Levels and determinants of quality of antenatal care in Bangladesh: Evidence from the Bangladesh Demographic and Health Survey

Background: Assessing the quality of antenatal care (ANC) is imperative for improving care provisions during pregnancy to ensure the health of mother and baby. In Bangladesh, there is a dearth of research on the quality of ANC using nationally representative data to understand its levels and determinants. The current study aims to assess the quality of ANC and identify the sociodemographic factors associated with the usage of quality ANC services in Bangladesh. Methods: We conducted secondary data analysis using the last two Bangladesh Demographic and Health Surveys (BDHS) (2014 and 2017-18). A total of 8,277 ever-married women were included in the analysis (3,631 from 2014 and 4,646 from 2017-18 BDHS). We constructed the quality ANC index using a principal component analysis on different ANC components: weight, blood pressure measurement, blood and urine tests, counseling about pregnancy complications and a minimum of four ANC visits of which one is by a medically trained provider. Multinomial logistic regression was used to determine the strength of association. Results: Receiving all the six components of quality ANC increased from about 13% in 2014 (BDHS 2014) to 18% in 2017/18 (BDHS 2017-18) with a significant difference of p < 0.001. Women from the poorest group, being rural areas, with no education, high birth order and unexposed to media were less likely to receive high-quality ANC than women from the richest group, from urban areas, with a higher level of education, low birth order and exposure to media. Conclusion: There is a need to improve the quality of ANC services in Bangladesh. An education program for women, with regular knowledge-enhancing sessions for pregnant mothers, may help them understand the value of ANC visits. Documentaries about maternal and child healthcare can be broadcast on television, YouTube, Facebook, radio and other digital platforms regularly.


46
Assessing the quality of antenatal care (ANC) is imperative for improving care provisions during 47 pregnancy to ensure the health of mother and baby (1). Systematic supervision of a woman 3 48 during pregnancy is the core intervention within the continuum of care for mothers and babies 49 known as ANC (2). Globally, complications during pregnancy have consistently been a leading 50 cause of maternal death, stillbirth and neonatal death (3). Poor-quality ANC is one contributing 51 factor to this (3). The latest Sustainable Development Goals' (SDGs) focused on reducing maternal 52 deaths to 70 per 100,000 live births globally by 2030 (4). In addition, the Bangladeshi government 53 has set the goal of reducing the maternal mortality ratio (MMR) from 176 per 100,000 live births 54 to 105 by 2021 (5). It has been established that appropriate quality ANC can save lives (3) and 55 lower maternal mortality by up to 20% (6, 7). Quality ANC has a protective effect against adverse 56 pregnancy outcomes such as low birth weight and premature birth (8-10). In addition, skilled 57 birth attendance (doctors, nurses, and midwives) utilisation during delivery and postnatal care 58 increases with quality ANC (11). Although 64% of pregnant women worldwide attended at least 59 four WHO-recommended ANC activities in 2007, the corresponding figure for 60 Bangladesh was only 47% (BDHS) in 2017-18 (12). 61 Assessing the quality of ANC necessitates information on service usage of the recommended 62 contents of ANC (13)(14)(15)(16)(17). Some literature has defined quality ANC in terms of its required 63 components, which should be delivered during pregnancy for a better, healthier life (3,12,18,64 19). Recently, the 2016 WHO ANC model has been published, and it recommended a minimum 65 of eight ANC contacts as well as some other components relevant to routine ANC for a successful 66 pregnancy. These include five types of interventions to upgrade the delivery of quality ANC: 67 nutritional interventions, maternal and foetal assessment, preventive measures, interventions 68 for physiological symptoms and health systems interventions (3). A study in Kenya assessed the 69 quality of ANC by evaluating service provisions and the mother's experience of care, which is an 70 approach based on the WHO model for ANC (3) and Kenya's National Guidelines for Quality 71 Obstetrics and Perinatal Care, which includes blood pressure measurement, foetal growth exams, 72 urine tests, iron and folic acid supplementation, tetanus toxoid immunisations and at least two 73 doses of intermittent preventive treatment for malaria during pregnancy (18). Physical 74 examination (blood pressure, pulse, temperature) along with minimum basic investigations 75 (haemoglobin, urine, blood group) are considered as appropriate care during ANC according to . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)  Singh et al. (2019) revealed that few studies were conducted to understand the factors that 97 influence ANC quality in low-and middle-income countries (28). Only a few researchers used 98 existing demographic and health survey data to examine the determinants and contents of ANC 99 contacts in Bangladesh (7,29,30). The aim of this study is therefore to (a) determine the quality 100 of ANC using the definition mentioned in the BDHS 2017-18 using the last two BDHS surveys 101 (2014 and 2017-18) and (b) examine the factors associated with the usage of the quality of ANC. 102 It is beneficial to conduct this study as there is no study that levels the quality of ANC in 103 Bangladesh using the most updated information.
