Article Text

Download PDFPDF

Original research
Trends and patterns of inequalities in using facility delivery among reproductive-age women in Bangladesh: a decomposition analysis of 2007–2017 Demographic and Health Survey data
  1. Md Ashfikur Rahman1,
  2. Sumaya Sultana1,
  3. Satyajit Kundu2,3,4,
  4. Md Akhtarul Islam5,
  5. Harun Or Roshid5,
  6. Zahidul Islam Khan6,
  7. Mortuza Tohan1,
  8. Nusrat Jahan1,
  9. Bayezid Khan1,
  10. Md Hasan Howlader1
  1. 1Development Studies, Khulna University, Khulna, Bangladesh
  2. 2Global Health Institute, North South University, Dhaka, Bangladesh
  3. 3School of Public Health, Southeast University, Nanjing, People's Republic of China
  4. 4Faculty of Nutrition and Food Science, Patuakhali Science and Technology University, Patuakhali, Bangladesh
  5. 5Statistics Discipline, Khulna University, Khulna, Bangladesh
  6. 6Statistics, East West University, Dhaka, Bangladesh
  1. Correspondence to Md Ashfikur Rahman; ashfikurr{at}


Objectives The prime objectives of the study were to measure the prevalence of facility delivery, assess socioeconomic inequalities and determine potential associated factors in the use of facility delivery in Bangladesh.



Setting The study involved investigation of nationally representative secondary data from the Bangladesh Demographic and Health Survey between 2007 and 2017–2018.

Participants The participants of this study were 30 940 (weighted) Bangladeshi women between the ages of 15 and 49.

Methods Decomposition analysis and multivariable logistic regression were both used to analyse data to achieve the study objectives.

Results The prevalence of using facility delivery in Bangladesh has increased from 14.48% in 2007 to 49.26% in 2017–2018. The concentration index for facility delivery utilisation was 0.308 with respect to household wealth status (p<0.001), indicating that use of facility delivery was more concentrated among the rich group of people. Decomposition analysis also indicated that wealth quintiles (18.31%), mothers’ education (8.78%), place of residence (7.75%), birth order (5.56%), partners’ education (4.30%) and antenatal care (ANC) seeking (8.51%) were the major contributors to the prorich socioeconomic inequalities in the use of facility delivery. This study found that women from urban areas, were overweight, had any level of education, from wealthier families, had ANC, and whose partners had any level of education and involved in business were more likely to have facility births compared with their respective counterparts.

Conclusions This study found a prorich inequality in the use of facility delivery in Bangladesh. The socioeconomic disparities in facility delivery must be addressed if facility delivery usage is to increase in Bangladesh.

  • Health economics
  • Health policy
  • Public health

Data availability statement

Data are available in a public, open access repository. Data can be freely accessible upon request to the DHS website (

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Strengths and limitations of this study

  • This study used data from four nationally representative surveys with appropriate statistical technique to estimate the prevalence of using facility delivery and its associated factors, as well as the inequalities in using facility delivery over socioeconomic determinants; therefore, the study results could be generalisable across the country.

  • Our capacity to infer causality was constrained by the inherent drawbacks of a cross-sectional study design.

  • Some important factors related to the health of the respondents were not included due to unavailability and missing information of those variables in the Bangladesh Demographic and Health Survey data.

  • Using the robust technique concentration index, a relative measure of inequality was employed to quantify wealth-related inequalities in facility delivery utilisation.

  • Cluster effect and sample weighting were taken into consideration in the analysis of the present study.


Maternal mortality ratio (MMR) is still a major health concern around the world, particularly in developing nations like Bangladesh. Maternal mortality, according to the WHO, is defined as a ‘woman’s death while pregnant or within 42 days of delivery or termination of pregnancy from any cause linked to, or aggravated by, pregnancy or its management, but excludes deaths from incidental or unintentional causes’.1 Globally, in 2010, there were reportedly 287 000 maternal deaths, with low-income and middle-income countries (LMICs) accounting for majority of these deaths.2 In 2017, approximately 295 000 women died both during and after pregnancy and delivery, with 94% of these deaths happening in low-resource settings and with the vast majority of these deaths preventable.3 In 2017, MMR in low-income countries was 462 per 100 000 live births vs 11 per 100 000 live births in high-income countries.3 This high number of maternal deaths in some areas of the world reflects inequalities in access to quality health services and highlights the gap between the rich and the poor.

