Abstract
Background
Non-institutional deliveries, defined as childbirth outside healthcare facilities, remain a significant global concern. It poses serious maternal and child health problems and significantly contributes to maternal and infant mortality. India witnessed a sustained decrease in non-institutional deliveries in thepast 20 years. However, the rate of decline has tapered recently while the country still has a sizeable number of women delivering out of the facility. To better understand the enduring proportions of non-institutional deliveries, this study preliminarily analyses its predictors. The prime objective of the study is to unravel the inequality in the prevalence of non-institutional deliveries and understand if they are unfairly concentrated among certain households in India. It also aims to provide policy-relevant insights into the socioeconomic factors contributing to its concentration among specific households and their implications for maternal and child health.
Methodology
Using the National Family Health Survey-5 (NFHS-5) data (N = 1,75,569 deliveries), we developed a regression model to understand the existing non-institutional deliveries and their predictors. We employed the Erreygers’ Concentration Index (ECI) to quantify the degree of concentration (inequality) of non-institutional deliveries among households. Furthermore, a decomposition analysis was run to analyse the factors contributing to the concentration of non-institutional deliveries in a particular groups of households. It breaks down the overall inequality at population level into its constituent parts to identify the sources of inequality. This approach helps discover prime causes of inequality, such as differences in income, education, or other relevant factors.
Findings
Out of the total sampled deliveries in the reference period, around 14% were non-institutional. Inequality analysis (ECI=-0.2174; p-value < 0.0001) suggests that non-institutional deliveries were unequally and unfairly concentrated in low-income households. The concentration of non-institutional deliveries in poor households was majorly contributed by factors like education (13.85%), wealth (13.91%), mass media exposure (12.27%) region (9.76%), birth order (3.17%), distance to health facilities (2.77%), caste (2.82%), timing of first ANC visit (1.07%), and women considering having to take transport as a problem (1.60%).
Conclusions
This research employs inequality analyses of non-institutional deliveries and contributes to the existing literature by establishing its unfair concentration among poor households in India. It expands our understanding of the factors driving non-institutional deliveries among the disadvantaged. The findings highlight the importance of targeted interventions and policies to reduce the concentration of last-mile non-institutional deliveries among vulnerable women living in marginalized households.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12905-025-03819-8.
Keywords: Non-institutional deliveries, Wealth-based inequality, Erreygers concentration index (ECI), Decomposition analysis, India
Introduction
Non-institutional deliveries, defined as childbirth outside healthcare facilities, often occur in unhygienic and unsafe environments. It is found to be associated with higher maternal and neonatal morbidity and mortality rates and remains a significant health concern in lower- and middle-income countries (LMICs) [1–5]. Every day, approximately 800 women succumb to pregnancy-related complications and childbirth, with Southern Asia accounting for approximately one-fifth of global maternal deaths in 2020 [6]. In cases of non-institutional deliveries unattended by a skilled birth attendant, the chances of delivery-related complications resulting in deaths are high [7]. Other issues that may arise in such deliveries include occurrence of uterine perforation [8], newborns left unattended, or improper care by the family members [9, 10] leading to a high risk of maternal and infant mortality.
Previous studies ascertained various factors of non-institutional deliveries, such as the issue of unavailability of healthcare services and shortage of skilled birth attendants in rural areas [11–13], long distance to a health facility [14], high transportation costs, and inability to pay hospital bills [2], awareness and access to ANC [2, 3, 5], low educational attainment [2, 15], household wealth [16, 17], late registration to ANC [18], higher birth order [3, 4]. Socio-cultural factors such as religion, caste, region, and place of residence were also found to be significant factors in the choice of place of delivery [3–5, 18–20]. Past research also highlighted a significant inequality in institutional deliveries among households regarding disparities in access to and utilisation of institutional deliveries across socioeconomic, regional, and cultural dimensions. Lower socioeconomic status, lack of education, and limited financial resources have been identified as barriers to accessing institutional deliveries [21, 22]. Cultural beliefs and traditional practices surrounding childbirth also contributed into disparities, as some communities still prioritized home births or relied on traditional birth attendants [23–25].
It is widely acknowledged that most pregnancy and birth-related complications can be timely managed with skilled birth attendance in a health facility. Hence, worldwide large-scale programs and policies to eliminate non-institutional deliveries have been implemented in the last few decades [1]. In India, these policies and programs were conceptualised and implemented at the central level, mainly through the National Health Policy [26] and the National Rural Health Mission [26]. The launch of Janani Suraksha Yojana (JSY) in this direction is considered a policy breakthrough, wherein pregnant women delivering at a health facility are incentivized through monetary rewards [27]. As a result of focused policy efforts, India witnessed a sustained decrease in non-institutional deliveries in the past three decaded (Fig. 1), although the rate of decline has tapered recently. Between 2015 and 2021, there has been a noticeable fall in the decline rate of non-institutional deliveries in India, dropping to an average annual rate of 2.5% in NFHS-5 from 4% in NFHS-4. While this fall in the decline of non-institutional deliveries in India represents a national average, regionally its concentrated in few states more than others (see supplementary.1). This decline rate has notably been slower recently compared to that of the initial ten-year period of Janani Suraksha Yojana (JSY), where the rate of decline was twice as high (Fig. 1). As per the latest data, 11.4% of births in the last five years preceding the survey (NFHS-5, 2019-21) occurred outside healthcare facilities. This recent trend is worrisome as India contributes around 12% of global maternal deaths annually [6]. It has also been highlighted that existing policy tools like JSY were of limited effectiveness in addressing such inequalities as incentives were insufficient compared to the actual cost of delivering at a health facility [28, 29]. The declining annual rates of reduction in non-institutional deliveries hints towards a serious issue recognised as a ‘last mile problem’ that affects the implementation and success of such social welfare programs. We argue that a research approach with a focus on such a problem will be helpful in devising effective policy solutions for it.
Fig. 1.
Trends of non-institutional deliveries in India. Data Source: National Family Health Survey (NFHS-5), 2019-21, India
Most of the previous studies in India concerning safe deliveries have solely focused on attributes and factors of institutional deliveries. However, in the current research, we have hypothesised that the remaining 11.4% of the non-institutional deliveries appears more concerning in light of the fact that social development interventions, programs, and policies have been indifferent and worked sluggishly for “the most vulnerable 10% of the people”. Considering the large population of India, even such a small figure represents a huge population that is being underserved. India may now need to restructure its policy measures to eliminate non-institutional deliveries entirely and achieve universal institutional deliveries. To further this agenda, in the current study, we have focused on last-mile non-institutional delivery and analysed it in two stages. In the first stage, we attempt to discover the predictors of non-institutional deliveries in India. In the following second stage, we inquire whether existing non-institutional deliveries are evenly spread across all sorts of households (and women) irrespective of socioeconomic status or are unfairly concentrated among underprivileged populations. Thereafter, we use a wealth-based inequality decomposition analysis to determine the contribution of various socioeconomic variables to this inequality.
Data and methods
Data source
This study used the data from the fifth round of the National Family Health Survey (NFHS), 2019-21, India [30]. The International Institute for Population Sciences (IIPS) conducts the NFHS in collaboration with the Ministry of Health and Family Welfare, Government of India. The NFHS data is one of the most reliable and nationally representative sources of information on women’s and children’s health indicators in India.
In the context of childbirth, NFHS-5 employed a five-year recall period. In order to minimise recall bias, we analysed ‘births that occurred in the last three years’ preceding the survey. This curation process leaves us with a total sample of 1,75,569 birth cases, which has information pertaining to ‘place of delivery’ and consists of 1,51,519 institutional and 24,050 (13.70%) non-institutional deliveries. We extracted relevant data from the ‘Births Recode’ data file to facilitate our analysis, excluding birth cases falling outside the three-year timeframe.
Outcome variable
‘Place of delivery’ is the outcome variable of this study. In the survey, the respondent mothers were asked, “Where did you give birth to (name of the child)?“. Based on their chosen option in the data, we have recategorised the place of delivery into two categories. Institutional delivery recoded as “0” includes births that occurred in any public or private institution, including “government/municipal hospital, government dispensary, UHC/UHP/UFWC, CHC/rural hospital/block PHC, PHC/Additional PHC, Sub-centre, other public sector health facility, private hospital/maternity home or clinic, NGO or trust hospital/clinic”. Home delivery or non-institutional delivery, recoded as “1”, includes births that occurred in “respondent’s home, parent’s home, or other’s home”. Since our variable of interest in this study is non-institutional delivery, we have coded it as “1” in our binary outcome variable i.e. ‘Place of delivery’.
Independent variables
Variables related to various socioeconomic and demographic characteristics are included in the analysis as independent variables to identify factors related to non-institutional deliveries. The analysis also covered the variables related to access to a health facility as independent variables. All the predictor (background) variables included in the study/analysis are listed in Table 1 with brief details. These variables are women’s age group, place of residence, the highest level of education, sex of head of the household, religion, caste categories, household wealth index, timing of 1st Antenatal Care (ANC) visit, health insurance coverage, birth order number, region and problems in accessing health facility.
