The Utility of Multistate Models: A Flexible Framework for Time-to-Event Data
Tóm tắt
Survival analyses are common and essential in medical research. Most readers are familiar with Kaplan–Meier curves and Cox models; however, very few are familiar with multistate models. Although multistate models were introduced in 1965, they only recently receive more attention in the medical research community. The current review introduces common terminologies and quantities that can be estimated from multistate models. Examples from published literature are used to illustrate the utility of multistate models. A figure of states and transitions is a useful depiction of a multistate model. Clinically meaningful quantities that can be estimated from a multistate model include the probability in a state at a given time, the average time in a state, and the expected number of visits to a state; all of which describe the absolute risks of an event. Relative risk can also be estimated using multistate hazard models. Multistate models provide a more general and flexible framework that extends beyond the Kaplan-Meier estimator and Cox models. Multistate models allow simultaneous analyses of multiple disease pathways to provide insights into the natural history of complex diseases. We strongly encourage the use of multistate models when analyzing time-to-event data.
Tài liệu tham khảo
Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53(282):457–81.
Cox D. Regression models and life tables. J Royal Stat Soc Series B (Methodological). 1972;34(2):187–220.
Cox DR, Miller HD. The theory of stochastic processes. Wiley publications in statistics. New York,: Wiley; 1965.
Andersen PK, Borgan O, Gill RD, Keiding N. Statistical models based on counting processes. Springer Science & Business Media; 2012. This seminal book establishes the underlying mathematical theory for non-parametric estimation and semi-parametric models of time-to-event data based on martigale process.
Cook RJ, Lawless JF. Multistate models for the analysis of life history data. Chapman and Hall/CRC; 2018.
R Development Core Team. R: a language and environment for statistical computing. Vienna, Austria: R Foundationo for Statistical Comoputing; 2010.
StataCorp. Stata Statistical Software: Release 17. College Station, TX: StataCorp LLC; 2021.
Hougaard P. Analysis of multivariate survival data. Statistics for biology and health. New York: Springer; 2000.
Allen AM, Therneau TM, Larson JJ, Coward A, Somers VK, Kamath PS. Nonalcoholic fatty liver disease incidence and impact on metabolic burden and death: a 20 year-community study. Hepatology. 2018;67(5):1726–36. https://doi.org/10.1002/hep.29546.
Aalen OO, Johansen S. Empirical transition matrix for nonhomogeneous Markov-Chains Based on censored observations. Scand J Stat. 1978;5(3):141–50.
Therneau TM. A Package for Survival Analysis in S. 2.38 ed2015.
Therneau TM, Grambsch PM. Modeling survival data : extending the Cox model. Statistics for biology and health. New York: Springer; 2000. This book introduces extensions of the Cox model to more complex data. It is a useful reference on survival analysis, filled with practical guidance, implementation in S-Plus or SAS, and contains some examples of multistate models.
Le-Rademacher JG, Peterson RA, Therneau TM, Sanford BL, Stone RM, Mandrekar SJ. Application of multi-state models in cancer clinical trials. Clin Trials. 2018;15(5):489–98. https://doi.org/10.1177/1740774518789098.
Stone RM, Mandrekar SJ, Sanford BL, Laumann K, Geyer S, Bloomfield CD, et al. Midostaurin plus chemotherapy for acute myeloid leukemia with a FLT3 mutation. N Engl J Med. 2017;377(5):454–64. https://doi.org/10.1056/NEJMoa1614359.
Jack CR Jr, Therneau TM, Lundt ES, Wiste HJ, Mielke MM, Knopman DS, et al. Long-term associations between amyloid positron emission tomography, sex, apolipoprotein E and incident dementia and mortality among individuals without dementia: hazard ratios and absolute risk. Brain Commun. 2022;4(2):fcac017. https://doi.org/10.1093/braincomms/fcac017.
Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc. 1999;94(446):496–509.
Andersen PK, Keiding N. Interpretability and importance of functionals in competing risks and multistate models. Stat Med. 2012;31(11–12):1074–88. https://doi.org/10.1002/sim.4385. This article provides an overview of survival analysis, competing risks, and illness-death models under the multistate model framework. The authors cautioned that although some functions of the transition hazards can be in these models mathematically correct, their interpretations and applicability may not be meaningful.
Beyersmann J, Allignol A, Schumacher M. Competing risks and multistate models with R. Springer Science & Business Media; 2011. This book gives a gentle introduction to the mathematics of multistate models and how it extends the usual survival analysis in the first part of the book. It draws connections between multistate models and competing risks model. The remainder of the book goes through a small set of examples, using a variety of R packages for data analysis.
Geskus RB. Data analysis with competing risks and intermediate states. Chapman and Hall/CRC; 2019. This is one of the most recent books on multistate model with an emphasis competing risks data. This book focuses more on the interpretation and application of multistate model and competing risks data using the R programming language.
Hernan MA. The hazards of hazard ratios. Epidemiology. 2010;21(1):13–5. https://doi.org/10.1097/EDE.0b013e3181c1ea43.
Aalen OO, Cook RJ, Roysland K. Does Cox analysis of a randomized survival study yield a causal treatment effect? Lifetime Data Anal. 2015;21(4):579–93. https://doi.org/10.1007/s10985-015-9335-y.