Application of Multi-State Markov Models to Alzheimer's Disease Data

Qianwei Lin, Huiling Deng


Objective: To explore the impact of the probability of metastasis between stages, mean residence time and APOE4 allele count on disease progression during the progression of Alzheimer's disease. Methods: 3191 patients initially diagnosed with Alzheimer's disease in the Uniform Data Set UDS maintained by the National Alzheimer's Collaborative Center (NACC) were selected, and a multi-state Markov model with death as the outcome was developed based on the MMSE standard cut-off point delineation criteria with three stages of Alzheimer's disease: mild, moderate and severe. Results: The metastatic intensity and probability of metastatic death gradually increased as the disease progressed through mild, moderate and severe stages; the mean length of stay in mild, moderate and severe Alzheimer's disease patients was 2.905, 1.875 and 1.819 years, respectively; with one APOE4 allele [HR 1.176, 95% CI (1.031,1.340)] and [HR 1.426, 95%CI(1.202,1.693)] were risk factors for mild to moderate transfer. Conclusions: Alzheimer's disease has a long course with multi-stage progression, risk factors affecting disease progression are more complex, the APOE4 allele is a risk factor for Alzheimer's disease, and having 2 APOE4 alleles is a greater risk than 1 APOE4 allele.


Multi-State Markov Model; Alzheimer's Disease; APOE4 Allele; Disease Progression; Probability of Metastasis

Full Text:



Cummings J, Lee G, Zhong K, et al. Alzheimer's disease drug development pipeline: 2021[J]. Alzheimer's & Dementia: Translational Research & Clinical Interventions, 2021, 7(1): e12179.

Scheltens P, De Strooper B, Kivipelto M, et al. Alzheimer's disease[J]. The Lancet, 2021, 397(10284): 1577–1590.

Zhao J, Fu Y, Yamazaki Y, et al. APOE4 exacerbates synapse loss and neurodegeneration in Alzheimer’s disease patient iPSC-derived cerebral organoids[J]. Nature communications, 2020, 11(1): 1-14.

Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”: a practical method for grading the cognitive state of patients for the clinician[J]. Journal of psychiatric research, 1975, 12(3): 189-198.

Hughes CP, Berg L, Danziger W, et al. A new clinical scale for the staging of dementia[J]. The British journal of psychiatry, 1982, 140(6): 566-572.

Jackson C. Multi-state modeling with R:the msm package (version 1.0). 2011.

Andersen, PK, Wandall, E.N.S. & Pohar Perme, M. Inference for transition probabilities in non-Markov multi-state models. Lifetime Data Analysis (2022) 28:585–604.

Zheng X, Xiong J, Zhang Y, et al. Multistate Markov model application for blood pressure transition among the Chinese elderly population: a quantitative longitudinal study[J]. BMJ open, 2022, 12(7): e059805.

Taguchi A, Hara K, Tomio J, et al. Multistate markov model to predict the prognosis of high-risk human papillomavirus-related cervical lesions[J]. Cancers, 2020, 12(2): 270.

Dessie ZG, Zewotir T, Mwambi H, et al. Modelling of viral load dynamics and CD4 cell count progression in an antiretroviral naive cohort: using a joint linear mixed and multistate Markov model[J]. BMC infectious diseases, 2020, 20(1): 1-14.

Farahani MV, Dizaji PA, Rashidi H, et al. Application of Multi-State Model in Analyzing of Breast Cancer Data[J]. Journal of Research in Health Sciences, 2019, 19(4): e00465.

Smith EMD, Eleuteri A, Goilav B, et al. A Markov Multi-State model of lupus nephritis urine biomarker panel dynamics in children: Predicting changes in disease activity[J]. Clinical Immunology, 2019, 198: 71-78.

Wattmo C, Wallin AK, Minthon L. Progression of mild Alzheimer’s disease: knowledge and prediction models required for future treatment strategies[J]. Alzheimer's research & therapy, 2013, 5(5): 1-15.



  • There are currently no refbacks.

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.