Asian Pacific Journal of Cancer Prevention

Asian Pacific Journal of Cancer Prevention (아시아태평양암예방학회)
권호 표지
DOI QR코드

DOI QR Code

Assessing Markov and Time Homogeneity Assumptions in Multi-state Models: Application in Patients with Gastric Cancer Undergoing Surgery in the Iran Cancer Institute

  • Zare, Ali (Department of Epidemiology and Biostatistics, Tehran University of Medical Sciences) ;
  • Mahmoodi, Mahmood (Department of Epidemiology and Biostatistics, Tehran University of Medical Sciences) ;
  • Mohammad, Kazem (Department of Epidemiology and Biostatistics, Tehran University of Medical Sciences) ;
  • Zeraati, Hojjat (Department of Epidemiology and Biostatistics, Tehran University of Medical Sciences) ;
  • Hosseini, Mostafa (Department of Epidemiology and Biostatistics, Tehran University of Medical Sciences) ;
  • Naieni, Kourosh Holakouie (Department of Epidemiology and Biostatistics, Tehran University of Medical Sciences)
  • Published : 2014.01.15
Download PDF
⟨ Previous Next ⟩

Abstract

Background: Multi-state models are appropriate for cancer studies such as gastrectomy which have high mortality statistics. These models can be used to better describe the natural disease process. But reaching that goal requires making assumptions like Markov and homogeneity with time. The present study aims to investigate these hypotheses. Materials and Methods: Data from 330 patients with gastric cancer undergoing surgery at Iran Cancer Institute from 1995 to 1999 were analyzed. To assess Markov assumption and time homogeneity in modeling transition rates among states of multi-state model, Cox-Snell residuals, Akaikie information criteria and Schoenfeld residuals were used, respectively. Results: The assessment of Markov assumption based on Cox-Snell residuals and Akaikie information criterion showed that Markov assumption was not held just for transition rate of relapse (state 1 ${\rightarrow}$ state 2) and for other transition rates - death hazard without relapse (state 1 ${\rightarrow}$ state 3) and death hazard with relapse (state 2 ${\rightarrow}$ state 3) - this assumption could also be made. Moreover, the assessment of time homogeneity assumption based on Schoenfeld residuals revealed that this assumption - regarding the general test and each of the variables in the model- was held just for relapse (state 1 ${\rightarrow}$ state 2) and death hazard with a relapse (state 2 ${\rightarrow}$ state 3). Conclusions: Most researchers take account of assumptions such as Markov and time homogeneity in modeling transition rates. These assumptions can make the multi-state model simpler but if these assumptions are not made, they will lead to incorrect inferences and improper fitting.