Reducing maternal mortality has long been a top global health concern. It is a Millennium Development Goal (MDG) target and a key component of the United Nations Secretary-General’s Global Strategy for Women’s and Children’s Health, which was unveiled in September 2010.4 5 The MDGs (1990–2015) underlined the significance of reducing mother and infant mortality by 75% and promoting a global MMR reduction of 38%.6 According to Sustainable Development Goal 3, MMR will be reduced to less than 70 deaths per 100 000 live births by 2030. Bangladesh, Nepal and Pakistan have all made significant progress in reducing MMR during the last few decades. Between 2010 and 2017, Bangladesh’s MMR dropped to 173 per 100 000 live births, Nepal to 186 per 100 000 and Pakistan to 140 per 100 000.2 In comparison with other LMICs around the world, MMR rates in these nations are still extremely high. Bangladesh is a developing country with eight administrative regions (Dhaka, Chittagong, Rajshahi, Khulna, Barishal, Sylhet, Rangpur, Mymensingh) and a total of more than 168 million people; data were collected from these eight regions using multistage cluster sampling.

To reduce maternal mortality, the factors behind these deaths have to be identified. Majority of these deaths are attributable to pregnancy-related delivery complications that are largely preventable by moving childbirth from home to a healthcare facility.7–10 Other disorders that might have existed before pregnancy which are not treated as part of a woman’s treatment may become more severe during pregnancy. Previous research has identified a number of key factors that contribute to low healthcare utilisation, including poor health-seeking behaviour, weak health systems, low socioeconomic status, cultural and personal health beliefs, lack of access to appropriate health services, high costs, long distances, lack of transportation options and poor quality of treatment.11 12 In South Asia, women who give birth at home are more likely to be exposed to unsafe and unclean conditions, putting the lives of the mothers and their newborns in danger.13 Several studies have found that using facility-based delivery services, family planning, and antenatal and postnatal care enable reductions in maternal deaths.8 14

The main rationale of this article is to analyse the socioeconomic inequalities in the utilisation of facility delivery in Bangladesh over time, based on its context using four rounds of data set to measure trends and contributing factors. Investigating the extent to which socioeconomic inequalities exist in facility delivery can aid in identifying the underlying causes of these disparities, thereby informing appropriate parties on how to address them. There are a few research that have analysed the socioeconomic factors of maternal health inequalities in Bangladesh using demographic and health survey data over a period of time. The prime objectives of this study are threefold: (1) to analyse the factors associated with facility delivery in Bangladesh using the Bangladesh Demographic and Health Survey (BDHS) data from 2007 to 2017–2018 in order to estimate the prevalence and trends of using facility delivery over time using four rounds of data set; (2) to measure socioeconomic inequalities in the use of facility delivery; and (3) to identify the primary components that explain socioeconomic inequalities in facility delivery over a period of time through a decomposition analysis.


Data sources

Secondary data from BDHS were used in this study (BDHS 2007, 2011, 2014 and 2017–2018). Demographic and health surveys are undertaken on a regular basis to determine the health status of the population. Demographic and health surveys provide a comprehensive picture of the study population, covering overall maternal and child health, as well as a range of other healthcare subject areas. The data set has been made freely available on the internet for academics and researchers to use. Survey strategy, methodology, sampling and questionnaires are all detailed in the final report.

