Abstract
India launched the National Rural Health Mission (NRHM) in 2005 to improve maternal and child health by providing good quality health services to all, especially deprived sections of society, to reduce inequality in access to health services. With the backdrop of NRHM, we analysed the extent to which the utilisation of maternal health care services (MHCSs) in the three stages of the continuum of care—antenatal care (ANC), care during child delivery and postnatal care (PNC)—–has improved among the poor vis-à-vis the rich in India, and the corresponding narrowing down in inequality in the period 2006–2016. Data from the 3rd round of the National Family Health Survey (NFHS) in 2005–2006, capturing the pre-NRHM period and the 4th round of NFHS 2015–2016, capturing the post-NRHM era ten years after the implementation of the flagship programme, are used for the analysis. We estimated absolute as well as relative measures of inequality, absolute gap and coverage ratio between the poor and rich, slope index of inequality and concentration index. Our findings show that maternal health care coverage increased significantly among the poor for all components of MHCSs. Even so, the extent of utilisation of services remains significantly lower among the poor in 2015–2016 compared to the coverage among the rich in 2005–2006. Although inequality declined at the national level over the decade, it still persists. High equity has been achieved in using skilled birth attendance during child delivery and institutional delivery during 2015–2016, however, inequality continues to be higher for ANC indicators including consumption of iron and folic acid supplements for at least 100 days, receipt of four or more antenatal check-ups and comprehensive health check-ups at least once during antenatal visits and receipt of first check-up in the first trimester.
Introduction
India has been experiencing high maternal death ratio for a long time. Although the maternal mortality ratio has declined from 301 in 2001–2003 to 130 per 100,000 live births in 2014–2016 (Vital Statistics of India, 2006, 2016), India is yet to achieve the National Population Policy’s goal of reducing the maternal mortality ratio to 100 per 100,000 live births by 2010 or the Millennium Development Goal of 109 by 2015. The nation still accounts for 15% of global maternal death (Say et al., 2014). Inequality in maternal health outcomes and the utilisation of maternal health care services (MHCS) has also been a serious concern in India as it has been in other developing nations (Barros et al., 2012; Boerma et al., 2008; Countdown, 2015; Houweling et al., 2007; Paredes, 2016; Wong et al., 2017). Earlier studies showed that wide economic and socio-demographic inequalities existed in maternal and child health outcomes as well as in utilisation of MHCS (Jejeebhoy & Santhya, 2014; Paul et al., 2011) in India. Until 2006, non-poor mothers were more likely to benefit from MHCS than poor mothers (Mohanty & Pathak, 2009; Pathak et al., 2010). There was reduction at the rate of 1% per year in the gap between the rich and poor in utilising MHCS during the period 1992–2006 (Boerma et al., 2008).
To improve the maternal health scenario by providing accessible, affordable, accountable and good quality health services to all, including the poorest and those in the most remote areas of the country, the National Rural Health Mission (NRHM) was launched in 2005 (MOHFW, 2005). It marked a clear commitment to improve the scenario by revamping several existing schemes and introducing new schemes such as the Janani Suraksha Yojana, a conditional cash transfer scheme intended to encourage institutional delivery. The government also introduced the Accredited Social Health Activists (ASHAs) scheme to serve as a link between the community and public health system. The NRHM was subsequently restructured as the National Health Mission (NHM) in 2013 to address the health needs of the urban population as well. Further, the national Reproductive, Maternal, Newborn, Child Health + Adolescent (RMNCH+A) initiative has proposed a continuum of care model that focuses on the pre-pregnancy stage shortly after marriage and extends to the child care phase of the life cycle (MOHFW, 2013).
The utilisation of MHCS increased during the period from 2006 to 2016 after the implementation of the NRHM/NHM (IIPS & ICF, 2017). However, only a few studies have examined the increase in utilisation contributed by an increased utilisation by the poor due to the lack of data at the national level. For example, Joe et al. (2018) found that inequality in institutional birth decreased significantly at the national level during 2004–2014. Other studies focused on the use of selected MHCSs covering antenatal care (ANC) or child delivery care in specific states or regions. Using different rounds of District Level Health Survey data, Vellakkal et al. (2017) examined whether inequality in ANC and child delivery care decreased after the adoption of NRHM in eight high focused states and seven north-eastern states where socio-economic inequality was high. On account of comparatively high fertility and mortality indicators, the Government of India declared the following states—Jharkhand, Uttar Pradesh, Bihar, Chhattisgarh, Uttarakhand, Rajasthan, Orissa and Madhya Pradesh—accounting for 48% of Indian’s population as high focus states. The study found that coverage and equity of institutional delivery and ANC services increased moderately in most of these states during 2011–2012 compared to the pre-NRHM era (before 2006). Another study focusing on the same set of states except the north-eastern ones showed similar findings; institutional birth increased from 20% to 49% in the high focus states during 2005–2010, however, inequality persisted (Randive et al., 2013). The decline in inequality in institutional delivery is higher than in ANC in the early post-NRHM period—2007–2008. Equity in the utilisations of these services improved further during 2011–2012, although the extent varied across different states (Vellakkal et al., 2017). Ray et al. (2018) found that inequality persisted in Haryana in the use of ANC services as well as institutional delivery in 2013. ANC services here included four or more antenatal checkups, or receiving two or more tetanus toxoid (TT) vaccine during pregnancy, or consumption of 100+ iron and folic acid (IFA) tablets or syrups during pregnancy.
