Model selection for maternal hypertensive disorders with symmetric hierarchical Dirichlet processes. (English) Zbl 07656978

Franzolini, Beatrice; Lijoi, Antonio; Prünster, Igor

Ann. Appl. Stat. 17, No. 1, 313-332 (2023).

Summary: Hypertensive disorders of pregnancy occur in about 10% of pregnant women around the world. Though there is evidence that hypertension impacts maternal cardiac functions, the relation between hypertension and cardiac dysfunctions is only partially understood. The study of this relationship can be framed as a joint inferential problem on multiple populations, each corresponding to a different hypertensive disorder diagnosis, that combines multivariate information provided by a collection of cardiac function indexes. A Bayesian nonparametric approach seems particularly suited for this setup, and we demonstrate it on a dataset consisting of transthoracic echocardiography results of a cohort of Indian pregnant women. We are able to perform model selection, provide density estimates of cardiac function indexes and a latent clustering of patients: these readily interpretable inferential outputs allow to single out modified cardiac functions in hypertensive patients, compared to healthy subjects, and progressively increased alterations with the severity of the disorder. The analysis is based on a Bayesian nonparametric model that relies on a novel hierarchical structure, called symmetric hierarchical Dirichlet process. This is suitably designed so that the mean parameters are identified and used for model selection across populations, a penalization for multiplicity is enforced, and the presence of unobserved relevant factors is investigated through a latent clustering of subjects. Posterior inference relies on a suitable Markov chain Monte Carlo algorithm, and the model behaviour is also showcased on simulated data.

MSC:

62Pxx

Applications of statistics

Keywords:

Bayesian nonparametrics; clustering populations; Dirichlet process; hierarchical partitions; hierarchical process; hypertensive disorders of pregnancy; model based clustering

Software:

ANOVA DDP; BayesDA

References:

