MHS utilization status was the outcome variable in this study. The independent variable of primary interest was HFI.
MHS
A literature review assisted in the selection of proximate socioeconomic, demographic, and individual level factors for MHS.
Maternal healthcare utilization status is usually measured by two composite elements based on the services: (1) childbirth (place of delivery) and (2) before and after childbirth (ANC and postnatal care (PNC) visits) (
Baer et al. 2015). Based on these two elements, MHS utilization status was categorised into three groups. Group 1 consisted of those who received service neither at childbirth nor ANC and PNC services (at least one ANC and PNC attendance), group 2 consisted of those who received either of the two, and group 3 consisted of those who received both types of services.
First, a participant was considered as not using MHS provided the place of delivery was the home; this was dichotomized as the home (whether of the participant or relatives) versus a health institution (i.e., medical college hospital, upazila health complex, maternal and child welfare center, or health and family welfare centre). Second, a participant was considered as not using MHS before or after delivery unless she had at least one ANC and one PNC visit.
To select the potentially relevant covariates in the context of maternal health behavior, an extensive literature review was conducted focusing on the three most proximate themes: demographic status, socioeconomic status, and media use. As per availability in the dataset, the following were included in the analysis: age, region, residency, religion, education level of the participant and her husband, working status of the participant, type of employment of the husband, household wealth index, household having electricity, membership in any microcredit institution, autonomy in decision-making (being able to decide how spend personal income and on own health care), and media use (newspaper, TV, and radio).
Age
Age was trichotomized into three groups: 11–24 years, 25–34 years, and 35–49 years. Participants in all seven administrative regions including Barisal, Chittagong, Dhaka, Khulna, Rajshahi, Rangpur, and Sylhet were included in this study. Place of residency was categorized as rural or urban. Religion was categorized into Islam (the dominant faith in the country), and others (Hinduism, Buddhism, and Christianity). The working status of participants was categorized as employed or unemployed.
Educational status
Growing evidence suggests that educational status is one of the most direct indicators of population health and health-related behavior. The educational statuses of both the participant and their husband were categorized into four groups based on the total number of years of formal education received: 0 = nil, 1 = primary (1–5 years), 2 = secondary (6–10 years), and 3 = higher (>10 years).
Occupation
Type of occupation was categorized as follows: (1) farming = farmer, agricultural worker, fisherman, poultry farmer, and cattle raising; (2) blue-collar jobs = carpenter, mason, driver, construction worker, rickshaw puller, brick breaking, and road building; (3) white-collar jobs = businessman, physician, lawyer, accountant, teacher, and government service holder; (4) others = student, religious leader, retired, and unemployed.
Wealth index
Household wealth status was categorized as follows: poor, poorest, middle, rich, and richest. Households were ranked by the score they were assigned based on the possession of durable goods (e.g., TV, radio, refrigerator, and construction material). Categorization was performed by dividing the household wealth scores into quintiles. The richest households were those in the highest quintile, and poorest were in the lowest quintile. The procedures for calculating wealth status in DHS studies have been explained in previous studies (
Yaya et al. 2017).
Microcredit institution membership
Microfinance programs have been shown to have a positive impact on women’s empowerment and to promote overall socioeconomic status. Mothers who are involved with such programs are expected to have better economic solvency and therefore better adherence to maternal care protocols. Participants of this study were categorized according to their membership with any of the four microcredit institutions (Association for Social Advancement (ASA), Bangladesh Rural Advancement Committee (BRAC), Bangladesh Rural Development Board (BRDB), and Grameen Bank) as follows: 0 = no membership; 1 = membership with at least one institution or with multiple institutions.
Media use
Utilization of paper and electronic media has been shown to promote health awareness among adults. In this study, media use status was measured by the habit of watching TV, listening to radio, and reading newspaper as follows: 0 = not using at all; 1 = using sometimes/regularly.