Table 1.
Background variables
| Variable | Description | Coding |
|---|---|---|
| Age Group: | Respondents age in five-year groups from 15 to 49 years. |
1 = 15–19, 2 = 20–24, 3 = 25–29, 4 = 30–34, 5 = 35–39, 6 = 40–44, 7 = 45–49 |
| Place of Residence: | The type of place of residence where the respondent lives. | 1 = Urban, 2 = Rural |
| Education: | The respondent obtained the highest level of education. |
0 = No education, 1 = Primary, 2 = Secondary, 3 = Higher |
| Sex of head: | Sex of head of the household. | 1 = male, 2 = female |
| Religion: | The religion of the respondent. |
1 = Hindu, 2 = Muslim, 3 = Christian 4 = Sikh, 5 = Other |
| Caste: | Caste categories of the respondents. | 1 = Scheduled Caste, 2 = Scheduled Tribe, 3 = Other Backward Caste, 4 = General/Other |
| Wealth Index: | The wealth index of the respondent’s household. | 1 = Poorest, 2 = Poorer, 3 = Middle, 4 = Richer, 5 = Richest |
| Timing of 1st ANC: | Timing of 1st Antenatal Care visit. | 1 = first trimester, 2 = second trimester, 3 = third trimester |
| Health insurance: | Covered by health insurance. | 0 = No, 1 = Yes |
| Birth order: | Birth order numbers were recorded and categorized into four categories. | 1 = 1st, 2 = 2nd & 3rd, 3 = 4th & 5th, 4 = over 5th |
| Mass media exposure: | Questions related to television, newspaper/magazine, and radio use has been asked to respondents in NFHS-5. Responses were not at all, less than once a week, and at least once a week. If someone responded, not at all (never) use any of the three media sources, we considered them as having no exposure to mass media. |
0 = No 1 = Yes |
| Getting permission to go: | If getting permission to go to get medical help an issue/problem? |
0 = No problem, 1 = Little problem, 2 = Big problem |
| Getting money needed for treatment: | If getting money needed for treatment is an issue/problem? |
0 = No problem, 1 = Little problem, 2 = Big problem |
| Distance to the health facility: | If the distance to the health facility to get the medical help is an issue/problem? |
0 = No problem, 1 = Little problem, 2 = Big problem |
| Having to take transport: | If having to take transport is an issue/problem to get the medical help? |
0 = No problem, 1 = Little problem, 2 = Big problem |
| Not wanting to go alone: | If not wanting to go alone is an issue/problem in getting medical help? |
0 = No problem, 1 = Little problem, 2 = Big problem |
| Concern no female health provider: | If concern of no female health provider is an issue/problem in getting medical help? |
0 = No problem, 1 = Little problem, 2 = Big problem |
| Region: | Region of the country where the respondent is from. States and union territories are classified into 6 regions. |
1 = Northern, 2 = Central, 3 = Eastern, 4 = Western,5 = Southern, 6 = North-Eastern |
Statistical analysis
As the first step of the analysis, Pearson’s Chi-square test was used to measure the association of each background variable with the outcome variable to summarise the data and check its significance for inferential statistics. A binary logistic regression model was then applied to find the factors significantally associated with the probablity of non-institutional deliveries. Since the outcome variable is dichotomous, viz. “place of delivery,” it is categorised as ‘Institutional’ and ‘non-institutional’ delivery to describe the effects of each predictor on the outcome variable. The Binary Logistic model can be described using the following equation;
Logit
α + (β1 × 1 + β2 × 2 +………. + βn Xn) + ε.
Where Y is the outcome variable of interest (non-institutional delivery is ‘Yes’), π is the probability of occurrence of the outcome variable. In this case, 
represents the odds. Logit is the natural log (ln) of odds of Y (non-institutional deliveries). α is the Y-intercept, βn is the nth predictor’s regression coefficient, Xn is the nth independent variable, and ε is the standard error of the equation. The value of coefficient β determines the direction of the relationship between X and the logit of Y. The model has all significant variables (p < 0.05), and the odds ratio for each independent variable is included. The results are presented as an odds ratio with p-values and a 95% confidence interval.
Erreygers concentration index (ECI): measuring the wealth-based inequality in non-institutional deliveries
Concentration indices (CI) such as Generalized Concentration Index, Wagstaff Normalized Concentration Index [2], and ECI [31–33] are very popular and widely used in inequality analysis [34]. Since one of our objectives in this study is to measure the inequality (or unfair concentration) in place of delivery of births in reference to non-institutional delivery, a concentration index has been applied. We chose to apply the ECI in our study because it was most suitable for binary outcome variables [35]. The wealth factor score of households was used as the rank measure of socioeconomic status to calculate ECI. The wealth score was utilised from the NFHS-5 dataset, which was obtained by calculating a composite value for a set of household assets. The concentration curve (CC) plots the cumulative proportions of non-institutional deliveries against the cumulative proportions of the ranking variable, which is the household wealth (assets) score.
The standard concentration index (CI) can be computed as follows under the Erreygers Corrected Concentration Index (ECI) method.
CI = 
cov (yi, ri).
Where y is the variable of interest of the i individual, 
is the mean of yi, and ri is the fractional rank of the ith person [3]. Since our study’s ‘place of delivery’ variable is binary, a normalisation process is required to measure inequality. This study uses the ECI, which can be represented as:
ECI (y) = 
CI
Where 
is the mean of the place of delivery variable, CI is the standard concentration index, ymax is the upper bound limit of the variable, which is non-institutional deliveries, in this case, coded as ‘1’, and ymin is the lower bound limit (i.e., 0) of the variable [36]. The ECI value ranges from − 1 to + 1, but when y is a binary variable, as in our case, ECI ranges from 
to 1- 
[3]. The ECI would be zero when there is no wealth-based inequality in place of delivery, which means that non-institutional deliveries are equally distributed across the population. If the ECI is positive, it would mean that the non-institutional deliveries are more concentrated amongst the richer population, and it is negative when non-institutional deliveries are concentrated amongst the poorer population or are pro-poor [31, 37].
Decomposition of erreygers concentration index (ECI)
The ECI has been decomposed to identify the contribution of different background variables in the inequalities visible in place of delivery in reference to non-institutional deliveries. In order to estimate the contribution, we first need to estimate the coefficient of the predictor variables using a regression model. A Generalized Linear Model (GLM) with a binomial family and probit link is used as it is found to be the best choice for decomposing inequality in the case of a binary outcome variable [38]. GLM with a binomial distribution and probit link function preserved the non-linear relationship and distribution structure of the variable under consideration. The probit link was computed using the Stata command: ‘glm y x, family (binomial) link (probit)’.
The analysis utilised a probit regression model to estimate (1) Covariate-specific concentration indices, (2) Elasticity, which indicates the direction and magnitude of the relationship between each independent variable and the outcome variable, and (3) The percentage contribution of each covariate to overall inequality. Covariate-specific concentration indices show how a factor is distributed across socioeconomic variables. Elasticity provides insights into the likelihood of relationships between variables. For ratio-scale variables, elasticity allows for traditional economic interpretations, such as how the probability of non-institutional delivery changes when a covariate increases by 1%. For dummy variables, the elasticity sign indicates whether belonging to a specific category confers an advantage or disadvantage compared to a reference category. The covariate’s contribution to inequality in absolute terms is quantified by multiplying its specific concentration index by its elasticity. By dividing this absolute contribution by the total magnitude of inequality (i.e., measured by ECI), we obtain the percentage contribution of each covariate to overall inequality.
Stata 16 statistical software was used (StataCorp, 2019) for all the statistical analysis, including ECI and its decomposition [31, 39].
Results
Out of the total 1,75,569 deliveries reported (in reference period of last three years) in our sample, we observed that about 14% of them were non-institutional deliveries. The study identifies various demographic, household, and social characteristics that act as risk factors for non-institutional deliveries. Women in young age groups like 20–24 had the lowest proportion of non-institutional deliveries, but the proportion increased in the higher age brackets and was found to be the highest (36%) among women aged 45–49 years. Similarly, non-institutional deliveries were found to be more commonly occurring among uneducated women (26.22%), and the proportion declines as the educational level of women increases, reaching as low as 2.6% among women who had higher education. Many household factors influence the woman’s ability to get institutionalised deliveries.
Interestingly, the proportion of out-of-facility delivery was higher among women when the head of the household was a woman (17%) than when the household head was a man (13%). Non-institutional deliveries were more common in rural areas (15.38%) than in urban areas, which can be attributed to several social and infrastructural disadvantages in rural areas. Amongst the religious and caste groups, non-institutional deliveries were more prevalent among Christians (37%) and Scheduled Tribes (25%). The proportion of non-institutional deliveries was found to be the lowest among Sikh and other/general caste populations. Over 15% of all deliveries occurred out of facilities in the case of the couple who did not have health insurance. Furthermore, a majority of deliveries were non-institutional when pregnant women had their first ANC visit in the second trimester of their pregnancy.