Outcome variable

The outcome variable in our study was the place of delivery (0=home, 1=facility). If a woman gave birth in a hospital run by the government, a district hospital, a maternal and child welfare centre, an upazila health complex, a health and family welfare centre, a private hospital or clinic, a private medical college or hospital, a rural health centre, a basic health unit, a primary healthcare centre, an outreach clinic, or a clinic run by a family planning association, the location of the birth was considered a ‘facility’. If a woman gave birth at the respondent’s, a relative’s or a neighbour’s home, it was regarded as a ‘home delivery’.

Explanatory variables

The following variables were chosen based on literature review15–22: place of residence, division, mother’s age, mother’s education, mother’s employment status, number of antenatal care (ANC) visits, husband’s education, husband’s occupation and household wealth status; health-related characteristics, mother’s body mass index, age at first birth and ANC seeking were coded if the mother had taken at least four or more ANC during their last pregnancy. New division was generated using two divisions, Mymensingh and Rangpur, because these were not created during the earlier surveys in 2007 and 2011.

Statistical analysis

Data were weighted using an appropriate method suggested by the demographic and health survey platform; we used the svy command. The background characteristics of the study populations are described using descriptive statistics, and weighted prevalence with 95% CI is reported. The association between predictor variable and delivery location was investigated using χ2 testing. Multivariable logistic regression was used to estimate the net influence of predictor variables on the outcome variable after confounding variables were removed. We adjusted the multivariable and decomposition models based on p values <0.05. In the adjusted model, the factors that were statistically significant at the p<0.05 level in the univariate analysis were taken into consideration for final adjustment in the multivariate model. Unadjusted/crude OR (UOR) and adjusted OR (AOR) are presented in this article; however, only adjusted results are interpreted in the main text. All analyses were carried out using Stata/MP V.16.

Inequality measurement

The concentration curve (CC) and the concentration index (CIX) were employed in their relative formulations (with no corrections) to study the inequalities in facility utilisation across analysable socioeconomic factors of the population (women).23 The CIX in this study represents horizontal inequity because each woman in the study was assumed to have the same need for a facility birth. The CC was calculated by plotting the cumulative proportion of women ranked by their wealth index score (poorest first) against the cumulative proportion of facility deliveries on the y-axis. Absolute equality was shown by a 45° slope from the origin. The use of institutional delivery is equal among women if the CC intersects with the line of equality. If, on the other hand, the CC subtends the line of equality below (above), then there is inequality in the use of institutional delivery, which is skewed against women from low (high) socioeconomic backgrounds. Further, the greater the degree of inequality, the more the CC deviates from the line of equality. The CIX was calculated to estimate the level of wealth-related inequality. The CIX is widened as twice the region between the line of equality and the CC.23

The following are some of the benefits of adopting the CIX as a measure of healthcare inequality: it considers the socioeconomic dimension of healthcare inequalities because individuals are classified based on their socioeconomic status rather than their health status; it captures the experience of the entire population; and it is sensitive to changes in population distribution across socioeconomic groups. The CIX takes a value between − 1 and + 1. When institutional delivery is evenly spread across socioeconomic categories, the CIX equals 0. The usage of institutional delivery is concentrated among the upper socioeconomic classes if the CIX has a positive value (prorich). A negative CIX score, on the other hand, indicates that institutional delivery is mostly used by the poor (propoor).24 The CIX was calculated using the ‘convenient covariance’ formula provided by Wagstaff et al,23 as shown in the following equation:

Embedded Image

Here CIX is the Concentration Index, h is the health factor variable (place of delivery), μ is the weighted mean of factor variable (place of delivery), r is the fractional rank of individual in the distribution of wealth index, and cov(h, r) represents the covariance between h and r. The user-written STATA commands ‘Lorenz’25 and ‘conindex’26 were used to produce the CC and measure the CIX, respectively.

Decomposition of CIX

The relative CIX was decomposed to identify the proportion of inequality due to underlying determinant inequality. The findings were evaluated and interpreted using the Wagstaff et al23 and O’Donnell et al26 approach. The contribution of each determinant of facility delivery to overall wealth-related inequality is determined as the product of the determinant’s sensitivity to facility delivery (elasticity) and the degree of wealth-related inequality in that determinant (CIX of determinant). The residual is the portion of the CIX that is not explained by the determinants.