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

105
A secondary analysis was conducted by using data from the last two BDHSs (2014 and 2017-18), 106 which are the seventh and eighth in the series undertaken in Bangladesh. They covered all the 107 districts across the administrative divisions of Bangladesh through nationally representative 108 cross-sectional surveys (12,20). The BDHS is a retrospective study following a multi-stage 109 stratified cluster sampling design. ANC-related information was collected for the last live birth ANC by considering these components: weight measurement, blood pressure measurement, 121 blood and urine tests, medical advice regarding signs of pregnancy complications and number of 122 ANC visits. For ANC to be considered quality, four visits must take place, one of which must be 123 from a medically trained provider (MTP). We constructed the quality ANCindex based on the 124 above-mentioned components using principal component analysis (PCA). After that, we 125 considered the first principle component as the index obtained from the PCA. We divided this 126 into three categories (low, middle and high). This categorical variable was the outcome variable 127 used in this study, i.e., the quality of ANC tertile. 128 Several socioeconomic and demographic characteristics were considered as explanatory 129 variables. These included division (Dhaka,Barishal,Chattogram,Khulna,Rajshahi,Rangpur,130 Sylhet and Mymensingh), place of residence (urban or rural), wealth quintile (the five categories 131 -richest, richer, middle, poorer and poorest -were derived using PCA based on ownership of . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted May 31, 2022. ; https://doi.org/10.1101/2022.05.31.22275822 doi: medRxiv preprint 132 household goods), mother's education level (higher, primary, secondary and no education), 133 maternal age in years (15-19, 20-24, 25-29, 30-34 and 35-49), birth order (1, 2-3, 4-5 and 6+), 134 religion (Islam or others) and media exposure (exposed or unexposed). Respondents were 135 considered exposed to media if they performed any of these three actions at least once a week: 136 reading a newspaper or magazine, watching television or listening to the radio. Otherwise, they 137 were considered unexposed to media. 138 Statistical analysis 139 We used descriptive statistics to report the characteristics of the mothers accessing quality ANC. 140 Also, the p-value from a chi-square test was reported for both surveys for observing trends over 141 ordered characteristics of quality ANC, according to socioeconomic and demographic 142 characteristics. A 5% level of significance was considered to indicate a significant trend. 143 Multinomial logistic regression was performed for the pooled data from the 2014 and 2017-18 144 BDHSs to assess the factors affecting the usage of quality ANC while considering the low category 145 of quality ANC as a reference. We reported a crude and adjusted odds ratio (OR) and a 146 corresponding 95% confidence interval (CI) for the multinomial logistic regression. The statistical 147 package Stata 14 (31) was used to perform the analysis.