Non-institutional deliveries are most prevalent in India’s northeastern (29%) region, followed by Eastern (18%). The proportion of non-institutional deliveries was lowest in the Southern region, where less than 2% of deliveries in the past three years were out of facilities. The central region has reported impressive changes in term of decline in non-instititutional deliveries than the rest, whereas north-eastern region has the poorest decline rate (see Supplementary file). It was also found to be most common among the poorest households. Around a third of all deliveries among the poorest households were non-institutional. The proportion of non-institutional deliveries goes as low as 2.5% in the richest households (Table 2). We also analysed certain conditions that women face when acquiring healthcare services for deliveries. Among such issues are “getting permission to go,” “getting the money needed for the treatment,” “distance to a health facility,” “having to take transport,” “not wanting to go alone,” and “concern of not having a female health care provider” acted as barriers for institutional deliveries. It was found that the proportion of non-institutional deliveries was highest among women who cited that these issues were a “big problem” for seeking health services.
Table 2.
Prevalence and predictors of non-institutional deliveries in the last three years preceding the survey, 2019-21 (NFHS-5)
| Background variables | Sample (N) 175,569 |
Non-institutional deliveries 24,050 (13.70%) |
Chi-square value | AOR (95% CI) |
|---|---|---|---|---|
| Age-groups | 1601.41*** | |||
| 15–19 | 5310 | 656 (12.35) | Ref. | |
| 20–24 | 58,689 | 6634 (11.30) | 0.841 (0.745–0.950) *** | |
| 25–29 | 69,299 | 9095 (13.12) | 0.704 (0.621–0.799) *** | |
| 30–34 | 29,402 | 4667 (15.87) | 0.546 (0.476–0.625) *** | |
| 35–39 | 10,300 | 2208 (21.44) | 0.473 (0.406–0.551) *** | |
| 40–44 | 2069 | 607 (29.34) | 0.522 (0.426–0.639) *** | |
| 45–49 | 500 | 183 (36.60) | 0.466 (0.334–0.650) *** | |
| Place of residence | 1778.75*** | |||
| Urban | 33,642 | 2217 (06.59) | Ref. | |
| Rural | 141,927 | 21,833 (15.38) | 1.012 (0.938–1.091) * | |
| Highest educational level | 9779.35*** | |||
| No education | 38,605 | 10,122 (26.22) | Ref. | |
| Primary | 22,564 | 4625 (20.50) | 0.818 (0.767–0.873) *** | |
| Secondary | 90,548 | 8682 (09.59) | 0.553 (0.522–0.585) *** | |
| Higher | 23,852 | 621 (02.60) | 0.334 (0.295–0.378) *** | |
| Sex of head of the household | 296.31*** | |||
| Male | 148,688 | 19,475 (13.10) | Ref. | |
| Female | 26,879 | 4575 (17.02) | 1.100 (1.039–1.165) *** | |
| Religion | 8070.35*** | |||
| Hindu | 129,137 | 13,825 (10.71) | Ref. | |
| Muslim | 25,481 | 4017 (15.76) | 1.487 (1.384–1.597) *** | |
| Christian | 14,183 | 5255 (37.05) | 2.802 (2.541–3.089) *** | |
| Sikh | 2944 | 115 (03.91) | 1.046 (0.800-1.367) | |
| Others | 3824 | 838 (21.91) | 1.468 (1.283–1.680) *** | |
| Caste categories | 4249.39*** | |||
| Scheduled Caste (SC) | 36,882 | 4579 (12.42) | Ref. | |
| Scheduled Tribe (ST) | 35,764 | 9130 (25.53) | 1.193 (1.112–1.280) *** | |
| Other Backward Classes | 67,697 | 6845 (10.11) | 0.866 (0.814–0.921) *** | |
| None of them (Others) | 26,466 | 2266 (08.56) | 0.971 (0.892–1.057) | |
| Wealth index | 12691.83*** | |||
| Poorest | 49,735 | 13,429 (27.00) | Ref. | |
| Poorer | 41,520 | 6011 (14.48) | 0.681 (0.645–0.720) *** | |
| Middle | 33,815 | 2763 (08.17) | 0.554 (0.515–0.596) *** | |
| Richer | 28,653 | 1305 (04.55) | 0.440 (0.400-0.485) *** | |
| Richest | 21,846 | 542 (02.48) | 0.330 (0.287–0.379) *** | |
| Timing of first ANC visit | 872.92*** | |||
| First trimester | 85,451 | 7180 (08.40) | Ref. | |
| Second trimester | 23,439 | 3484 (14.86) | 1.314 (1.252–1.380) *** | |
| Third trimester | 5098 | 560 (10.98) | 1.203 (1.087–1.332) *** | |
| Covered by health insurance | 170.96*** | |||
| No | 129,820 | 18,610 (14.34) | Ref. | |
| Yes | 45,749 | 5440 (11.89) | 0.761 (0.721–0.803) *** | |
| Birth order | 8793.88*** | |||
| 1st | 73,445 | 5597 (07.62) | Ref. | |
| 2nd & 3rd | 80,705 | 11,682 (14.47) | 2.078 (1.956–2.208) *** | |
| 4th & 5th | 16,851 | 4804 (28.51) | 3.375 (3.110–3.663) *** | |
| Over 5th | 4568 | 1967 (43.06) | 5.079 (4.509–5.721) *** | |
| Mass media exposure | 6995.34*** | |||
| No | 53,820 | 12,928 (24.02) | Ref. | |
| Yes | 121,749 | 11,122 (09.14) | 0.785 (0.747–0.825) *** | |
| Getting permission to go | 170.64*** | |||
| No problem | 104,829 | 13,600 (12.97) | Ref. | |
| Little problem | 41,289 | 5765 (13.96) | 0.925 (0.871–0.982) ** | |
| Big problem | 29,451 | 4685 (15.91) | 0.871 (0.812–0.933) *** | |
| Getting money needed for treatment | 2252.47*** | |||
| No problem | 75,932 | 7523 (09.91) | Ref. | |
| Little problem | 56,652 | 8052 (14.21) | 0.984 (0.926–1.047) | |
| Big problem | 42,985 | 8475 (19.72) | 0.968 (0.902–1.038) | |
| Distance to the health facility | 2576.71*** | |||
| No problem | 62,998 | 5713 (09.07) | Ref. | |
| Little problem | 62,872 | 8626 (13.72) | 1.147 (1.064–1.237) *** | |
| Big problem | 49,699 | 9711 (19.54) | 1.285 (1.174–1.407) *** | |
| Having to take transport | 2848.92*** | |||
| No problem | 66,589 | 6086 (09.14) | Ref. | |
| Little problem | 61,761 | 8435 (13.66) | 1.021 (0.946–1.101) | |
| Big problem | 47,219 | 9529 (20.18) | 1.174 (1.071–1.286) *** | |
| Not wanting to go alone | 1196.60*** | |||
| No problem | 78,209 | 8431 (10.78) | Ref. | |
| Little problem | 61,417 | 9157 (14.91) | 1.029 (0.969–1.092) | |
| Big problem | 35,943 | 6462 (17.98) | 0.993 (0.919–1.072) | |
| Concern no female health provider | 810.51*** | |||
| No problem | 59,705 | 6348 (10.63) | Ref. | |
| Little problem | 55,051 | 7853 (14.26) | 0.987 (0.929–1.050) | |
| Big problem | 60,813 | 9849 (16.20) | 0.972 (0.911–1.036) | |
| Region | 10305.42*** | |||
| Northern | 31,705 | 2258 (07.12) | Ref. | |
| Central | 47,478 | 6791 (14.30) | 1.981 (1.815–2.164) *** | |
| Eastern | 35,111 | 6503 (18.52) | 1.848 (1.685–2.026) *** | |
| Western | 15,220 | 949 (06.24) | 0.928 (0.817–1.055) * | |
| Southern | 21,590 | 416 (01.93) | 0.358 (0.303–0.423) *** | |
| North-eastern | 24,465 | 7133 (29.16) | 2.053 (1.834–2.297) *** | |
|
AOR = Adjusted Odds Ratio, CI = Confidence Interval, Ref. = Reference Category*** p < 0.01, ** p < 0.05, * p < 0.1 Data Source: National Family Health Survey (NFHS-5), 2019-21, India | ||||
Correlates and risk factors of non-institutional deliveries
Table 2 presents the results of a multivariate logistic regression analysis focused on the factors associated with non-institutional deliveries in a given population. This highlights that compared to teenage [15–19] mothers, the probability of non-institutional deliveries decreases in all age groups. There was about 16-53% less likelihood of non-institutional deliveries with increasing age, and it was found to be lowest in age group of 45–49 years (AOR = 0.466, 95% CI = 0.334–0.650). A gradual decrease in the likelihood of non-institutional deliveries was observed with women’s increasing education level. It was also observed that non-institutional delivery was about 19-67% less likely among mothers who had primary (AOR = 0.818, 95% CI = 0.767–0.873), secondary (AOR = 0.553, 95% CI = 0.522–0.585) or higher (AOR = 0.334, 95% CI = 0.295–0.378) education as compared to uneducated mothers. It was found that compared to households where the head was a male, there was about 10% more likelihood of non-institutional deliveries when the head was a female (AOR = 1.100, 95% CI = 1.039–1.165). Compared to Hindus, there was about 48-280% more likelihood of non-institutional deliveries among Muslims (AOR = 1.487, 95% CI = 1.384–1.597) and Christians (AOR = 2.802, 95% CI = 2.541–3.089). Amongst caste categories, ST women (AOR = 1.193, 95% CI = 1.112–1.280) were 19% more likely to have non-institutional deliveries as compared to SC, but OBC women (AOR = 0.866, 95% CI = 0.814–0.921) were up to 13% less likely to have non-institutional deliveries. In contrast to the northern region of India, central (AOR = 1.981, 95% CI = 1.815–2.164), eastern (AOR = 1.848, 95% CI = 1.685–2.026), and north-eastern (AOR = 2.053, 95%CI = 1.834–2.297) regions had 80-100% more likelihood of pregnant women having non-institutional deliveries whereas western (AOR = 0.928, 95% CI 0.817–1.055) and southern (AOR = 0.358. 95% CI 0.303–0.423) region were 8-65% less likely for the same.