The ‘elasticity’ column indicates the change in the dependent variable (socioeconomic disparity in facility delivery) resulting from a one-unit change in the explanatory factors. A positive or negative elasticity score indicates an upward or downward trend in facility delivery in response to a favourable change in the determinants.

Patient and public involvement

No patients were involved.


Background characteristics of the study participants

Table 1 displays the socioeconomic and demographic characteristics of women aged 15–49. The table displays the results produced from 30 940 observations recorded in 2007, 2011, 2014 and 2017–2018, as well as the overall results derived from the data for all the years considered.

Table 1

Background characteristics of the study participants

From overall data, we can conclude that majority of women (67%) resided in rural areas, with the majority hailing from Chattogram (19%) and Dhaka (17%). Of the women, 22% belonged to the poorest group and 19.80% to the poorer group. The highest proportion of women was aged 15–24 years (49%), 43% had secondary education, 98% had improved water and 58% had improved sanitation, and only 25% were employed. In addition, 59% of women had normal BMI and 37% have already given birth. Among the mothers, 68% did not have ANC, yet majority (81%) had a normal last birth. In addition, majority of partners had a primary education (31%) and were primarily employed in non-agricultural occupations (52%).

Prevalence of facility delivery

Table 2 shows that, in 2007, 17% of women had facility delivery, which increased over the years. In 2017–2018, the percentage of facility delivery was 50%. Most women who were underweight went through home delivery (81%), while those who were overweight were more likely to have facility births (60%). Among women who had no ANC, 79% had home delivery and 60% had facility delivery who had any number of ANC. With regard to birth orders, home delivery was found to be more frequent in all categories and increased with increasing number of births, whereas the possibility of a facility birth was highest during the first birth but decreased with increasing number of births. However, women who had their last birth by caesarean section showed a high percentage (98%) of facility birth. The percentage of home delivery was found to be greater than facility birth in both urban (53%) and rural (76%) areas of Bangladesh. However, urban areas (47%) had more facility births than rural areas (24%). The percentage of home delivery was found to be higher than facility birth even when the observations were categorised according to divisions, with Khulna division found to have more facility births (46%) compared with other divisions. Facility births were also found to be more common among the wealthiest families (62%), while in all the other groups home delivery was found to be more frequent. Women and partners with higher education are more likely to have facility births, at 75% and 67%, respectively. Women’s working status and improved sanitation and water facilities do not seem to increase the rate of facility births; in all these cases, the percentage of home delivery was found to be higher. Moreover, the prevalence rates in table 2 show that women residing in urban areas (49%), women with higher education (72%), women whose last birth was by caesarean section (98%) and women richest in wealth index (61%) were more likely to have facility delivery compared with their counterparts. Figure 1 shows that facility births have become more prevalent over time from 2007 (14%) to 2017–2018 (49%).

Table 2

Prevalence of using facility delivery across different socioeconomic variables

Figure 1

Prevalence of using facility delivery over time in Bangladesh (weighted).

Factors associated with facility delivery (regression model)

The CIs for the bivariate and multivariate regression models at 95% are presented in table 3 as UOR and AOR, respectively. The analyses showed that in all 3 years (2011, 2014, 2017–2018), facility births increased compared with 2007 as the reference category, where in 2017 it was about four times higher. In both bivariate and multivariate analyses, it was found that women living in urban areas, from Dhaka and Khulna divisions, were overweight, had any level of education, belonged to wealthier families, had ANC, and whose partners had any level of education and were involved in business were more likely to have facility births compared with their respective counterparts. On the other hand, women from divisions other than Dhaka and Khulna, belonged to age groups 25–34 years and 35–49 years, were underweight, were employed, had any number of children, had improved water and sanitation, and whose partners were involved in agricultural or non-agricultural works were found to have lower odds of facility birth.