148 Table 1 shows the background characteristics of the mothers who were included in this analysis. 150 The Dhaka division had the highest percentage of women (31%) participating in the survey, while 151 the Barishal and Mymensingh divisions had the lowest percentages of women (5%). Around 71% 152 of the mothers were from rural areas, and 17% were from the poorest socioeconomic group. In 153 both surveys, more than half of the mothers (51%) had a secondary level of education. The 20-154 24 year age group accounted for the majority of the maternal age group (35%). A large number 155 of mothers (48%) reported a birth order of 2 to 3. Considering religion, around 92% of women 156 were Muslim. Around 58% of the mothers were exposed to media.

149
157 Table 1: Distribution of the mothers according to background characteristics, presented as frequency 158 and percentage . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted May 31, 2022. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. percentage of mothers with high quality ANC increased for all age groups and birth orders. A 180 change in trends of receiving quality of ANC status by the mother's exposure to media status was 181 observed in both surveys; the percentage of mothers exposed to media receiving a high quality 182 of ANC increased by around 15% (31% in 2014 and 46% in 2017-18).
183  and media exposure were significant determinants for quality of ANC index. Mothers living in . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted May 31, 2022. Mothers who were not exposed to media were 201 less likely to receive high-quality ANC compared with mothers exposed to media (AOR: 0.65, 95% 202 CI: 0.57-0.75).
203 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted May 31, 2022.  This study reveals that socioeconomic and demographic characteristics are related to receiving 217 better quality ANC and that a considerable proportion of women did not receive adequate ANC 218 with quality. A low percentage of mothers received all six components of quality ANC, with a . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted May 31, 2022. ; https://doi.org/10.1101/2022.05.31.22275822 doi: medRxiv preprint 219 significant difference between the two surveys (13% in 2014 and 18% in 2017-18 BDHS). Women 220 with high birth order, lower levels of education, belonging to the poorer group, living in a rural 221 area and not exposed to media were less likely to receive high-quality ANC. Around 57% of 222 women from the richest group and around 57% of women with a higher education level received 223 high-quality ANC, whereas only around 12% of women from the poorest group and around 12% 224 of women with no education received high-quality ANC. 225 This study found that mothers with a higher level of education have a higher likelihood of 226 receiving quality ANC, which is consistent with other studies' findings from different parts of the 227 world (7, 32-35). A study conducted in Bangladesh documented that the mothers' educational 228 level strongly affects the optimal uptake of ANC services (7). Uneducated women have less 229 independent decision-making power than educated women (36)   is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted May 31, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022 13 248 quality ANC (34). A study in Bangladesh reported that women in urban areas were more likely to 249 receive necessary ANC services than women in rural areas (37). In addition, another study 250 conducted in Bangladesh reported that the lower socioeconomic status of women in rural areas 251 might be associated with a lower uptake of ANC services (7). 252 A prime determining factor of receiving quality ANC is wealth status. Women with a higher wealth 253 status are more likely to receive a high quality of ANC. This study found that women from the 254 poorest group are less likely to receive high-quality ANC compared with women from the richest 255 group. This finding is consistent with previous studies (7,28,34). A study conducted using 91 256 national household surveys documented that poor women receive a lower quality of care (1). 257 Wealth status could influence the health-seeking nature of women in multiple ways, and without 258 the overall betterment of their standard of living, efforts to encourage poorer women's utilisation 259 of maternal health services might not obtain useful results (7). 260 The age of women also affects their uptake of ANC services. The findings from this study show 261 that young women have a lower likelihood of accessing high-quality ANC. Unlike teenage 262 mothers, older women in Bangladesh frequently accessed ANC services (30). A study conducted 263 in Ghana reported that women's age is associated with the extent of their ANC uptake. Older 264 women are more likely to develop difficulties at birth and have higher uptake of ANC (26). In 265 another study, Tessema & Minyihun (2021) (39) documented that older women understand the 266 benefits of visiting health facilities. 267 We also found that birth orders are related to uptake of quality ANC services. Women with higher 268 birth order are less likely to utilise high-quality ANC services, implying that women with previous 269 birth experiences demand less ANC. This finding is similar to the results obtained from the study 270 conducted by Nketiah et al. (2013) (26), documenting that in the case of higher births, women 271 are less likely to access ANC than during earlier pregnancies. In addition, a study conducted in 272 Ethiopia reported that women with multiple births are less likely to seek ANC services (2). In 273 Bangladesh, it is apparent that women who give birth for the first time have a higher likelihood 274 of using recommended ANC services (37). This could be due to the belief that they do not need 275 ANC services as they have earlier experiences with pregnancy and childbirth.