[1]	AKSU, E., CUGLAN, B., TOK, A., CELIK, E., DOGANER, A., SOKMEN, A. and SOKMEN, G. (2021). Cardiac electrical and structural alterations in preeclampsia. J. Matern.-Fetal Neonatal Med. 1-10.
[2]	AMBROŽIC, J., LUCOVNIK, M., PROKŠELJ, K., TOPLIŠEK, J. and CVIJIC, M. (2020). Dynamic changes in cardiac function before and early postdelivery in women with severe preeclampsia. J. Hypertens. 38 1367-1374.
[3]	Antoniak, C. E. (1974). Mixtures of Dirichlet processes with applications to Bayesian nonparametric problems. Ann. Statist. 2 1152-1174. · Zbl 0335.60034 · doi:10.1214/aos/1176342871
[4]	BELLAMY, L., CASAS, J.-P., HINGORANI, A. D. and WILLIAMS, D. J. (2007). Pre-eclampsia and risk of cardiovascular disease and cancer in later life: Systematic review and meta-analysis. BMJ 335 974.
[5]	BERAHA, M., GUGLIELMI, A. and QUINTANA, F. A. (2021). The semi-hierarchical Dirichlet process and its application to clustering homogeneous distributions. Bayesian Anal. 16 1187-1219. · Zbl 07808151 · doi:10.1214/21-BA1278
[6]	Camerlenghi, F., Dunson, D. B., Lijoi, A., Prünster, I. and Rodríguez, A. (2019a). Latent nested nonparametric priors (with discussion). Bayesian Anal. 14 1303-1356. · Zbl 1436.62108 · doi:10.1214/19-BA1169
[7]	Camerlenghi, F., Lijoi, A., Orbanz, P. and Prünster, I. (2019b). Distribution theory for hierarchical processes. Ann. Statist. 47 67-92. · Zbl 1478.60151 · doi:10.1214/17-AOS1678
[8]	CHRISTENSEN, J. and MA, L. (2020). A Bayesian hierarchical model for related densities by using Pólya trees. J. R. Stat. Soc. Ser. B. Stat. Methodol. 82 127-153. · Zbl 1440.62240
[9]	CIFARELLI, D. M. and REGAZZINI, E. (1978). Problemi Statistici Non Parametrici in Condizioni di Scambialbilita Parziale e Impiego di Medie Associative. Quaderni Istituto Matematica Finanziaria, Torino.
[10]	CIPOLLI, W. III, HANSON, T. and MCLAIN, A. C. (2016). Bayesian nonparametric multiple testing. Comput. Statist. Data Anal. 101 64-79. · Zbl 1466.62044 · doi:10.1016/j.csda.2016.02.016
[11]	DAHL, D. B. and NEWTON, M. A. (2007). Multiple hypothesis testing by clustering treatment effects. J. Amer. Statist. Assoc. 102 517-526. · Zbl 1172.62316 · doi:10.1198/016214507000000211
[12]	DALAL, S. R. (1979a). Dirichlet invariant processes and applications to nonparametric estimation of symmetric distribution functions. Stochastic Process. Appl. 9 99-107. · Zbl 0415.60035 · doi:10.1016/0304-4149(79)90043-7
[13]	DALAL, S. R. (1979b). Nonparametric and robust Bayes estimation of location. In Optimizing Methods in Statistics (Proc. Internat. Conf., Indian Inst. Tech., Bombay, 1977) 141-166. Academic Press, New York. · Zbl 0497.62033
[14]	DAVIS, E. F., LAZDAM, M., LEWANDOWSKI, A. J., WORTON, S. A., KELLY, B., KENWORTHY, Y. et al. (2012). Cardiovascular risk factors in children and young adults born to preeclamptic pregnancies: A systematic review. Pediatrics 129 1552-1561.
[15]	De Iorio, M., Müller, P., Rosner, G. L. and MacEachern, S. N. (2004). An ANOVA model for dependent random measures. J. Amer. Statist. Assoc. 99 205-215. · Zbl 1089.62513 · doi:10.1198/016214504000000205
[16]	DEMARTELLY, V. A., DREIXLER, J., TUNG, A., MUELLER, A., HEIMBERGER, S., FAZAL, A. A., NASEEM, H., LANG, R., KRUSE, E. et al. (2021). Long-term postpartum cardiac function and its association with preeclampsia. J. Am. Heart Assoc. 10 e018526.
[17]	DENTI, F., GUINDANI, M., LEISEN, F., LIJOI, A., WADSWORTH, W. D. and VANNUCCI, M. (2021). Two-group Poisson-Dirichlet mixtures for multiple testing. Biometrics 77 622-633. · Zbl 1520.62180 · doi:10.1111/biom.13314