Table 3

Factors associated with facility delivery in Bangladesh

The analysis shows that women in the age group 25–34 years were about 1.54 times (CI 1.39 to 1.71) and in the age group 35–49 years about 2.43 times (CI 2.01 to 2.93) more likely to have facility birth compared with the age group 15–24 years. Women residing in urban areas were 1.44 times (CI 1.32 to 1.58) more likely to have facility birth. Overweight women were found to be 1.84 times (CI 1.66 to 2.04) more likely to have facility birth, whereas underweight women were 0.83 times (CI 0.75 to 0.91) less likely. Women who had any number of ANC were 2.38 times (CI 2.20 to 2.58) more likely to have facility births, but this tends to decrease with having more children over time. Education played a great role in the uptake of facility delivery, where findings show that with the increase in the level of education, more women tend to receive facility births. Similar result was found with the increase in the level of education of partners. In terms of wealth status, AOR was observed to increase as wealth status increased.

Decomposition of CIX for facility delivery

Table 4 illustrates the effects of key socioeconomic and demographic characteristics on facility utilisation and the disparities. The column labelled ‘Elasticity’ represents the amount of change in the dependent variable (socioeconomic inequality in facility delivery) caused by a one-unit change in the explanatory factors. Elasticity with a positive or negative sign indicates a rising or falling trend in the facility’s output in conjunction with a positive change in the factor.27 28 This study indicates that the value of the CIX for facility delivery was 0.30846363 (p<0.001) among Bangladeshi households with a higher socioeconomic status, indicating socioeconomic inequality in facility delivery in favour of the wealthy. The column ‘CIX’ displays the distribution of the determinants in terms of wealth quintiles. The positive or negative direction of the CI indicates whether the factors were more prevalent in the wealthy or poor group. The percentage contribution indicates how much each variable in the model contributes to socioeconomic disparities as a whole. A positive percentage contribution indicates that a factor contributes to the increase in observed socioeconomic gaps in the provision of healthcare facilities. A negative percentage contribution, on the other hand, indicates a component that is anticipated to reduce socioeconomic inequalities connected to facility delivery. Wealth quintiles (18.31%), mother’s education (8.78%), place of residence (7.75%), birth order (5.56%) and partner’s education (4.30%), as well as ANC seeking (8.51%), were the significant contributors to the prorich socioeconomic inequalities in facility delivery.

Table 4

Decomposition of concentration index for measuring socioeconomic inequalities

Figure 1 depicts the overall prevalence of the likelihood of using facility delivery during the course of the year. With the passage of time, it is apparent that facility delivery has increased. In 2007, the prevalence was only 14.48%, but climbed by at least 10-fold in 2011 (24.49%). In 2017–2018, nearly half of all women used facility delivery with skilled birth attendants (SBAs).

Using a Lorenz curve (CC), figure 2 also shows the disparities in facility delivery among the 4 distinct years. We can see that all four CCs fell below the line of equality, suggesting that facility delivery is more common among women from affluent households. Nevertheless, it seemed as though the CC was moving in the direction of equality. The difference between the line of equality and the CC was found to be at its widest in 2007, but narrowed in 2017.

Figure 2

Lorenz curve for inequality estimation.


The current study examined the socioeconomic inequalities associated with facility births among the Bangladeshi population using the most recent demographic and health survey data. An essential instrument for influencing policy choices that are influenced by inequalities is now analysis of socioeconomic inequality. Facility delivery is more common and concentrated among the richest Bangladeshis living in metropolitan areas, although it has substantially declined since the previous round of research. Household financial status, women’s education, ANC seeking, birth order, partners’ education and living in urban regions all had a substantial impact on the prorich socioeconomic inequalities in facility delivery.