Most significantly, the odds of having non-institutional deliveries decrease with increasing wealth quintile. It was found that richer (AOR = 0.440, 95% CI = 0.400-0.485) and richest (AOR = 0.330, 95% CI = 0.287–0.379) households were nearly 56-77% less likely to have non-institutional deliveries compared to the poorest people. Similarly, there was about 24% less likelihood of non-institutional deliveries if the woman/couple had health insurance (AOR = 0.761, 95% CI = 0.721–0.803) than those without health insurance. There was about 15-29% more likelihood of non-institutional deliveries if the pregnant woman cited distance to the health facility as an issue. Exposure to mass media was found to be a protective factor for non-institutional delivery. Women who had access to mass media (AOR = 0.785, 95% CI = 0.747–0.825) were about 22% less likely to have non-institutional delivery than women who had no exposure to mass media.
Those who had registered for ANC in their second (AOR = 1.314, 95%CI = 1.252–1.380) or third (AOR = 1.203, 95% CI = 1.087–1.332) trimester were 20-30% more likely to have non-institutional deliveries as compared to those women who got early registration in their first trimester of pregnancy. Our study also reveals 2 to 5 times higher odds of non-institutional deliveries if the occurring birth was not the first child. The probability of non-institutional deliveries increases as the birth order increases, such as about five times more chances of out-of-facility delivery if women were to deliver their fifth child.
Concentration curve (CC) showing wealth-based inequality in non-institutional deliveries
The concentration curve (CC) plot demonstrates an unfair concentration of non-institutional deliveries among women from poor households, as shown in Fig. 2. It shows the cumulative proportion of non-institutional deliveries against the wealth factor score of households. The CC lies above the line of equality, indicating that non-institutional deliveries were more prevalent among poor households. The CC plot is directly related to the concentration index, which represents twice the area between the CC and the 45-degree equality line [37]. In Table 3, the value of ECI was negative (-0.2174; p-value < 0.001), indicating that poor women were significantly more likely to have non-institutional deliveries.
Fig. 2.
Concentration curve (CC): presenting the distribution of non-institutional deliveries in relatively poor households in India. Data Source: National Family Health Survey (NFHS-5), 2019-21, India
Table 3.
Concentration index and decomposition analysis results: contributing factors of wealth-based inequality in non-institutional deliveries
| Category | Elasticity | Concentration Index | Absolute Contribution | % Contribution | Total % Contribution | ||
|---|---|---|---|---|---|---|---|
| Age-groups | |||||||
| 15–19 | Ref. | 0 | -0.0106 | 0 | 0 | ||
| 20–24 | -0.0121 | 0.0096 | -0.0001 | 0.0534 | |||
| 25–29 | -0.0388 | 0.0721 | -0.0028 | 1.287 | |||
| 30–34 | -0.0412 | 0.0065 | -0.0003 | 0.1234 | |||
| 35–39 | -0.0236 | -0.0427 | 0.001 | -0.464 | |||
| 40–44 | -0.0054 | -0.0234 | 0.0001 | -0.058 | |||
| 45–49 | -0.0018 | -0.0114 | 0 | -0.0096 | 0.9322 | ||
| Place of residence | |||||||
| Urban | Ref. | 0 | 0.3753 | 0 | 0 | ||
| Rural | 0.0029 | -0.3753 | -0.0011 | 0.498 | 0.498 | ||
| Highest educational level | |||||||
| No education | Ref. | 0 | -0.4077 | 0 | 0 | ||
| Primary | -0.0092 | -0.1092 | 0.001 | -0.4632 | |||
| Secondary | -0.0868 | 0.2561 | -0.0222 | 10.2242 | |||
| Higher | -0.0341 | 0.2609 | -0.0089 | 4.0965 | 13.8575 | ||
| Sex of head of the household | |||||||
| Male | Ref. | 0 | 0.0394 | 0 | 0 | ||
| Female | 0.0046 | -0.0394 | -0.0002 | 0.0824 | 0.0824 | ||
| Religion | |||||||
| Hindu | Ref. | 0 | -0.0043 | 0 | 0 | ||
| Muslim | 0.0185 | 0.0482 | 0.0009 | -0.4112 | |||
| Christian | 0.0323 | -0.0685 | -0.0022 | 1.0161 | |||
| Sikh | 0.0003 | 0.0388 | 0 | -0.0046 | |||
| Other | 0.0028 | -0.0142 | 0 | 0.0183 | 0.6186 | ||
| Caste categories | |||||||
| Scheduled Caste | Ref. | 0 | -0.0541 | 0 | 0 | ||
| Scheduled Tribe | 0.0122 | -0.2862 | -0.0035 | 1.6028 | |||
| OBC | -0.017 | 0.1423 | -0.0024 | 1.1132 | |||
| None of them (Other) | -0.0012 | 0.198 | -0.0002 | 0.1134 | 2.8294 | ||
| Wealth index | |||||||
| poorest | Ref. | 0 | -0.8703 | 0 | 0 | ||
| poorer | -0.0296 | -0.1159 | 0.0034 | -1.5782 | |||
| middle | -0.033 | 0.2179 | -0.0072 | 3.3086 | |||
| richer | -0.0352 | 0.3733 | -0.0131 | 6.0445 | |||
| richest | -0.0338 | 0.395 | -0.0134 | 6.1393 | 13.9142 | ||
| Timing of first ANC visit | |||||||
| First trimester | Ref. | 0 | 0.1389 | 0 | 0 | ||
| Second trimester | 0.0177 | -0.1311 | -0.0023 | 1.0665 | |||
| Third trimester | 0.0026 | -0.0078 | 0 | 0.0094 | 1.0759 | ||
| Covered by health insurance | |||||||
| No | Ref. | 0 | 0.0096 | 0 | 0 | ||
| Yes | -0.0233 | -0.0096 | 0.0002 | -0.1028 | -0.1028 | ||
| Birth order | |||||||
| 1st | Ref. | 0 | 0.1706 | 0 | 0 | ||
| 2nd & 3rd | 0.0912 | -0.0357 | -0.0033 | 1.4958 | |||
| 4th & 5th | 0.0328 | -0.0966 | -0.0032 | 1.4571 | |||
| Over 5th | 0.0124 | -0.0383 | -0.0005 | 0.2185 | 3.1714 | ||
| Mass media exposure | |||||||
| No | Ref. | 0 | -0.5427 | 0 | 0 | ||
| Yes | -0.0492 | 0.5427 | -0.0267 | 12.2775 | 12.2775 | ||
| Getting permission to go | |||||||
| No problem | Ref. | 0 | 0.1195 | 0 | 0 | ||
| Little problem | -0.0057 | -0.0389 | 0.0002 | -0.1015 | |||
| Big problem | -0.0071 | -0.0806 | 0.0006 | -0.2636 | -0.3651 | ||
| Getting money needed for treatment | |||||||
| No problem | Ref. | 0 | 0.3084 | 0 | 0 | ||
| Little problem | -0.0019 | -0.0643 | 0.0001 | -0.0566 | |||
| Big problem | -0.0023 | -0.244 | 0.0006 | -0.2543 | -0.3109 | ||
| Distance to a health facility | |||||||
| No problem | Ref. | 0 | 0.2968 | 0 | 0 | ||
| Little problem | 0.0141 | -0.0615 | -0.0009 | 0.4 | |||
| Big problem | 0.022 | -0.2353 | -0.0052 | 2.3749 | 2.7749 | ||
| Having to take transport | |||||||
| No problem | Ref. | 0 | 0.3218 | 0 | 0 | ||
| Little problem | 0.0016 | -0.0726 | -0.0001 | 0.054 | |||
| Big problem | 0.0136 | -0.2492 | -0.0034 | 1.5559 | 1.6099 | ||
| Not wanting to go alone | |||||||
| No problem | Ref. | 0 | 0.246 | 0 | 0 | ||
| Little problem | 0.0036 | -0.0971 | -0.0003 | 0.16 | |||
| Big problem | -0.0002 | -0.1489 | 0 | -0.0146 | 0.1454 | ||
| Concern no female health provider | |||||||
| No problem | Ref. | 0 | 0.2142 | 0 | 0 | ||
| Little problem | -0.0005 | -0.0492 | 0 | -0.0113 | |||
| Big problem | -0.0021 | -0.165 | 0.0004 | -0.1616 | -0.1729 | ||
| Region | |||||||
| Northern | Ref. | 0 | 0.2217 | 0 | 0 | ||
| Central | 0.0562 | -0.0675 | -0.0038 | 1.7444 | |||
| Eastern | 0.0347 | -0.2519 | -0.0088 | 4.0252 | |||
| Western | -0.0026 | 0.0681 | -0.0002 | 0.0827 | |||
| Southern | -0.0276 | 0.1595 | -0.0044 | 2.0271 | |||
| North-eastern | 0.0316 | -0.1299 | -0.0041 | 1.887 | 9.7664 | ||
| Total | -0.1363 | 62.602 | |||||
| Residual | 37.398 | ||||||
| Index: Erreygers Normalized Concentration Index | |||||||
| No. of obs. | 175,569 | Std. error | 0.0018 | ||||
| Index value | -0.2174 | p-value | < 0.0001 | ||||
| Source: National Family Health Survey (NFHS-5), 2019-21, India | |||||||
Decomposing the wealth-based inequality in place of delivery (in reference to non-institutional delivery)
We conducted a decomposition analysis to determine the factors contributing to the unfair concentration of non-institutional deliveries among poor households in India, as presented in Table 3. The table shows the elasticity (product of coefficient and mean of each background variable), concentration index of each explanatory variable, and the absolute and percentage contribution of these factors in inequality in place of delivery in reference to non-institutional deliveries. A positive value of elasticity represents the variable that causes a higher likelihood of non-institutional deliveries.
