Evaluation of the natural history of disease by combining incident and prevalent cohorts: application to the Nun Study

Springer Science and Business Media LLC - Tập 29 - Trang 752-768 - 2023
Daewoo Pak1, Jing Ning2, Richard J. Kryscio3, Yu Shen2
1Division of Data Science, Yonsei University, Wonju, Korea
2Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, USA
3Department of Statistics, University of Kentucky, Lexington, USA

Tóm tắt

The Nun study is a well-known longitudinal epidemiology study of aging and dementia that recruited elderly nuns who were not yet diagnosed with dementia (i.e., incident cohort) and who had dementia prior to entry (i.e., prevalent cohort). In such a natural history of disease study, multistate modeling of the combined data from both incident and prevalent cohorts is desirable to improve the efficiency of inference. While important, the multistate modeling approaches for the combined data have been scarcely used in practice because prevalent samples do not provide the exact date of disease onset and do not represent the target population due to left-truncation. In this paper, we demonstrate how to adequately combine both incident and prevalent cohorts to examine risk factors for every possible transition in studying the natural history of dementia. We adapt a four-state nonhomogeneous Markov model to characterize all transitions between different clinical stages, including plausible reversible transitions. The estimating procedure using the combined data leads to efficiency gains for every transition compared to those from the incident cohort data only.

Tài liệu tham khảo

Contador I, Stern Y, Bermejo-Pareja F, Sanchez-Ferro A, Benito-Leon J (2017) Is educational attainment associated with increased risk of mortality in people with dementia? A population-based study. Curr Alzheimer Res 14(5):571–576 Cox D, Miller H (1965) The theory of stochastic processes. Chapman and Itall, New York Gorfine M, Keret N, Ben Arie A, Zucker D, Hsu L (2021) Marginalized frailty-based illness-death model: application to the UK-biobank survival data. J Am Stat Assoc 116(535):1155–1167 Hubbard R, Inoue L, Fann J (2008) Modeling nonhomogeneous Markov processes via time transformation. Biometrics 64(3):843–850. https://doi.org/10.1111/j.1541-0420.2007.00932.x Kalbfleisch J, Lawless J (1985) The analysis of panel data under a Markov assumption. J Am Stat Assoc 80(392):863–871. https://doi.org/10.1080/01621459.1985.10478195 Kulathinal S, Säävälä M, Auranen K, Saarela O (2020) Estimation of marriage incidence rates by combining two cross-sectional retrospective designs: Event history analysis of two dependent processes. arXiv preprint arXiv:2009.01897 Lee C, Ning J, Kryscio R, Shen Y (2019) Analysis of combined incident and prevalent cohort data under a proportional mean residual life model. Stat Med 38(12):2103–2114. https://doi.org/10.1002/sim.8098 McVittie J, Wolfson D, Stephens D, Addona V, Buckeridge D (2020) Parametric models for combined failure time data from an incident cohort study and a prevalent cohort study with follow-up. Int J Biostatist. https://doi.org/10.1515/ijb-2020-0042 Omar R, Stallard N, Whitehead J (1995) A parametric multistate model for the analysis of carcinogenicity experiments. Lifetime Data Anal 1(4):327–346. https://doi.org/10.1007/BF00985448 Pak D, Li C, Todem D, Sohn W (2017) A multistate model for correlated interval-censored life history data in caries research. J R Stat Soc Ser C 66(2):413–423. https://doi.org/10.1111/rssc.12186 Pak D, Li C, Todem D (2019) Semiparametric analysis of correlated and interval-censored event-history data. Stat Methods Med Res 28(9):2754–2767. https://doi.org/10.1177/0962280218788383 Pérez-Ocón R, Ruiz-Castro J, Gámiz-Pérez M (2001) Non-homogeneous Markov models in the analysis of survival after breast cancer. J R Stat Soc Ser C 50(1):111–124. https://doi.org/10.1111/1467-9876.00223 Saarela O, Kulathinal S, Karvanen J (2009) Joint analysis of prevalence and incidence data using conditional likelihood. Biostatistics 10(3):575–587. https://doi.org/10.1093/biostatistics/kxp013 Shen Y, Ning J, Qin J (2009) Analyzing length-biased data with semiparametric transformation and accelerated failure time models. J Am Stat Assoc 104(487):1192–1202. https://doi.org/10.1198/jasa.2009.tm08614 Stern Y, Tang M, Denaro J, Mayeux R (1995) Increased risk of mortality in Alzheimer’s disease patients with more advanced educational and occupational attainment. Ann Neurol Official J Am Neurol Assoc Child Neurol Soc 37(5):590–595. https://doi.org/10.1002/ana.410370508 Titman A (2011) Flexible nonhomogeneous Markov models for panel observed data. Biometrics 67(3):780–787. https://doi.org/10.1111/j.1541-0420.2010.01550.x Tyas S, Salazar J, Snowdon D, Desrosiers M, Riley K, Mendiondo M, Kryscio R (2007) Transitions to mild cognitive impairments, dementia, and death: findings from the nun study. Am J Epidemiol 165(11):1231–1238. https://doi.org/10.1093/aje/kwm085 Wei S, Xu L, Kryscio R (2014) Markov transition model to dementia with death as a competing event. Comput Stat Data Anal 80:78–88. https://doi.org/10.1016/j.csda.2014.06.014 Wolfson D, Best A, Addona V, Wolfson J, Gadalla S (2019) Benefits of combining prevalent and incident cohorts: an application to myotonic dystrophy. Stat Methods Med Res 28(10–11):3333–3345. https://doi.org/10.1177/0962280218804275