. However, the large number of nuisance parameters in the nonparametric baseline hazard function makes the estimating procedure very complicated and time-consuming. We approximated the baseline hazard function to simplify the estimating procedure. Simulation studies showed that our method performed similarly to that of the previous semiparametric likelihood-based method, but with much faster speed. Approximating the baseline hazard not only reduced the computational burden but also made it possible to implement the estimating procedure in a standard software, such as SAS.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1

Similar content being viewed by others

References

  1. Cheng G, Zhang Y, Lu L (2011) Efficient algorithms for computing the non and semi-parametric maximum likelihood estimates with panel count data. J Nonparametric Stat 23(2):567–579. https://doi.org/10.1080/10485252.2010.548521

    Article  MathSciNet  MATH  Google Scholar 

  2. Diggle PJ, Heagerty P, Liang K-Y, Zeger SL (2002) Analysis of longitudinal data. Oxford statistical science series, OUP, Oxford

    MATH  Google Scholar 

  3. Feng S, Wolfe RA, Port FK (2005) Frailty survival model analysis of the national deceased donor kidney transplant dataset using Poisson variance structures. J Am Stat Assoc 100(471):728–735. https://doi.org/10.1198/016214505000000123

    Article  MathSciNet  MATH  Google Scholar 

  4. Fitzmaurice GM, Molenberghs G, Lipsitz SR (1995) Regression models for longitudinal binary responses with informative drop-outs. J R Stat Soc Ser B (Methodological) 57(4):691–704

    MathSciNet  MATH  Google Scholar 

  5. Fitzmaurice G, Davidian M, Verbeke G, Molenberghs G (2008) Longitudinal data analysis. Chapman & Hall/CRC handbooks of modern statistical methods, Chapman and Hall/CRC, Boca Raton

    Book  Google Scholar 

  6. Gray RJ (1992) Flexible methods for analyzing survival data using splines, with applications to breast cancer prognosis. J Am Stat Assoc 87(420):942–951

    Article  Google Scholar 

  7. Jongbloed G (1998) The iterative convex minorant algorithm for nonparametric estimation. J Comput Graph Stat 7(3):310–321. https://doi.org/10.1080/10618600.1998.10474778

    Article  MathSciNet  Google Scholar 

  8. Liang K-Y, Zeger SL (1986) Longitudinal data analysis using generalized linear models. Biometrika 73(1):13–22. https://doi.org/10.1093/biomet/73.1.13

    Article  MathSciNet  MATH  Google Scholar 

  9. Liu L, Huang X (2008) The use of gaussian quadrature for estimation in frailty proportional hazards models. Stat Med 27(14):2665–2683. https://doi.org/10.1002/sim.3077

    Article  MathSciNet  Google Scholar 

  10. Liu L, Wolfe RA, Huang X (2004) Shared frailty models for recurrent events and a terminal event. Biometrics 60(3):747–756. https://doi.org/10.1111/j.0006-341X.2004.00225.x

    Article  MathSciNet  MATH  Google Scholar 

  11. Liu L, Huang X, O’Quigley J (2008) Analysis of longitudinal data in the presence of informative observational times and a dependent terminal event, with application to medical cost data. Biometrics 64(3):950–958. https://doi.org/10.1111/j.1541-0420.2007.00954.x

    Article  MathSciNet  MATH  Google Scholar 

  12. Liu L, Huang X, Yaroshinsky A, Cormier JN (2016) Joint frailty models for zero-inflated recurrent events in the presence of a terminal event. Biometrics 72(1):204–214. https://doi.org/10.1111/biom.12376

    Article  MathSciNet  MATH  Google Scholar 

  13. Royston P, Parmar MKB (2002) Flexible parametric proportional-hazards and proportional-odds models for censored survival data, with application to prognostic modelling and estimation of treatment effects. Stat Med 21(15):2175–2197. https://doi.org/10.1002/sim.1203

    Article  Google Scholar 

  14. Sun J, Zhao X (2013) Statistical analysis of panel count data. Springer, New York. https://doi.org/10.1007/978-1-4614-8715-9