Using four consecutive nationally representative BDHS data, this study revealed that there exist numerous socioeconomic inequalities in using facility delivery. The level of socioeconomic inequalities in facility births in Bangladesh is one of the uppermost among the South and East Asian countries.29 The results of our study show that rural areas had the maximum number of respondents (67.30%) and most of the women in these areas had normal last births (81.04%). Rural areas also had lower (23.91%) facility births than urban areas (47.26%). The results of this study also indicate that respondents from the lower age group (15–24 years) and who were overweight had more facility deliveries. Moreover, respondents from the wealthiest families and from the Khulna division were found to be more occupied with facility births. During the last 10 years, starting from 2007 to 2017–2018, the percentage of facility delivery has increased from 16.76% to 50.49%, although this is still low.18 This study showed that the respondents in 2017 had a higher likelihood of having facility births than the respondents in 2007, but this is still not sufficient. Facility birth is increasing, but at a slower rate, and several studies have shown similar results.17 30 31 Regional differences in using facilities are observed in this study and indicate that respondents from the Khulna and Dhaka divisions were more likely to have facility births than respondents from the new division. Regional differences and inequalities in using facility delivery are common, and results similar to our study exist.31 32 Young-aged respondents have higher likelihood of having a facility delivery than respondents from a higher age group. Several studies also showed the same results, and this may be because older women consider home delivery convenient and not risky.31 33 Also, there is a big difference between younger and older women in their knowledge and healthcare facility-seeking behaviour. Younger women are more interested in seeking knowledge and healthcare facilities.33 Women from urban areas were more likely to use facility births in comparison with respondents from rural areas in developing countries like Bangladesh.15–17 31 34 Moreover, overweight respondents have a higher likelihood of having facility delivery compared with respondents of normal weight. Existing studies show that respondents with non-normal weight have a higher likelihood of having facility delivery.35–37 A respondent may have more complicacy due to being overweight; consequently, overweight respondents tend to use more facility delivery.

Education is another significant factor influencing inequalities in using facility delivery. Respondents with a primary, secondary or higher level of education were more likely to receive facility births than respondents with no education. Education plays a key role in making a woman independent and autonomous in making her own healthcare decisions as she becomes more concerned about her health. This behaviour eventually enhanced women’s concern about facility delivery.22 33 38 Surprisingly, employed respondents were less likely to have the chance to use facility delivery than those who were not working.36 37 Employed respondents may experience time constraints, decreasing their opportunity to receive facility delivery.32 39

Again, respondents with educated partners have higher odds of using facility delivery than respondents with uneducated partners. There are similar results about existing inequalities in receiving facility delivery being influenced by education of the respondent and the husband.15 40–43 Education improves health awareness, and families with more education are more likely to use healthcare services. The socioeconomic disparities in facility delivery are also strongly influenced by the affluent position of the household. This study reveals that respondents from middle-class and affluent families were more likely to have facility delivery than those from low-income households. Clearly, educated respondents with educated partners have a greater likelihood of obtaining a high-paying job or earning more money and being able to afford maternal healthcare services such as delivery facilities.33 38 This finding of education and wealth index influencing inequalities in receiving facility delivery is consistent with previous studies conducted in different countries.44–47 These inequalities are influenced by different socioeconomic and demographic factors and their interactions.15 30 48 Moreover, majority of a low-income family’s money is typically spent on food and everyday necessities, and the cost of healthcare facilities and education is a hardship for this population; hence, they use home-based facilities for delivery. Therefore, low-educated and underprivileged individuals are typically denied access to facilities.

Additionally, this study revealed that respondents with improved water supply and sanitation facilities have higher odds of using facility delivery compared with respondents without improved water supply and sanitation facilities, which is a match with previous studies.49 Better sanitation and water facilities are primarily related to respondents’ level of education and socioeconomic standing, demonstrating a direct correlation between the two variables. Compared with respondents with a second or higher birth order, first-time mothers are more likely to have facility delivery for their first child.44 50 51 Also, similar to the results of other studies, this study showed that respondents with ANC have a higher likelihood of taking facility delivery than respondents with no ANC visit.36 52 53 An ANC visit creates consciousness among the respondents about the danger signs of labour and pregnancy complications, leading them to use facility delivery.36