Our results show that the wealth index contributes about 13.91% of the inequality, followed by the mother’s educational level, mass media exposure, and regional location, which explains about 13.85%, 12.27%, and 9.76% of the inequality in place of delivery, respectively. The Eastern region alone contributes to 4% of inequality. Together, these four factors contributed most to the inequality in ‘place of delivery’ across India. The model explains about 62.60% of the inequality in place of delivery in India. The other factors that significantly contribute into inequality in non-institutional deliveries were the mother’s childbirth order number (3.17%), caste (2.82%), consideration of distance to health facility as a problem (2.77%), timing of first ANC visit (1.07%), and consideration of having to take transport as a problem (1.60%).
About 37% of the inequality remains unexplained by the variables we have incorporated in this model. The residual part of inequality refers to the portion of the inequality in the place of delivery that cannot be explained by the systematic variation across the background variables used in this analysis and thus cannot be decomposed. It can be assumed that there must be other variables responsible for this unexplained part of inequality, although we used most of the variables available in the birth recode data file that might potentially affect the non-institutional deliveries. Some other socioeconomic or health infrastructure-related factors might affect this inequality, which remains unexplored in this dataset. Residuals also contributed to some extent to the pro-poor inequality.
Discussion
The current study presents a comprehensive analysis of non-institutional deliveries in India. Firstly, the prevalence and predictors of non-institutional deliveries in India were examined. Further, using the Erreygers Concentration Index, wealth-based inequality in non-institutional deliveries was measured, revealing that such deliveries in India are concentrated among poor households. This warranted a decomposition analysis to discover the contributing factors of the assessed inequality in non-institutional deliveries in India. Education, wealth status, and exposure to mass media were found to be major contributing factors to inequality in non-institutional deliveries in India, along with a range of other variables such as timing of the first antenatal care (ANC) visit, birth order, distance and transport-related issues, region of residence, and social factors like religion and caste.
Education was found to be one the biggest contributors to the inequality, contributing 14% alone. Non-institutional delivery was found to be most prevalent among uneducated women, and its odds were found to be decreasing with increasing educational level. Previously, education and wealth have been found to be major predictors of place of delivery in various studies [2, 5, 16, 17, 19, 24, 28, 40–43], similar to ours. It suggests that education and wealth can be major factors in closing the gap in institutional delivery utilisation. Education can play a crucial role in raising awareness about the benefits of skilled birth attendance and the risks associated with home deliveries [42, 43]. Sahoo et al. establish that educated couples are more open to opting for modern medicines and trained healthcare providers [44]. Similarly, exposure to mass media was another significant contributor (12.27%) to inequality. The probability of non-institutional delivery was significantly low among women who had mass media exposure. Means of mass media such as mobile phones and television can be utilised by policymakers and governments to spread awareness on various health concerns [45–48]. Awareness regarding policy and programmes promoting institutional deliveries such as Janani Suraksha Yojana (JSY) incentives, ASHA workers’ role in ANC, and positive outcomes of institutional deliveries on such media platforms, must reach out to the last mile population. It is evident from our findings that exposure to mass media can be one of the important factors in reducing the inequality in place of delivery. This is particularly relevant in the context of population living at margins, in rural and tribal areas where the level of education is generally lower and traditional practices are more prevalent. Table 4 presents reasons cited by women for opting for non-institutional deliveries, and a considerable proportion of women, specifically 28%, expressed that delivering in a health facility is simply not necessary, highlighting the lack of awareness. Women with lower educational levels and those from poorer economic backgrounds were more inclined towards non-institutional deliveries.
Table 4.
Reasons for not delivering in a health facility: Percentage of women aged 15–49 who had a live birth in the five years preceding the survey by reasons for not delivering the most recent live birth in a health facility
| Reason for not delivering in the health facility | Total |
|---|---|
| Costs too much | 15.1 |
| Facility not open | 9.7 |
| Too far/ no transportation | 16.6 |
| Don’t trust facility/ poor quality service | 5.0 |
| No female provider at the facility | 4.0 |
| Husband/ family did not allow | 19.3 |
| Not necessary | 28.1 |
| Not customary | 3.6 |
| Other | 16.8 |
| Note: Percentages may not add up to 100 because multiple responses were permitted. Number of Women = 17,016. | |
| Data Source: National Family Health Survey (NFHS-5), 2019-21; India Report (14) | |
Among all independent variables included in the analysis wealth index was found to be the largest contributor to non-institutional delivery inequality. It was also found that higher wealth at household level lowers the odds of non-institutional delivery among pregnant women in India, as its highest prevalence was reported among women of the poorest households and the lowest among the wealthiest households. The Indian government has long promoted institutional deliveries through health policies and programs. In particular, the Janani Suraksha Yojana (JSY) scheme offering incentives to mothers and ASHA workers for institutional delivery [27] and Janani Shishu Suraksha Karyakram (JSSK) to assure cashless institutional delivery to pregnant women including free transport and diet [49], have encouraged institutional deliveries and improved maternal health indicators significantly [50, 51]. As a result, non-institutional deliveries have decreased substantially in the past decade. Nevertheless, a substantial number of women in India still had to go through non-institutional deliveries. Our findings, aligned with those of Goli et al. (2021), indicate that out-of-pocket expenditure (OOPE) for institutional deliveries remains a substantial barrier [52]. An exploration of reasons for non-institutional deliveries (Table 4) corroborates the same, where ‘cost’ was cited as the prime reason by 15.1% of women for not delivering in a health facility. Previous research in lower- and middle-income countries consistently highlighted wealth as a key factor influencing the place of deliveries [12, 22, 29, 32, 33, 53–55]. Similar reasons can be attributed to late ANC visits, which further increases the odds of non-institutional delivery. It was also observed in our analysis that late ANC contributed about 1% to the non-institutional delivery inequality. Among other socio-demographic dimensions, higher childbirth orders was observed to be associated with out of facility delivery. It was found that children of higher birth order after 2nd child had increasing odds of being delivered out of a health facility, and it contributed about 3% to the existing inequality. OOPE of the delivery at institutional facilities, as mentioned earlier, can be inferred as one of the major influencing reasons for such a pattern. Non-institutional deliveries may also be further linked to the the availability and accessibility of healthcare services, especially in rural and remote areas. In this context, our study aligns with the findings of Srivastava and Joseph [43, 56], which highlighted that distance to health facilities and lack of transportation are significant barriers to institutional delivery. In our study, the issue of the distance between health facilities and transportation contributed about 4% to the existing inequality, increasing the odds of non-institutional delivery among women consistent with other studies [2]. In all, 16.6% of women cited distance or lack of transportation as one of the primary reasons for not being able to deliver at a health facility (Table 4).