[18]	DIACONIS, P. and FREEDMAN, D. (1986). On inconsistent Bayes estimates of location. Ann. Statist. 14 68-87. · Zbl 0595.62023 · doi:10.1214/aos/1176349843
[19]	DO, K.-A., MÜLLER, P. and TANG, F. (2005). A Bayesian mixture model for differential gene expression. J. Roy. Statist. Soc. Ser. C 54 627-644. · Zbl 1490.62353 · doi:10.1111/j.1467-9876.2005.05593.x
[20]	DOLEA, C. and ABOUZAHR, C. (2003). Global burden of hypertensive disorders of pregnancy in the year 2000 Technical report, GBD 2000 Working Paper, World Health Organization, Geneva.
[21]	DOSS, H. (1984). Bayesian estimation in the symmetric location problem. Z. Wahrsch. Verw. Gebiete 68 127-147. · Zbl 0531.62030 · doi:10.1007/BF00531774
[22]	Escobar, M. D. and West, M. (1995). Bayesian density estimation and inference using mixtures. J. Amer. Statist. Assoc. 90 577-588. · Zbl 0826.62021 · doi:10.1080/01621459.1995.10476550
[23]	Ferguson, T. S. (1973). A Bayesian analysis of some nonparametric problems. Ann. Statist. 1 209-230. · Zbl 0255.62037
[24]	FRANZOLINI, B., LIJOI, A. and PRÜNSTER, I. (2023). Supplement to “Model Selection for Maternal Hypertensive Disorders with Symmetric Hierarchical Dirichlet Processes.” https://doi.org/10.1214/22-AOAS1628SUPPA, https://doi.org/10.1214/22-AOAS1628SUPPB
[25]	GARCIA-GONZALEZ, C., GEORGIOPOULOS, G., AZIM, S. A., MACAYA, F., KAMETAS, N., NIHOYANNOPOULOS, P., NICOLAIDES, K. H. and CHARAKIDA, M. (2020). Maternal cardiac assessment at 35 to 37 weeks improves prediction of development of preeclampsia. Hypertens. 76 514-522.
[26]	GELMAN, A., CARLIN, J. B., STERN, H. S., DUNSON, D. B., VEHTARI, A. and RUBIN, D. B. (2013). Bayesian Data Analysis. CRC Press, Boca Raton, FL.
[27]	GHOSAL, S., GHOSH, J. K. and RAMAMOORTHI, R. V. (1999). Consistent semiparametric Bayesian inference about a location parameter. J. Statist. Plann. Inference 77 181-193. · Zbl 1054.62528 · doi:10.1016/S0378-3758(98)00192-X
[28]	GOPALAN, R. and BERRY, D. A. (1998). Bayesian multiple comparisons using Dirichlet process priors. J. Amer. Statist. Assoc. 93 1130-1139. · Zbl 1063.62530 · doi:10.2307/2669856
[29]	GUINDANI, M., MÜLLER, P. and ZHANG, S. (2009). A Bayesian discovery procedure. J. R. Stat. Soc. Ser. B. Stat. Methodol. 71 905-925. · Zbl 1411.62224 · doi:10.1111/j.1467-9868.2009.00714.x
[30]	Gutiérrez, L., Barrientos, A. F., González, J. and Taylor-Rodríguez, D. (2019). A Bayesian nonparametric multiple testing procedure for comparing several treatments against a control. Bayesian Anal. 14 649-675. · Zbl 1421.62046 · doi:10.1214/18-BA1122
[31]	HALL, M. E., GEORGE, E. M. and GRANGER, J. P. (2018). The heart during pregnancy. Rev. Esp. Orientac. 64 1045-1050.
[32]	IGBERASE, G. and EBEIGBE, P. (2006). Eclampsia: Ten-years of experience in a rural tertiary hospital in the Niger Delta, Nigeria. J. Obstet. Gynaecol. 26 414-417.
[33]	IGLESIAS, P. L., ORELLANA, Y. and QUINTANA, F. A. (2009). Nonparametric Bayesian modelling using skewed Dirichlet processes. J. Statist. Plann. Inference 139 1203-1214. · Zbl 1156.62041 · doi:10.1016/j.jspi.2008.07.009
[34]	Lijoi, A., Nipoti, B. and Prünster, I. (2014). Bayesian inference with dependent normalized completely random measures. Bernoulli 20 1260-1291. · Zbl 1309.60048 · doi:10.3150/13-BEJ521
[35]	LIJOI, A., PRÜNSTER, I. and REBAUDO, G. (2022). Flexible clustering via hidden hierarchical Dirichlet priors. Scand. J. Stat.. · Zbl 07677035 · doi:10.1111/sjos.12578