Policy implications and specific recommendations

This research found a prorich inequality existing in Bangladeshi women’s use of birthing facilities. Therefore, public health policies and interventions should be implemented to increase the number of births that take place in these settings, such as provision of birth centres, training and assurance of SBAs, use of mass media for health education and raising awareness, implementation of mandatory female education, and participation of men in pregnancy and childbirth. Despite Bangladesh having achieved commendable success in using facility delivery among reproductive-age women, it is undeniable that women with less education and poor wealth status are highly deprived of getting facility delivery. The following are therefore recommended:

  • Immediate priority should be given to multisectoral interventions to upgrade facility delivery services covering the entire country, mostly the remote areas of Bangladesh.

  • Women with poor health conditions, with less education and of poor financial status should be covered with aiding facilities for using facility delivery services to motivate them as well as their families.

  • This study finds that for the first child most women use facility delivery services, but this rate goes down as the number of births goes up; hence, policymakers can introduce incentives to mothers who use facility delivery for their second child.

  • Further cohort study is recommended since a cross-sectional study has inherent limitations in establishing causality.

  • The government can spend more on women’s education and uplifting their positions to support the decision of availing facility delivery for every woman.

  • Further study can be conducted on facility delivery improvement strategies that are being followed by different countries to suggest better specific action plans for Bangladesh.

  • Identifying how women’s and their partners’ education helps improve the rate of facility delivery, as well as the far-reaching effect of education, should be beneficial for policymakers to be exact with their policies.

Limitations and strengths

The study has some limitations that include some important factors related to the health of the respondents and the delivery facility occurred due to the unavailability and missing information, such as cost of facility or caesarean birth, insurance, distance, waiting time, healthcare practitioners’ behaviour and availability of transportation facilities. Also, the cross-sectional nature of the study did not allow drawing a causal conclusion. Nonetheless, the study showed many strengths by using data from a large sample of a nationwide representative, population-based survey. Since this study was undertaken based on nationally representative consecutive data sets, the findings are more generalisable. Another strength of the study is the use of a more thorough decomposition analysis to determine the factors that influence socioeconomic inequalities in use of facility delivery. This is a robust method used to estimate health-related inequalities and is widely used in the public health literature. In addition, using the CIX as a measure of inequality index in healthcare has the following benefits: it captures the experience of the entire population; it takes into account the socioeconomic dimension of facility delivery because the classification of individuals is based on their socioeconomic status rather than their health status; and it is sensitive to changes in population distribution across socioeconomic groups.


This study indicates that women from urban areas, were overweight, had any level of education, from wealthy households, had ANC, and whose partners had any level of education and were involved in business were more likely to deliver in a hospital. This study also found a prorich inequality in facility delivery utilisation in Bangladesh, indicating that facility delivery utilisation was more prevalent among wealthier people. Existing socioeconomic inequalities in facility delivery must be addressed in order to boost the utilisation of facility delivery in Bangladesh. In light of these findings, it is essential to establish an intervention that targets these important linked factors in order to increase births in hospitals. Moreover, policy decision-making can prioritise the design and implementation of various poverty alleviation projects to eliminate socioeconomic disparities in facility delivery in Bangladesh.

Data availability statement

Data are available in a public, open access repository. Data can be freely accessible upon request to the DHS website (

Ethics statements

Patient consent for publication

Ethics approval

The Institutional Review Board and country-specific review committees ethically authorised all DHS survey protocols.


The authors of the present study greatly acknowledge the Demographic and Health Survey (DHS) for providing access to freely use their database.



  • Contributors MAR accepts full responsibility for the work and/or the conduct of the study, had access to the data, and controlled the decision to publish. MAR also takes responsibility for the integrity and accuracy of the data analysis. MAR and SS performed the statistical analysis. MAR, SS, SK, MAI, ZIK and HOR produced the first draft of the manuscript. MAR, SK, MHH, BK and NJ reviewed and undertook the scientific editing of the manuscript both for statistical correctness and language appropriateness. All authors read and approved the final version for publication.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.