On a similar trend, women from rural areas were more likely to have non-institutional deliveries compared to their urban counterparts [15, 57]. This disparity could stem from factors such as limited healthcare infrastructure, lesser accessibility to healthcare facilities, and ingrained cultural norms in rural areas. Regionally, Empowered Action Group (EAG) states in India, a geographical demarcation of population with poor socioeconomic indicators, have extreme disparities in access to healthcare facilities [58], which also translates into spatial concentration of health outcomes inequality. The eastern and north-eastern states of India were reported to have higher odds of non-institutional deliveries possibly due to the challenging terrain and poor transportation networks. This compounds into inadequate access to health care facilities and reflects into geographical region contributing to about 10% in the observed inequality of non-institutional delivery. Total 14 districts spread across Meghalaya, Nagaland, and Manipur states reported over 50% of births being non-institutional, and the majority of the districts in north-eastern India have reported higher non-institutional delivery rates than the national average of 11%. As shown in Fig. 3, there is a visible concentration of higher rates of non-institutional deliveries in the north-eastern and eastern states of India, which demands further research to identify the factors resulting in such high concentration.
Fig. 3.
Distribution of non-institutional deliveries in India. Data Source: National Family Health Survey (NFHS-5), 2019-21, India
It is evident that there is a complex interplay of multitudes of variables that contribute to the existing unfair concentration of remaining non-institutional deliveries among poor households and resultant inequality in India. Besides, as reasons for not delivering in a health facility, women have reported various social and familial restrictions and perceptions that lead to non-institutional deliveries. As reported in Tables 1 and 49.3% of women reported facing restrictions from their husbands or families, thereby hindering their ability to opt for a healthcare facility delivery. Additionally, a smaller yet noteworthy percentage of women (5%) did not trust the facility or showed dissatisfaction with the quality of healthcare services provided therein.
Addressing the observed unfair concentration and inequality of non-institutional deliveries requires a holistic approach that combines financial support, educational interventions, improvements in healthcare infrastructure, and sensitisation on cultural and social norms. By focusing on these areas, policies and programs can be more effectively tailored to meet the needs of the most vulnerable segments of the population, thereby moving closer to the goal of universal institutional deliveries. Moreover, our study suggests that future research should delve deeper into the sociological, economic, and geographical aspects of non-institutional deliveries. Understanding the root causes of these deliveries, particularly among the poorest and most marginalised groups, is crucial for designing targeted interventions. Policymakers have to now realize that the last 10% of the non-institutional deliveries may be concentrated to extremely vulnerable and hard to reach population and the existing policy measures may not be effective. The population in which these last-mile non-institutional deliveries are concentrated may have a distinct set of barriers and bottlenecks. It is, therefore, imperative to recognise that conventional policy tools, while successful to a certain extent, may not be entirely effective in their current forms for addressing the remaining cases of non-institutional deliveries their unfair concentration and inequality. This calls for decentralized innovative strategies and local interventions that can adapt to the population specific maternal healthcare needs in India. Longstanding legacy factors such as the unavailability of health facilities, high costs due to travel and transportation, and cultural preferences also warrant a revisit in the revised approach to policy-making aimed at achieving universal institutional deliveries in India.
Strengths and limitations
The study uses a large-scale representative sample distributed well across the country, enabling us to generalise the findings nationally. Our study uses the most recent and reliable data available to understand the wealth-based concentration of non-institutional deliveries in India and its contributing factors. To the best of our knowledge, this is the first study in India that examines the determinants of non-institutional deliveries, their concentrations, and factors associated with this concentration (inequality) in India. Unlike many existing studies on institutional deliveries, this research presents a comprehensive analysis of last-mile non-institutional deliveries for their targeting and elimination. The study is based on secondary data, i.e., NFHS-5, and thus may have missed a few relevant variables (due to their non-inclusion in the data set) contributing to the non-institutional delivery and its concentration.
Conclusion
The study reveals that 14% of Indian women had non-institutional deliveries in the past three years. The analysis in the study suggests a significant inequality in the non-institutional deliveries in the form of its unfair concentration among the households at the socioeconomic margins in India. It also recognises some of the most prominent factors explaining the concentration among underprivileged women. Non-institutional deliveries are concentrated in poorer households, and such concentration is majorly contributed by the wealth of the households, the education of mothers, exposure to mass media, and their regional location. This study also provides several other contributing factors of inequality in non-institutional deliveries: birth order, distance to health facilities, caste, transport as an issue, and late antenatal care. To achieve universal institutional delivery and safe childbirth in India, we recommend realigning policy measures to focus on and last-mile non-institutional deliveries by carefully identifying their socioeconomic context and geography. It is only by ensuring zero non-institutional deliveries and improved maternal and child care for all that India would achieve the ideals of health equity, social justice, and human rights.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Acknowledgements
NA.
Author contributions
RC, SM, and AS conceptualised the idea. RC, JP and SM developed the final manuscript. JP and SS did the statistical analysis and helped in writing the first draft. AS and SS reviewed the analysis and helped in editing.
Funding
None.
Data availability
All the datasets processed and analysed for the current study are publicly available on the DHS webpage and can be accessed from https://dhsprogram.com/data/available-datasets.cfm.
Declarations
Ethics approval and consent to participate
All the methods were carried out in accordance with relevant guidelines and regulations.
Consent for publication
NA.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Biswas A, Anderson R, Doraiswamy S, Abdullah ASM, Purno N, Rahman F et al. Timely referral saves the lives of mothers and newborns: Midwifery led continuum of care in marginalized teagarden communities - A qualitative case study in Bangladesh. F1000Research [Internet]. 2018 Mar 23 [cited 2022 Aug 15];7:365. Available from: https://f1000research.com/articles/7-365 [DOI] [PMC free article] [PubMed]
- 2.Patel R, Marbaniang SP, Srivastava S, Kumar P, Chauhan S. Why women choose to deliver at home in India: a study of prevalence, factors, and socio-economic inequality. BMC Public Health [Internet]. 2021 Dec 1 [cited 2022 Aug 7];21(1):1–14. Available from: https://bmcpublichealth.biomedcentral.com/articles/10.1186/s12889-021-11779-5 [DOI] [PMC free article] [PubMed]
- 3.El Taleb MV, Cabases JM, Zine Eddine El Idrissi MD, Mills S. Changes in Inequality in Use of Maternal Health Care Services: Evidence from Skilled Birth Attendance in Mauritania for the Period 2007–2015. Int J Environ Res Public Health [Internet]. 2022;19(6):3566. Available from: https://www.mdpi.com/1660-4601/19/6/3566 [DOI] [PMC free article] [PubMed]
- 4.Kumar R, Mandava S. Institutional deliveries in India: a study of associates and inequality. Int J Soc Econ. 2022;49(5):726–43. [Google Scholar]
- 5.Devkota B, Maskey J, Pandey AR, Karki D, Godwin P, Gartoulla P et al. Determinants of home delivery in Nepal - A disaggregated analysis of marginalised and non-marginalised women from the 2016 Nepal demographic and health survey. PLoS ONE. 2020;15(1). [DOI] [PMC free article] [PubMed]
- 6.World Health Organization. Trends in maternal mortality 2000 to 2020: estimates by WHO, UNICEF, UNFPA, World Bank Group and UNDESA/Population Division [Internet]., Geneva: World Health Organization; 2023. Available from: https://www.who.int/publications/i/item/9789240068759
- 7.Abebe F, Berhane Y, Girma B. Factors associated with home delivery in Bahirdar, Ethiopia: A case control study. BMC Res Notes [Internet]. 2012 Nov 24 [cited 2022 Aug 15];5(1):1–6. Available from: https://bmcresnotes.biomedcentral.com/articles/10.1186/1756-0500-5-653 [DOI] [PMC free article] [PubMed]