Keywords

References

  1. Akaike H (1980). Likelihood and the Bayes procedure. Trabajos de Estadistica y de Investigacion Operativa, 31, 143-66. https://doi.org/10.1007/BF02888350
  2. Akaike H (1998). A new look at the statistical model identification. In Selected Papers of Hirotugu Akaike, pp. 215-222.
  3. American Joint Committee on Cancer (2002). AJCC Cancel Staging Manual. 6th ed. New York. NY: Springer. pp 99-106.
  4. Anderson DR (2008). Model based inference in the life sciences: a primer on evidence.
  5. Association JGC (2011). Japanese gastric cancer treatment guidelines 2010 (ver. 3). Gastric Cancer, 14, 113-23. https://doi.org/10.1007/s10120-011-0042-4
  6. Chen P-L, Bernard EJ, Sen PK (1999). A Markov chain model used in analyzing disease history applied to a stroke study. J App Stat, 26, 413-22. https://doi.org/10.1080/02664769922304
  7. Chen P-L, Tien H-C (2004). Semi-Markov models for multistate data analysis with periodic observations. Communications in Statistics: Theory and Methods, 33, 475-86. https://doi.org/10.1081/STA-120028679
  8. Collett D (2003) Modelling survival data in medical research: CRC press.
  9. De Stavola BL (1988). Testing departures from time homogeneity in multistate Markov processes. App Stat, 33, 242-50.
  10. Escobar LA, Meeker Jr WQ (1992). Assessing influence in regression analysis with censored data. Biometrics, 48, 507-28. https://doi.org/10.2307/2532306
  11. Faddy M (1976). A note on the general time-dependent stochastic compartmental model. Biometrics, 443-8.
  12. Foucher Y, Giral M, Soulillou J, Daures J (2010). A flexible semi-Markov model for interval-censored data and goodness-of-fit testing. Stat Methods Med Res, 19, 127-45. https://doi.org/10.1177/0962280208093889
  13. Foulkes A, De Gruttola V (2003). Characterizing the progression of viral mutations over time. J Am Stat Assoc, 98, 859-67. https://doi.org/10.1198/016214503000000792
  14. Gentleman R, Lawless J, Lindsey J, Yan P (1994). Multi-state Markov models for analysing incomplete disease history data with illustrations for hiv disease. Stat Med, 13, 805-21. https://doi.org/10.1002/sim.4780130803
  15. Grambsch PM, Therneau TM (1994). Proportional hazards tests and diagnostics based on weighted residuals. Biometrika, 81, 515-526. https://doi.org/10.1093/biomet/81.3.515
  16. Healy B, Degruttola V (2007). Hidden Markov models for settings with interval-censored transition times and uncertain time origin: application to HIV genetic analyses. Biostatistics, 8, 438-52. https://doi.org/10.1093/biostatistics/kxl021
  17. Hosmer Jr DW, Lemeshow S, May S (2011). Applied survival analysis: regression modeling of time to event data: Wiley-Interscience.
  18. Hsieh HJ, Chen THH, Chang SH (2002). Assessing chronic disease progression using non-homogeneous exponential regression Markov models: an illustration using a selective breast cancer screening in Taiwan. Stat Med, 21, 3369-82. https://doi.org/10.1002/sim.1277
  19. Jackson CH (2011). Multi-state models for panel data: the msm package for R. J Stat Software, 38, 1-29.
  20. Kalbfleisch J, Lawless JF (1985). The analysis of panel data under a Markov assumption. J Am Stat Assoc, 80, 863-71. https://doi.org/10.1080/01621459.1985.10478195
  21. Kalbfleisch JD, Prentice RL (2011). The statistical analysis of failure time data: John Wiley & Sons.
  22. Kang M & Lagakos SW (2007). Statistical methods for panel data from a semi-Markov process, with application to HPV. Biostatistics 8, 252-264. https://doi.org/10.1093/biostatistics/kxl006
  23. Kay R (1986). A Markov model for analysing cancer markers and disease states in survival studies. Biometrics, 855-865.
  24. Klein JP, Moeschberger ML (2003). Survival analysis: techniques for censored and truncated data: Springer.
  25. Klein JP, Zhang MJ (2005). Survival analysis, software: Wiley Online Library.