[36]	MACEACHERN, S. N. (2000). Dependent Dirichlet processes Technical Report, Department of Statistics, The Ohio State Univ.
[37]	MALIK, A., JEE, B. and GUPTA, S. K. (2019). Preeclampsia: Disease biology and burden, its management strategies with reference to India. Pregnancy Hypertens. 15 23-31.
[38]	MARTIN, R. and TOKDAR, S. T. (2012). A nonparametric empirical Bayes framework for large-scale multiple testing. Biostatistics 13 427-439. · Zbl 1244.62066
[39]	MCCLURE, E. M., SALEEM, S., PASHA, O. and GOLDENBERG, R. L. (2009). Stillbirth in developing countries: A review of causes, risk factors and prevention strategies. J. Matern.-Fetal Neonatal Med. 22 183-190.
[40]	MOSER, S., RODRÍGUEZ, A. and LOFLAND, C. L. (2021). Multiple ideal points: Revealed preferences in different domains. Polit. Anal. 29 139-166.
[41]	MULIERE, P. and PETRONE, S. (1993). A Bayesian predictive approach to sequential search for an optimal dose: Parametric and nonparametric models. J. Italian Stat. Soc. 2 349-364. · Zbl 1446.62283
[42]	Neal, R. M. (2000). Markov chain sampling methods for Dirichlet process mixture models. J. Comput. Graph. Statist. 9 249-265. · doi:10.2307/1390653
[43]	PEDERSEN, S. S., VON KÄNEL, R., TULLY, P. J. and DENOLLET, J. (2017). Psychosocial perspectives in cardiovascular disease. Eur. J. Prev. Cardiol. 24 108-115.
[44]	Rodríguez, A., Dunson, D. B. and Gelfand, A. E. (2008). The nested Dirichlet process. J. Amer. Statist. Assoc. 103 1131-1144. · Zbl 1205.62062 · doi:10.1198/016214508000000553
[45]	SCOTT, J. G. and BERGER, J. O. (2006). An exploration of aspects of Bayesian multiple testing. J. Statist. Plann. Inference 136 2144-2162. · Zbl 1087.62039 · doi:10.1016/j.jspi.2005.08.031
[46]	Scott, J. G. and Berger, J. O. (2010). Bayes and empirical-Bayes multiplicity adjustment in the variable-selection problem. Ann. Statist. 38 2587-2619. · Zbl 1200.62020 · doi:10.1214/10-AOS792
[47]	SHAH, A., FAWOLE, B., M’IMUNYA, J. M., AMOKRANE, F., NAFIOU, I., WOLOMBY, J.-J. et al. (2009). Cesarean delivery outcomes from the WHO global survey on maternal and perinatal health in Africa. Int. J. Gynecol. Obstet. 107 191-197.
[48]	SORIANO, J. and MA, L. (2017). Probabilistic multi-resolution scanning for two-sample differences. J. R. Stat. Soc. Ser. B. Stat. Methodol. 79 547-572. · Zbl 1414.62149 · doi:10.1111/rssb.12180
[49]	TATAPUDI, R. and PASUMARTHY, L. R. (2017a). Data for: Maternal cardiac function in gestational hypertension, mild and severe preeclampsia and normal pregnancy: A comparative study. Available at https://data.mendeley.com/datasets/d72zr4xggx/1. https://doi.org/10.17632/d72zr4xggx.1 Licensed under a Creative Commons Attribution 4.0 International licence.
[50]	TATAPUDI, R. and PASUMARTHY, L. R. (2017b). Maternal cardiac function in gestational hypertension, mild and severe preeclampsia and normal pregnancy: A comparative study. Pregnancy Hypertens. 10 238-241.
[51]	Teh, Y. W., Jordan, M. I., Beal, M. J. and Blei, D. M. (2006). Hierarchical Dirichlet processes. J. Amer. Statist. Assoc. 101 1566-1581. · Zbl 1171.62349 · doi:10.1198/016214506000000302
[52]	TIMOKHINA, E., KUZMINA, T., STRIZHAKOV, A., PITSKHELAURI, E., IGNATKO, I. and BELOUSOVA, V. (2019). Maternal cardiac function after normal delivery, preeclampsia, and eclampsia: A prospective study. J. Pregnancy 2019 2090-2727.
[53]	ZUANETTI, D. A., MÜLLER, P., ZHU, Y., YANG, S. and JI, Y. (2018). Clustering distributions with the marginalized nested Dirichlet process. Biometrics 74 584-594 · Zbl 1414.62489 · doi:10.1111/biom.12778