- 8.Godara R, Garg P, Verma S, Gupta S. Disembowelment following unsafe home delivery: Young life lost! Trop Doct [Internet]. 2017 Jan 1 [cited 2022 Aug 7];47(1):56–8. Available from: https://journals.sagepub.com/doi/10.1177/0049475516632403 [DOI] [PubMed]
- 9.Zakir M, Wunnava PV. Factors affecting infant mortality rates: evidence from cross-sectional data. Appl Econ Lett. 1999;6(5):271–3. [Google Scholar]
- 10.Meltem Daysal N, Trandafir M, Van Ewijk R. Saving lives at birth: The impact of home births on infant outcomes. Am Econ J Appl Econ [Internet]. 2015 [cited 2022 Aug 15];7(3):28–50. Available from: 10.1257/app.20120359
- 11.Nwakoby BN. Use of Obstetric Services in Rural Nigeria. J R Soc Promot Health [Internet]. 1994 Sep 7 [cited 2022 Aug 15];114(3):132–6. Available from: https://journals.sagepub.com/doi/10.1177/146642409411400304 [DOI] [PubMed]
- 12.Joseph G, Da Silva ICM, Wehrmeister FC, Barros AJD, Victora CG. Inequalities in the coverage of place of delivery and skilled birth attendance: analyses of cross-sectional surveys in 80 low and middle-income countries. Reprod Health [Internet]. 2016 Jun 17 [cited 2023 Jun 6];13(1). Available from: /pmc/articles/PMC4912761/. [DOI] [PMC free article] [PubMed]
- 13.Winter R, Yourkavitch J, Wang W, Mallick L. Assessment of health facility capacity to provide newborn care in Bangladesh, Haiti, Malawi, Senegal, and Tanzania. J Glob Health [Internet]. 2017 Dec [cited 2023 Jun 9];8(1). Available from: www.jogh.org. [DOI] [PMC free article] [PubMed]
- 14.Moindi RO, Ngari MM, Nyambati VCS, Mbakaya C. Why mothers still deliver at home: Understanding factors associated with home deliveries and cultural practices in rural coastal Kenya, a cross-section study Global health. BMC Public Health [Internet]. 2016 Feb 3 [cited 2022 Aug 15];16(1):1–8. Available from: https://bmcpublichealth.biomedcentral.com/articles/10.1186/s12889-016-2780-z. [DOI] [PMC free article] [PubMed]
- 15.Yadav AK, Jena PK. Explaining changing patterns and inequalities in maternal healthcare services utilization in India. J Public Aff [Internet]. 2022 [cited 2022 Aug 15];22(3):e2570. Available from: https://onlinelibrary.wiley.com/doi/full/10.1002/pa.2570
- 16.Devasenapathy N, George MS, Jerath SG, Singh A, Negandhi H, Alagh G et al. Why women choose to give birth at home: A situational analysis from urban slums of Delhi. BMJ Open [Internet]. 2014 May 1 [cited 2022 Aug 15];4(5):e004401. Available from: https://bmjopen.bmj.com/content/4/5/e004401 [DOI] [PMC free article] [PubMed]
- 17.Ou CY, Yasmin M, Ussatayeva G, Lee MS, Dalal K. Maternal Delivery at Home: Issues in India. Adv Ther [Internet]. 2021 Jan 30 [cited 2022 Aug 15];38(1):386–98. Available from: https://link.springer.com/article/10.1007/s12325-020-01551-3. [DOI] [PMC free article] [PubMed]
- 18.Xu Y, Peng MYP, Ahuru RR, Anser MK, Osabohien R, Aziz A. Individual and community-level factors associated with non-institutional delivery of women of childbearing-age in Nigeria. Humanit Soc Sci Commun [Internet]. 2022 Jun 15 [cited 2022 Jul 16];9(1):1–7. Available from: https://www.nature.com/articles/s41599-022-01168-7
- 19.Kitui J, Lewis S, Davey G. Factors influencing place of delivery for women in Kenya: An analysis of the Kenya demographic and health survey, 2008/2009. BMC Pregnancy Childbirth [Internet]. 2013 Dec 17 [cited 2022 Jul 16];13(1):40. Available from: https://bmcpregnancychildbirth.biomedcentral.com/articles/10.1186/1471-2393-13-40. [DOI] [PMC free article] [PubMed]
- 20.Yadav AK, Jena PK. Maternal health outcomes of socially marginalized groups in India. Int J Health Care Qual Assur [Internet]. 2020 Feb 28 [cited 2022 Aug 15];33(2):172–88. Available from: https://www.emerald.com/insight/content/doi/10.1108/IJHCQA-08-2018-0212/full/html. [DOI] [PubMed]
- 21.Sarkar P, Tigga NS. Social Inequality and Institutional Deliveries. Soc Change [Internet]. 2018 Mar 1 [cited 2022 Aug 15];48(1):85–103. Available from: https://journals.sagepub.com/doi/10.1177/0049085717743840
- 22.Randive B, San Sebastian M, De Costa A, Lindholm L. Inequalities in institutional delivery uptake and maternal mortality reduction in the context of cash incentive program, Janani Suraksha Yojana: Results from nine states in India. Soc Sci Med [Internet]. 2014 Dec 1 [cited 2022 Sep 19];123:1–6. Available from: https://pubmed.ncbi.nlm.nih.gov/25462599/ [DOI] [PubMed]
- 23.Konje ET, Magoma MTN, Hatfield J, Kuhn S, Sauve RS, Dewey DM. Missed opportunities in antenatal care for improving the health of pregnant women and newborns in Geita district, Northwest Tanzania. BMC Pregnancy Childbirth. 2018;18(1). [DOI] [PMC free article] [PubMed]
- 24.Hazarika I. Factors that determine the use of skilled care during delivery in India: Implications for achievement of MDG-5 targets. Matern Child Health J [Internet]. 2011 Sep 22 [cited 2022 Aug 15];15(8):1381–8. Available from: https://link.springer.com/article/10.1007/s10995-010-0687-3. [DOI] [PubMed]
- 25.Sudhinaraset M, Treleaven E, Melo J, Singh K, Diamond-Smith N. Women’s status and experiences of mistreatment during childbirth in Uttar Pradesh: A mixed methods study using cultural health capital theory. BMC Pregnancy Childbirth [Internet]. 2016 Oct 28 [cited 2023 Jun 9];16(1):1–12. Available from: https://bmcpregnancychildbirth.biomedcentral.com/articles/10.1186/s12884-016-1124-4. [DOI] [PMC free article] [PubMed]
- 26.Nandan D. National Rural Health Mission. In: Public health in developing countries [Internet]. Woodhead Publishing Limited; 2011 [cited 2022 Aug 15]. pp. 1195–213. Available from: https://nhm.gov.in/index1.php?lang=1_level=1_sublinkid=969_lid=49
- 27.GoI. National Health Mission Department of Health & Family Welfare, Ministry of Health & Family Welfare, Government of India. 2005 [cited 2023 May 23]. Janani Suraksha Yojana:: National Health Mission. Available from: https://www.nhm.gov.in/index1.php?lang=1_level=3_sublinkid=841_lid=309
- 28.Sarkar A, Kharmujai O, Lynrah W, Suokhrie N. Factors influencing the place of delivery in rural Meghalaya, India: A qualitative study. J Fam Med Prim Care [Internet]. 2018 [cited 2022 Aug 15];7(1):98. Available from: https://journals.lww.com/jfmpc/Fulltext/2018/07010/Factors_influencing_the_place_of_delivery_in_rural.18.aspx [DOI] [PMC free article] [PubMed]
- 29.Mukherjee S, Singh A. Has the Janani Suraksha Yojana (A Conditional Maternity Benefit Transfer Scheme) Succeeded in Reducing the Economic Burden of Maternity in Rural India? Evidence from the Varanasi District of Uttar Pradesh. J Public Health Res [Internet]. 2018 Feb 5 [cited 2023 Jun 6];7(1):jphr.2018.957. Available from: /pmc/articles/PMC5941261/. [DOI] [PMC free article] [PubMed]
- 30.International Institute for Population Sciences (IIPS) & ICF. National Family Health Survey (NFHS-5) 2019-21: India [Internet]. Mumbai, India: IIPS. 2021 [cited 2023 May 5]. Available from: http://www.rchiips.org/nfhs
- 31.O’Donnell O, O’Neill S, Van Ourti T, Walsh B, conindex. Estimation of concentration indices. Stata J [Internet]. 2016 Mar 1 [cited 2022 Jul 2];16(1):112–38. Available from: /pmc/articles/PMC4819995/. [PMC free article] [PubMed]
- 32.Lukwa AT, Siya A, Odunitan-Wayas FA, Alaba O. Decomposing maternal socioeconomic inequalities in Zimbabwe; leaving no woman behind. BMC Pregnancy Childbirth [Internet]. 2022 Dec 1 [cited 2022 Aug 15];22(1):1–16. Available from: https://bmcpregnancychildbirth.biomedcentral.com/articles/10.1186/s12884-022-04571-9. [DOI] [PMC free article] [PubMed]
- 33.Joe W, Perkins JM, Kumar S, Rajpal S, Subramanian SV. Institutional delivery in India, 2004-14: Unravelling the equity-enhancing contributions of the public sector. Health Policy Plan [Internet]. 2018 Jun 1 [cited 2022 Aug 15];33(5):645–53. Available from: https://academic.oup.com/heapol/article/33/5/645/4962123 [DOI] [PubMed]
- 34.Kjellsson G, Gerdtham UG. On correcting the concentration index for binary variables. J Health Econ [Internet]. 2013;32(3):659–70. Available from: 10.1016/j.jhealeco.2012.10.012 [DOI] [PubMed]
- 35.Erreygers G. Correcting the concentration index. J Health Econ. 2009;28(2):504–15. [DOI] [PubMed] [Google Scholar]
- 36.Mutyambizi C, Booysen F, Stokes A, Pavlova M, Groot W. Lifestyle and socio-economic inequalities in diabetes prevalence in South Africa: A decomposition analysis. Devleesschauwer B, editor. PLoS ONE [Internet]. 2019;14(1):e0211208. Available from: 10.1371/journal.pone.0211208 [DOI] [PMC free article] [PubMed]
- 37.O’Donnell O, van Doorslaer E, Wagstaff A, Lindelow M. Analyzing Health Equity Using Household Survey Data. Anal Health Equity Using Househ Surv Data [Internet]. 2007 Oct 27 [cited 2022 Jul 2]; Available from: https://openknowledge.worldbank.org/handle/10986/6896
- 38.Yiengprugsawan V, Lim LL, Carmichael GA, Dear KB, Sleigh AC. Decomposing socioeconomic inequality for binary health outcomes: An improved estimation that does not vary by choice of reference group. BMC Res Notes [Internet]. 2010 [cited 2022 Jul 2];3. Available from: http://www.biomedcentral.com/1756-0500/3/57 [DOI] [PMC free article] [PubMed]
- 39.StataCorp. Stata statistical software: release 16. College station. TX: StataCorp LLC; 2019.
