Modelling zero-truncated overdispersed antenatal health care count data of women in Bangladesh

Abstract

Overdispersion in count data analysis is very common in many practical fields of health sciences. Ignorance of the presence of overdispersion in such data analysis may cause misleading inferences and thus lead to incorrect interpretations of the results. Researchers should account for the consequences of overdispersion and need to select the correct choice of models for the analysis of such data. In this paper, Generalized Linear Models (GLMs) are applied in modelling and analysis of antenatal care (ANC) count data extracted from the Bangladesh Demographic and Health Survey (BDHS) 2014. Pearson chi-square and different score tests are used to investigate the effect of overdispersion in the analysis. Overdispersion is found to be significant in the antenatal health care count data and so appropriate modelling is used to produce valid inferences for the regression parameters. The zero-truncated negative binomial regression (0-NBR) is found to be the best choice for analysing such data while excluding zero counts. Study findings reveal that place of residence, order of birth, exposure to mass media, wealth index and education of mother have significant impacts on the ANC status of women during pregnancy in Bangladesh.

Generalized Poisson Regression Type-II at Jambi City Health Office

“…Poisson regression assumes the random component is generalized Poisson distribution [1,[7][8][9][10][11][12].…”

“…In general, regression analysis has a response variable which is a continuous variable that follows the normal distribution. But there is also a regression analysis that has a response variable not a continuous variable but a discrete variable one of which is Poisson regression analysis [1][2][3][4].…”

“…The next assumption is equidispersion, a condition in which the mean (mean) and variance of the response variable are the same. However, not all data meet the assumption of equidispersion so that if the data have conditions where the mean and variance are not the same then Poisson regression analysis should not be used [1][2][3][4][5][6].…”

A Novel Bayesian Outlier Score Based on the Negative Binomial Distribution for Detecting Aberrantly Expressed Genes in RNA-Seq Gene Expression Count Data

“…F INDING outliers in data following the Negative Binomial distribution (NBD) is an active research topic, as NBD can be used to model non-negative integer data of various types, such as gene expression counts produced by RNA sequencing (RNA-Seq) [1], medical [2] and social data [3], etc.…”

Analyzing Overdispersed Antenatal Care Count Data in Bangladesh: Mixed Poisson Regression with Individual-Level Random Effects

“…Women education was also found to be the significant factor for ensuring adequate ANC visits, antenatal morbidity, and also to reducing child and maternal deaths (Becker, Peters, Gray, Gultiano, and Black 1993;Mohammad et al 2017;Chakraborty, Islam, Chowdhury, Bari, and Akhter 2003;Chowdhury et al 2007). A study on ANC attendance among Bangladeshi women, who started to visit health centre during their pregnancy in order to receive antenatal care, reported that the urban and educated women who belonging to rich family and having access to mass-media were more likely to attend adequate ANC visits compared with their counterparts (Hossain, Akter, Sultana, and Kabir 2020).…”