Bayesian Generalized Poisson Regression Modeling for Overdispersed Maternal Mortality Data
DOI:
https://doi.org/10.26877/dytq2f84Keywords:
Bayesian statistics, Maternal health, Maternal mortality rate, Overdispersion, Statistical modelingAbstract
Maternal mortality is a global health issue that reflects disparities in access to and the quality of healthcare services. This study applies the Bayesian Generalized Poisson Regression (BGPR) approach to address the problem of overdispersion in the data, which renders the standard Poisson regression model less appropriate. The Generalized Poisson model was chosen for its ability to handle overdispersion, while the Bayesian approach provides more stable parameter estimates, particularly when working with small sample sizes. The analysis results show that all independent variables have a statistically significant effect on maternal mortality. In addition, the BGPR model yields a lower Bayesian Information Criterion (BIC) value compared to the standard Poisson model, indicating better model performance. The BGPR model helps identify the key factors that truly contribute to maternal mortality, making the results useful for local governments or health institutions in setting priorities for intervention.
References
[1] World Health Organization, “Trends in maternal mortality 2000 to 2020,” World Health Organization, 2023. [Online]. Available: https://www.who.int/publications/i/item/9789240068759
[2] Bappenas, “Roadmap Tujuan Pembangunan Berkelanjutan (SDGs) Indonesia 2020–2030,” Bappenas, Jakarta, 2022. [Online]. Available: https://sdgs.bappenas.go.id/website/wp-content/uploads/2021/02/Roadmap_Bahasa-Indonesia_File-Upload.pdf
[3] Kementerian Kesehatan RI, “Profil Kesehatan Indonesia Tahun 2022,” Kementerian Kesehatan RI, Jakarta, 2023. [Online]. Available: https://kemkes.go.id/id/profil-kesehatan-indonesia-2022
[4] Ö. İŞçi̇ Güneri̇ and B. Durmuş, “Models for Overdispersion Count Data with Generalized Distribution: An Application to Parasites Intensity,” J. New Theory, no. 35, pp. 48–61, Jun. 2021, doi: 10.53570/jnt.902066.
[5] F. Famoye and K. P. Singh, “Zero-Inflated Generalized Poisson Regression Model with an Application to Domestic Violence Data,” J. Data Sci., vol. 4, no. 1, pp. 117–130, Jul. 2021, doi: 10.6339/JDS.2006.04(1).257.
[6] H. Zamani, P. Faroughi, and N. Ismail, “Bivariate generalized Poisson regression model: applications on health care data,” Empir. Econ., vol. 51, no. 4, pp. 1607–1621, Dec. 2016, doi: 10.1007/s00181-015-1051-7.
[7] A. Gelman, J. B. Carlin, H. S. Stern, D. B. Dunson, A. Vehtari, and D. B. Rubin, “Bayesian Data Analysis Third edition (with errors fixed as of 20 February 2025)”.
[8] R. Van De Schoot, S. D. Winter, O. Ryan, M. Zondervan-Zwijnenburg, and S. Depaoli, “A systematic review of Bayesian articles in psychology: The last 25 years.,” Psychol. Methods, vol. 22, no. 2, pp. 217–239, Jun. 2017, doi: 10.1037/met0000100.
[9] A. G. Assaf and M. Tsionas, “A Bayesian solution to multicollinearity through unobserved common factors,” Tour. Manag., vol. 84, p. 104277, Jun. 2021, doi: 10.1016/j.tourman.2020.104277.
[10] A. D. Chaniago and S. P. Wulandari, “Pemodelan Generalized Poisson Regression (GPR) dan Negative Binomial Regression (NBR) untuk Mengatasi Overdispersi pada Jumlah Kematian Bayi di Kabupaten Probolinggo,” J. Sains Dan Seni ITS, vol. 11, no. 6, pp. D448–D455, May 2023, doi: 10.12962/j23373520.v11i6.93240.
[11] A. A. H. Aminullah and P. Purhadi, “Pemodelan untuk Jumlah Kasus Kematian Bayi dan Ibu di Jawa Timur Menggunakan Bivariate Generalized Poisson Regression,” J. Sains Dan Seni ITS, vol. 8, no. 2, pp. D72–D78, Jan. 2020, doi: 10.12962/j23373520.v8i2.44693.
[12] N. Sari, “Solusi Overdispersi Menggunakan Generalized Poisson Regression (Studi Kasus : Penderita HIV di Provinsi Riau),” vol. 4, no. 2, 2018.