- 40.De Jonge E, Azad K, Hossen M, Kuddus A, Manandhar DS, Van De Poel E et al. Socioeconomic inequalities in newborn care during facility and home deliveries: A cross sectional analysis of data from demographic surveillance sites in rural Bangladesh, India and Nepal. Int J Equity Health [Internet]. 2018 Aug 15 [cited 2022 Aug 8];17(1):1–9. Available from: https://equityhealthj.biomedcentral.com/articles/10.1186/s12939-018-0834-9. [DOI] [PMC free article] [PubMed]
- 41.Bolam A, Manandhar DS, Shrestha P, Ellis M, Malla K, Costello AM. Factors affecting home delivery in the Kathmandu Valley, Nepal. Health Policy Plan [Internet]. 1998 Jan 1 [cited 2022 Aug 8];13(2):152–8. Available from: https://academic.oup.com/heapol/article/13/2/152/693579 [DOI] [PubMed]
- 42.Yadav AK, Sahni B, Jena PK. Education, employment, economic status and empowerment: Implications for maternal health care services utilization in India. J Public Aff [Internet]. 2021 [cited 2024 Jan 14];21(3):e2259. Available from: https://onlinelibrary.wiley.com/doi/abs/10.1002/pa.2259
- 43.Gabrysch S, Campbell OM. Still too far to walk: Literature review of the determinants of delivery service use. BMC Pregnancy Childbirth [Internet]. 2009 Aug 11 [cited 2024 Jan 16];9(1):34. Available from: 10.1186/1471-2393-9-34 [DOI] [PMC free article] [PubMed]
- 44.Sahoo J, Singh SV, Gupta VK, Garg S, Kishore J. Do socio-demographic factors still predict the choice of place of delivery: A cross-sectional study in rural North India. J Epidemiol Glob Health [Internet]. 2015 Jun 11 [cited 2022 Aug 15];5(4):S27–34. Available from: https://www.atlantis-press.com/journals/jegh/125906091 [DOI] [PMC free article] [PubMed]
- 45.Fatema K, Lariscy JT. Mass media exposure and maternal healthcare utilization in South Asia. SSM - Popul Health [Internet]. 2020 Aug 1 [cited 2024 Jan 17];11:100614. Available from: https://www.sciencedirect.com/science/article/pii/S2352827320302512 [DOI] [PMC free article] [PubMed]
- 46.Asp G, Pettersson KO, Sandberg J, Kabakyenga J, Agardh A. Associations between mass media exposure and birth preparedness among women in southwestern Uganda: a community-based survey. Glob Health Action [Internet]. 2014 Dec 1 [cited 2024 Jan 17];7(1):22904. Available from: 10.3402/gha.v7.22904 [DOI] [PMC free article] [PubMed]
- 47.Biswas RK, Rahman N, Islam H, Senserrick T, Bhowmik J. Exposure of mobile phones and mass media in maternal health services use in developing nations: evidence from Urban Health Survey 2013 of Bangladesh. Contemp South Asia [Internet]. 2021 Jul 3 [cited 2024 Jan 17];29(3):460–73. Available from: 10.1080/09584935.2020.1770698
- 48.Aboagye RG, Seidu AA, Ahinkorah BO, Cadri A, Frimpong JB, Hagan JE et al. Association between frequency of mass media exposure and maternal health care service utilization among women in sub-Saharan Africa: Implications for tailored health communication and education. Todd CS, editor. PLOS ONE [Internet]. 2022 Sep 29 [cited 2024 Jan 17];17(9):e0275202. Available from: 10.1371/journal.pone.0275202 [DOI] [PMC free article] [PubMed]
- 49.Janani Shishu Suraksha Karyakaram (JSSK):. National Health Mission [Internet]. [cited 2024 Jan 16]. Available from: https://nhm.gov.in/index4.php?lang=1_level=0_linkid=150_lid=171
- 50.Randive B, Diwan V, De Costa A. India’s Conditional Cash Transfer Programme (the JSY) to Promote Institutional Birth: Is There an Association between Institutional Birth Proportion and Maternal Mortality? PLoS ONE [Internet]. 2013 Jun 27 [cited 2022 Sep 19];8(6). Available from: /pmc/articles/PMC3694862/. [DOI] [PMC free article] [PubMed]
- 51.Salve HR, Charlette L, Kankaria A, Rai SK, Krishnan A, Kant S. Improving Access to Institutional Delivery through Janani Shishu Suraksha Karyakram: Evidence from Rural Haryana, North India. Indian J Community Med Off Publ Indian Assoc Prev Soc Med [Internet]. 2017 [cited 2024 Jan 16];42(2):73–6. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5427865/ [DOI] [PMC free article] [PubMed]
- 52.Goli S, Moradhvaj, Pradhan J, Reja T. The unending burden of high out-of-pocket expenditure on institutional deliveries in India. Public Health [Internet]. 2021 Apr 1 [cited 2024 Jan 16];193:43–7. Available from: https://www.sciencedirect.com/science/article/pii/S0033350621000512 [DOI] [PubMed]
- 53.Ogolla JO. Factors associated with home delivery in West Pokot County of Kenya. Adv Public Health. 2015;2015:1–6. [Google Scholar]
- 54.Kasaye HK, Endale ZM, Gudayu TW, Desta MS. January. Home delivery among antenatal care booked women in their last pregnancy and associated factors: Community-based cross sectional study in Debremarkos town, North West Ethiopia, 2016. BMC Pregnancy Childbirth [Internet]. 2017 Jul 14 [cited 2022 Aug 15];17(1):1–12. Available from: https://bmcpregnancychildbirth.biomedcentral.com/articles/10.1186/s12884-017-1409-2. [DOI] [PMC free article] [PubMed]
- 55.Berhan Y, Berhan A. Skilled health personnel attended delivery as a proxy indicator for maternal and perinatal mortality: a systematic review. Ethiop J Health Sci. 2014;24:69–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Srivastava J, Joseph A. Why Institutional Deliveries are Low in Balrampur District Uttar Pradesh: A Cross-Sectional Quantitative and Qualitative Exploration. J Obstet Gynecol India [Internet]. 2019 Jun 1 [cited 2024 Jan 16];69(3):225–31. Available from: 10.1007/s13224-018-1165-x [DOI] [PMC free article] [PubMed]
- 57.Yadav AK, Sahni B, Jena PK, Kumar D, Bala K, Trends. Differentials, and Social Determinants of Maternal Health Care Services Utilization in Rural India: An Analysis from Pooled Data. Womens Health Rep [Internet]. 2020 Jun 23 [cited 2024 Jan 17];1(1):179–89. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9380883/ [DOI] [PMC free article] [PubMed]
- 58.Yadav AK, Jena PK, Sahni B, Mukhopadhyay D. Comparative study on maternal healthcare services utilisation in selected Empowered Action Group states of India. Health Soc Care Community [Internet]. 2021 Nov 1 [cited 2022 Aug 15];29(6):1948–59. Available from: https://onlinelibrary.wiley.com/doi/full/10.1111/hsc.13309. [DOI] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
All the datasets processed and analysed for the current study are publicly available on the DHS webpage and can be accessed from https://dhsprogram.com/data/available-datasets.cfm.