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted May 31, 2022. Our findings on exposure to media as a factor influencing uptake of ANC services are in line with 277 other studies (7,37,40,41). For instance, a study conducted in Uganda reported that women 278 who have access to media are more likely to use ANC services than women who do not (41). 279 Another study in Nepal found that exposure to media positively impacts the utilisation of ANC 280 components (40). Women exposed to media are more likely to receive ANC services in 281 Bangladesh (37). Women who have access to media are more informed than their counterparts 282 about the usefulness of ANC services during pregnancy and ask the ANC providers for services 283 that may help them lead a healthy life (41). Also, mothers who watch television are more likely 284 to know the problems related to pregnancy and the importance of using ANC services (7). 286 We believe this is the first study designing the quality of the ANC tertile using components utilised 287 by women in Bangladesh using nationally representative demographic and health surveys, 288 meaning that its findings are generalisable at the national and sub-national levels. In addition, 289 we have conducted this study using the pooled data of two surveys (2014 and 2017-18 BDHSs), 290 which indicates a larger population-based study for examining the factors associated with the 291 uptake of quality ANC. 292 The data were collected for three years preceding the survey using the recall method; as a result, 293 there may be some recall bias in the study. However, in the sample, 36.8% of the births were 294 identified for the previous year, 34.8% within 1-2 years and 28.4% within 2-3 years (42). As the 295 recall period varies, the recall bias should be quite low. In the survey, ANC-related information is 296 only available for mothers with live births. Therefore stillbirths and mothers who died during 297 pregnancy or delivery are not included in the study. In addition, we do not have data relating to 298 all of the WHO 2016 ANC model recommendations. However, the BDHS is the best available 299 population based national representative data on the quality of ANC in Bangladesh. We used the 300 BDHS 2017-18 definition of quality ANC as a representation of quality ANC for this study. Due to 301 data limitations, we could not address some important confounding factors, such as cost of care, 302 availability of health facilities and knowledge about maternal healthcare services, that may 303 influence uptake of quality ANC.

Strengths and limitations
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

305
The findings obtained from the analysis provide essential information for policymakers to use to 306 implement the following necessary initiatives to lower maternal and neonatal death by ensuring 307 the quality of maternal health services during pregnancy. Special attention needs to be paid to 308 women in rural areas because they are more deprived than those in urban areas in terms of 309 medical facilities and other amenities. Obstacles to women's education, such as child marriage 310 and gender-based violence, should be addressed. Re-entry strategies for young mothers should 311 be implemented, just as there should be methods in place to assist women who dropped out of school to 312 return to their studies and the hygiene or sanitation needs of women in schools should be 313 reviewed. In addition, women from early reproductive age groups should be given extra 314 attention. Therefore, there is a need for an education program for women; regular knowledge-315 enhancing sessions for pregnant mothers may play a vital role in increasing the awareness of the 316 importance of attending ANC visits. The role of media exposure should also be given 317 consideration. Documentaries about maternal health and child healthcare can regularly be 318 broadcast on television, YouTube, Facebook and radio. Drama and Jatra shows could even be 319 staged in rural areas, which would not only will entertain but also will raise awareness. Moreover, 320 trained healthcare providers at the field level should be effectively engaged so that they can go 321 door to door and provide adequate services to pregnant mothers. If the findings and suggestions 322 presented in this paper are acted upon, mothers will become aware of healthcare services during 323 pregnancy. 324