[13] L. Alkema et al., “A Bayesian approach to the global estimation of maternal mortality,” Ann. Appl. Stat., vol. 11, no. 3, Sep. 2017, doi: 10.1214/16-AOAS1014.
[14] D. Jakperik, S. Nadarajah, and M. J. Adjabui, “Bayesian Additive Modeling of Maternal Mortality In Ghana,” Dec. 08, 2021, In Review. doi: 10.21203/rs.3.rs-1089149/v1.
[15] J.-F. Angers and A. Biswas, “A Bayesian analysis of zero-in&ated generalized Poisson model”.
[16] Dinas Kesehatan Provinsi Jawa Barat, “Profil Kesehatan Provinsi Jawa Barat Tahun 2024,” Dinas Kesehatan Provinsi Jawa Barat, Bandung, 2024. Accessed: Jan. 21, 2025. [Online]. Available: https://app-diskes.jabarprov.go.id/drive/s/zAjp8Ac58gKWBp3
[17] F. Suryadi, S. Jonathan, K. Jonatan, and M. Ohyver, “Handling overdispersion in poisson regression using negative binomial regression for poverty case in NTT,” Procedia Comput. Sci., vol. 216, pp. 517–523, 2023, doi: 10.1016/j.procs.2022.12.164.
[18] Y. Tiara, M. N. Aidi, E. Erfiani, and R. Rachmawati, “Overdispersion Handling In Poisson Regression Model By Applying Negative Binomial Regression,” BAREKENG J. Ilmu Mat. Dan Terap., vol. 17, no. 1, pp. 0417–0426, Apr. 2023, doi: 10.30598/barekengvol17iss1pp0417-0426.
[19] M. A. Mukid and A. Rusgiyono, “Simulasi Stokastik Menggunakan Algoritma Gibbs Sampling”.
[20] A. Chakrabarti and J. K. Ghosh, “AIC, BIC and Recent Advances in Model Selection,” in Philosophy of Statistics, Elsevier, 2011, pp. 583–605. doi: 10.1016/B978-0-444-51862-0.50018-6.
[21] J. Hu, H. Qin, T. Yan, and Y. Zhao, “Corrected Bayesian Information Criterion for Stochastic Block Models,” J. Am. Stat. Assoc., vol. 115, no. 532, pp. 1771–1783, Oct. 2020, doi: 10.1080/01621459.2019.1637744.
[22] L. P. Rahayu, K. Sadik, and I. Indahwati, “Overdispersion study of poisson and zero-inflated poisson regression for some characteristics of the data on lamda, n, p,” Int. J. Adv. Intell. Inform., vol. 2, no. 3, p. 140, Nov. 2016, doi: 10.26555/ijain.v2i3.73.
[23] Z. A. Budiawan and Y. Widyaningsih, “Poisson Regression Models To Analyze Factors That Influence The Number Of Tuberculosis Cases In Java,” BAREKENG J. Ilmu Mat. Dan Terap., vol. 17, no. 3, pp. 1761–1772, Sep. 2023, doi: 10.30598/barekengvol17iss3pp1761-1772.
[24] A. M. H. Van Der Veen, “Bayesian methods for type A evaluation of standard uncertainty,” Metrologia, vol. 55, no. 5, pp. 670–684, Oct. 2018, doi: 10.1088/1681-7575/aad103.
[25] D. K. Sari, N. W. S. Wardhani, and S. Astutik, “Parameter Estimation of Structural Equation Modeling Using Bayesian Approach,” CAUCHY J. Mat. Murni Dan Apl., vol. 4, no. 2, pp. 86–94, May 2016, doi: 10.18860/ca.v4i2.3492.
[26] A. Joshi, T. Ollila, I. Immonen, and S. Kulathinal, “Bayesian Simultaneous Credible Intervals for Effect Measures from Multiple Markers,” Stat. Biopharm. Res., vol. 15, no. 3, pp. 638–653, Jul. 2023, doi: 10.1080/19466315.2022.2110935.
[27] A. B. Lawson, Bayesian Disease Mapping: Hierarchical Modeling in Spatial Epidemiology, Third Edition, 3rd ed. Chapman and Hall/CRC, 2018.
[28] P. C. Consul and G. C. Jain, “A Generalization of the Poisson Distribution,” Technometrics, vol. 15, no. 4, pp. 791–799, Nov. 1973, doi: 10.1080/00401706.1973.10489112.
