Data source
This study is based on data from the Bangladesh Demographic and Health Survey (BDHS) 2017-2018, the eighth nationwide survey to provide information on the demographics and health status of women and children. The BDHS (2017-2018) is a national survey that includes a sample of about 20,250 randomly chosen households that is nationally representative. Individual interviews were available to all ever-married women between the ages of 15 and 49 who are regular residents of the chosen houses or who spent the night before the survey in the chosen homes. The survey was made to provide accurate estimates for important national indicators, as well as for urban and rural areas, as well as for each of the eight regions of the nation: Barishal, Chattogram, Dhaka, Khulna, Mymensingh, Rajshahi, Rangpur, and Sylhet.
Study variables
In this study, the outcome variable is Antenatal care visit (ANC) and it is measured as a binary outcome. Thus, antenatal care visit was categorized as Yes (taken 4 or more antenatal care visit, coded as 1) or No (less than 4 antenatal care visits, coded as 0) for all the models. The predictors (features) used in this study include, Regional place of respondent (Barisal, Chittagong, Dhaka, Khulna, Rajshahi, Rangpur, Sylhet, Mymensingh), Individual age in 5-year groups, Educational level of respondent (No education, Primary, Secondary, Higher),Wealth index status (Poorest, Poorer, Middle, Richer, Richest), Types of residency (urban, rural), Educational level of respondent’s partner (No education, Primary, Secondary, Higher), Respondent employment status (employed, not employed/housewife), Total number of children in household, Health insurance (yes, no), Sex of household head (Male, Female), and household size. The selection of these predictor variables was based on information from existing literature on the subject (Denny et al., 2022).
Analytic strategy
The independent variables for this study were categorical data of the respondents’ residence, age, education level, regional places of household head, partner educational level, wealth index, and number of household member, and number of children. A χ2 test was used to detect the variables related to ANC utilization during pregnancy. The SPSS V.23 software was operated to perform the data processing and chi-square test. The R programming language (version 4.1.2) was used for multivariate analysis. When we want to look at a dependence structure with a dependent variable and a set of explanatory variables (one or more), we can use the logistic regression framework. In many cases research focuses on models where the dependent variable is categorical. For example, the dependent variable in here completing ANC visits or not completing, and we could be interested in how this variable is related to age group, wealth status, educational level, type of residence, no of children, husband’s educational level, health insurance etc. In this case we could not carry out a multiple linear regression as many of the assumptions of this technique will not be met, as will be explained theoretically below. Instead we would carry out a binary logistic regression.
We discussed the multivariate regression analysis techniques and then apply in our data to identify the relationship among Bangladeshi women's antenatal health care (ANC) with various associated factors.
Chi-square Test of Independence
The chi-square test is a way of analyzing data for categorical variables. The information was presented in the form of counts. The Chi-square test for association (independence) in a two-way classification use this approach to determine if the likelihood of items or subjects being classified for one variable is dependent on the other. The goal of the chi-square test of independence is to determine whether or not there is a relationship between two category variables. The data displayed by a table known as contingency table. If (i=1, 2, ..., n) is a set of observed frequencies and (i = 1, 2, ..., n) is the corresponding set of expected frequencies, then the mathematical form of Chi-square is given by,
![](data:image/png;base64,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)
The Logistic Regression Model
Logistic regression is applied when the regressed variable, the dependent variable, or the response variable is qualitative or categorical in character (Hosmer W et al., 2000). A qualitative response variable might be binary (dichotomous) or have several categories. In this investigation, we would apply binary logistic regression. The logistic regression model is a generalized linear model with random components. In a variety of regression application, a response variable of interest has just two possible qualitative outcomes and can thus be represented by binary indicator variables with values 0 and 1. This response variable is denoted by Y, and its two possible values are 0 and 1, indicating failure and success, respectively. Logistic regression predicts a discrete result, such as group membership, from a predictor variable that might be continuous, discrete, dichotomous, or a combination of any of these.
The outcome variable antenatal care visit (ANC) was measured as a binary outcome. Thus, antenatal care visit was categorized as Yes (taken 4 or more antenatal care visit, coded as 1) or No (less than 4 antenatal care visits, coded as 0) for all the models.
In this study, we deal with a large number of explanatory variables. The explanatory variables used in the binary logistic regression analysis are as follows: Regional place of respondent (Barisal, Chittagong, Dhaka, Khulna, Rajshahi, Rangpur, Sylhet, Mymensingh), Highest educational level of respondent (No education, Primary, Secondary, Higher), Respondent’s economic status (Poorest, Poorer, Middle, Richer, Richest), Individual age in 5-year groups, Husband’s educational level (No education, Primary, Secondary, Higher), No. of household member, Total number of children in household, and Type of residence (Urban, Rural).
Odds Ratio
Logistic regressions work with odds, hence odds and odds ratio must be defined. The odds are just the probability ratio of the two probable events. If p is the likelihood of the event occurring, then 1 -p is the chance of the event not occurring.