  15. Wellner JA, Zhang Y (2007) Two likelihood-based semiparametric estimation methods for panel count data with covariates. Ann Stat 35(5):2106–2142

    Article  MathSciNet  Google Scholar 

  16. Yu G, Zhu L, Li Y, Sun J, Robison LL (2017) Regression analysis of mixed panel count data with dependent terminal events. Stat Med 36(10):1669–1680. https://doi.org/10.1002/sim.7217

    Article  MathSciNet  Google Scholar 

  17. Yu G, Zhu L, Sun J, Robison LL (2018) Regression analysis of incomplete data from event history studies with the proportional rates model. Stat Interface 11(1):91–97. https://doi.org/10.4310/SII.2018.v11.n1.a8

    Article  MathSciNet  MATH  Google Scholar 

  18. Yu G, Li Y, Zhu L, Zhao H, Sun J, Robison LL (2019) An additive-multiplicative mean model for panel count data with dependent observation and dropout processes. Scand J Stat 46(2):414–431

    Article  MathSciNet  Google Scholar 

  19. Zhu L, Tong X, Zhao H, Sun J, Deo KS, Leisenring W, Robison LL (2013) Statistical analysis of mixed recurrent event data with application to cancer survivor study. Stat Med 32(11):1954–1963. https://doi.org/10.1002/sim.5674

    Article  MathSciNet  Google Scholar 

  20. Zhu L, Tong X, Sun J, Manhua C, Deo KS, Leisenring W, Robison LL (2014) Regression analysis of mixed recurrent-event and panel-count data. Biostatistics 15(3):555–568. https://doi.org/10.1093/biostatistics/kxu009

    Article  Google Scholar 

  21. Zhu L, Zhao H, Sun J, Leisenring W, Robison LL (2015) Regression analysis of mixed recurrent-event and panel-count data with additive rate models. Biometrics 71(1):71–79. https://doi.org/10.1111/biom.12247

    Article  MathSciNet  MATH  Google Scholar 

  22. Zhu L, Zhang Y, Li Y, Sun J, Robison LL (2018) A semiparametric likelihood-based method for regression analysis of mixed panel-count data. Biometrics 74(2):488–497. https://doi.org/10.1111/biom.12774

    Article  MathSciNet  MATH  Google Scholar 

Download references

Acknowledgement

The work was partly supported by NIH [R03CA219450] to Zhu.

Author information

Authors and Affiliations

  1. Department of Biostatistics, St. Jude Children’s Research Hospital, Memphis, TN, 38105, USA

    Yimei Li

  2. Division of Clinical and Translational Sciences, Department of Internal Medicine, The University of Texas Health Science Center at Houston, Houston, TX, 77030, USA

    Liang Zhu

  3. Division of Biostatistics, Washington University in St. Louis, St. Louis, MO, 63310, USA

    Lei Liu

  4. Department of Epidemiology and Cancer Control, St. Jude Children’s Research Hospital, Memphis, TN, 38105, USA

    Leslie L. Robison

Authors
  1. Yimei Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Liang Zhu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Lei Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Leslie L. Robison
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Liang Zhu.

Appendix

Appendix

The sample SAS codes for the piecewise spline and cubic spline methods are given below. The sample SAS code for setups with random effects is available upon request from the corresponding author. The SAS dataset “one” is prepared so that each patient has multiple data lines, with each data line corresponding to a panel-count or binary record. Three covariates are denoted by “covar1”, “covar2”, and “covar3” in the recurrent event models. The cumulative baseline hazard is denoted by “cum_base_haz_s” (starting point) and “cum_base_haz_e” (end point) to calculate the increase of the baseline hazard in a duration. The parameters for the piecewise constant baseline hazards are denoted by r01, r02, ..., and r05, and the parameters for the cubic spline hazards are denoted by c00, c01, and c02.

figure a

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Li, Y., Zhu, L., Liu, L. et al. Regression Analysis of Mixed Panel-Count Data with Application to Cancer Studies. Stat Biosci 13, 178–195 (2021). https://doi.org/10.1007/s12561-020-09291-2

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12561-020-09291-2

Keywords

Search

Navigation