  26. Mathieu E, Loup P, Dellamonica P, Daures J (2005). Markov Modelling of Immunological and Virological States in HIV-1 Infected Patients. Biometrical J, 47, 834-46. https://doi.org/10.1002/bimj.200410164
  27. Meira-Machado L, de Una-Alvarez J, Cadarso-Suarez C, Andersen PK (2009). Multi-state models for the analysis of time-to-event data. Stat Methods Med Res, 18, 195-222.
  28. Mohagheghi MA (2004). Annual report of Tehran cancer registery 1999: Cancer institute publication.
  29. Mohagheghi MA, Mosavi-Jarrahi A, Malekzadeh R, Parkin M (2009). Cancer incidence in tehran metropolis: the first report from the tehran population-based cancer registry. Archives Iranian Med, 12, 15-23.
  30. Mousavi SM, Gouya MM, Ramazani R, et al (2009). Cancer incidence and mortality in Iran. Ann Oncol, 20, 556-63.
  31. Ocan-Riola R (2005). Non-homogeneous Markov processes for biomedical data analysis. Biometrical J, l47, 369-76.
  32. Omar RZ, Stallard N, Whitehead J (1995). A parametric multistate model for the analysis of carcinogenicity experiments. Lifetime Data Analysis, 1, 327-46. https://doi.org/10.1007/BF00985448
  33. Parkin DM, Bray F, Ferlay J, Pisani P (2005). Global cancer statistics, 2002. CA: Cancer J Clin, 55, 74-108. https://doi.org/10.3322/canjclin.55.2.74
  34. Perez-Ocon R, Ruiz-Castro JE, Gamiz-Perez ML (2000). Markov models with lognormal transition rates in the analysis of survival times. Test, 9, 353-70. https://doi.org/10.1007/BF02595740
  35. Perez-Ocon R, Ruiz-Castro JE, Gamiz-Perez ML (2001). Nonhomogeneous Markov models in the analysis of survival after breast cancer. J Royal Stat Society, 50, 111-24. https://doi.org/10.1111/1467-9876.00223
  36. Peto R, Peto J (1972). Asymptotically efficient rank invariant test procedures. J Royal Stat Soc, 185-207.
  37. Razavi SE, Aaghajani H, Haghazali M, et al (2009) The most common cancers in Iranian women. Iranian J Public Health, 38, 109-12.
  38. Ruiz-Castro JE & Perez-Ocon R (2004) A semi-Markov model in biomedical studies.
  39. Sadighi S, Mohagheghi M, Haddad P, et al (2008). Life expectancy with perioperative chemotherapy and chemoradiotherapy for locally advanced gastric adenocarcinoma. TUMJ, 66, 664-9.
  40. Sadighi S, Raafat J, Mohagheghi M, Meemary F (2005). Gastric carcinoma: 5 year experience of a single institute. Asian Pac J Cancer Prev, 6, 195-6.
  41. Saint-Pierre P, Combescure C, Daures J, Godard P (2003). The analysis of asthma control under a Markov assumption with use of covariates. Stat Med, 22, 3755-70. https://doi.org/10.1002/sim.1680
  42. Samadi F, Babaei M, Yazdanbod A, et al (2007). Survival rate of gastric and esophageal cancers in Ardabil province, North-West of Iran. Arch Iran Med, 10, 32-7.
  43. Sharples L, Taylor G, Faddy M (2001). A piecewisehomogeneous Markov chain process of lung transplantation. J Epidemiol and Biostatistics, 6, 349-55. https://doi.org/10.1080/13595220152601828
  44. Titman A (2008). Model diagnostics in multi-state models of biological systems, PhD Thesis, University of Cambridge, UK.
  45. Titman AC (2011). Flexible nonhomogeneous Markov models for panel observed data. Biometrics, 67, 780-7. https://doi.org/10.1111/j.1541-0420.2010.01550.x
  46. Titman AC (2012). Estimating parametric semi-Markov models from panel data using phase-type approximations. Statistics and Computing, 1-10.
  47. Titman AC, Sharples LD (2010a). Model diagnostics for multistate models. Stat Methods Med Res, 19, 621-51. https://doi.org/10.1177/0962280209105541
  48. Titman AC, Sharples LD (2010b). Semi-Markov Models with phase-type sojourn distributions. Biometrics, 66, 742-52. https://doi.org/10.1111/j.1541-0420.2009.01339.x
  49. Weissfeld LA, Schneider H (1990). Influence diganostics for the Weibull model fit to censored data. Stat & Probability Lett, 9, 67-73. https://doi.org/10.1016/0167-7152(90)90097-Q

Cited by

  1. Assessing Misdiagnosis of Relapse in Patients with Gastric Cancer in Iran Cancer Institute Based on a Hidden Markov Multi-state Model vol.15, pp.9, 2014, https://doi.org/10.7314/APJCP.2014.15.9.4109