Although a limited number of available studies examined how the pattern of inequality in selected MHCS—either ANC or institutional delivery—changed after the adoption of NRHM/NHM, a comprehensive analysis of inequality focusing on all three stages of the continuum of care is still missing at the national level. It remained an unexplored topic in India due to the unavailability of nationally representative data after 2006. However, it is an important question that needs to be answered given the low health sector expenditure as a proportion of gross domestic product in this country. We hypothesise that inequality in the use of MHCS declined over the past decade after the implementation of NRHM. This study, therefore, aims to examine the extent to which the use of MHCS has improved among the poorest vis-à-vis the rich in India, economic inequalities in access to those services have narrowed over time and components of MHCS for which there has been a reduction in inequality of access between the years 2006 (pre-NRHM/NHM) and 2016 (post-implementation of NRHM/NHM). Comparison of the 4th round of national Family Health Survey Data (NFHS) with the 3rd round gave us an opportunity to answer the research question.
Methods
We have estimated inequality measures using two rounds of repeated cross-sectional surveys conducted at the national level capturing the pre- and post-NRHM era. The sample households have been divided into five quintiles according to their wealth index 1 : poorest (Q1), poor (Q2), middle class (Q3), rich (Q4) and the richest (Q5). We estimated the proportion of coverage of three groups of women for each of MCH services indicators: (1) for all ever married women; (2) for ever married women in the poorest quintile (Q1); and (3) for ever married women in the wealthiest quintile (Q5), and compared those estimates across two time points—2006 and 2016, capturing the pre- and post-NRHM era, respectively, to understand how MHCS utilisation has changed over the decade. We have reported the standard errors of coverage along with the average.
We also assessed the extent to which inequality in utilisation of MHCS had narrowed between 2005–2006 and 2015–2016 by using both basic and refined measures of absolute and relative measures of inequality. Two measures of absolute inequality are estimated—absolute gap in coverage between the poorest (Q1) and richest (Q5) women and slope index of inequality (SII), which refers to the absolute difference in predicted values of an indicator between the poorest and richest, estimated using logistic regression (Pamuk, 1985; Regidor, 2004). We also calculated measures of relative inequality—the ratio between Q5 and Q1 and concentration index (CI), which quantifies the extent to which a health service coverage indicator is concentrated among the poorest or the richest (Wagstaff et al., 1991). The CI is expressed in a scale of 0–1, where a value of 0 indicates perfect equality while a higher positive value indicates that health service is availed of more by rich women; 1 indicates perfect inequality. While absolute gap and ratio are the basic measures considering only Q1 and Q5, SII and CI are the refined measures of inequality capturing the full rank of wealth distribution of target women.
We reported standard errors of inequality along with values corresponding to the 95% confidence interval. All the estimation has been done using appropriate sample weights at the national level. We have presented concentration curves of the most equitable and inequitable MHCS indicators as well, which depict a graphical presentation of the relative concentration of MCH service indicator against economic status.
Study Samples and Data Used
The data for the analysis have been taken from the 3rd and 4th round of NFHSs (IIPS & ICF, 2017; IIPS & Macro International, 2007), conducted in 2005–2006 and 2015–2016, respectively. NFHS employs the same sampling methodology and survey instrument as the internationally used Demographic and Health Surveys (DHS). These nationally representative surveys followed a multi-stage stratified sampling technique; while at the final stage, households were selected using the systematic sampling method. Like DHS, the NFHS is based on interviews of women aged 15–49 years and collects information on availability and accessibility of MHCSs along with information on their demographic and economic characteristics. Detailed information of MHCSs is available across the three stages of the care continuum –ANC, child delivery and PNC. During the surveys conducted, questions related to the trimester in which the mother first received ANC, number of times ANC was received during pregnancy, place of the child’s birth and how soon the mother received a check up after the birth of the child were asked. NFHS 3 and 4 covered 124,385 and 699,686 women, respectively. Our analysis focuses on care seeking for maternal health for the last birth among every married women aged 15–49 years, who had given birth in the five years before the survey. Thus, the final analytical sample size becomes 36,850 and 186,721 in 2005–2006 and 2015–2016, respectively. National level sampling weights are used to adjust for sample selection. The data analysis was performed with Stata 13.1 (StataCorp, 2013).
Key Outcome Variables
Definition of Outcome Variables.
Results
A comparison of the coverage of MHCS in 2005–2006 and 2015–2016 shows that utilisation has increased significantly since the launch of the NRHM, regardless of the indicator used (Table 2). However, the magnitude of improvements in coverage varied widely across different components, for example, indicators like institutional delivery (increased from 41% to 81%), skilled birth attendance (SBA) (increased from 50% to 83%), post-partum checkup within two days of delivery (increased from 36% to 65%) and receiving comprehensive tests and check-ups at least once during antenatal visits (increased from 35% to 64%) observed considerably high increase. In comparison, services like receiving two doses of TT injections (76–83%), at least four antenatal check-ups (37–51%), first check-up in the first trimester (44–58%), and consumption of IFA supplements for at least 100 days (15–30%) observed lesser increase.
Coverage of Maternal Health Care Services (in %).
Source: Author’s calculation.
*p < .01.
For most of the indicators, improvement in MHCS utilisation was low for poor women in 2006. It hovered mostly around 10–23% except two doses of TT injection. In the next 10 years, coverage increased significantly among them, regardless of the indicator. Like the national trend, institutional delivery, use of a SBA and post-partum checkup within two days of delivery increased by as much as 48%, 44% and 34%, respectively. On the contrary, the increase in the coverage of consumption of IFA tablets at least for 100 days (6–14%), at least four ANC visits (12–25%) and first ANC visit in the first trimester (23–38%) remained low.
Improvements in the utilisation of services among rich women varied across different components. For example, while indicators such as consumption of IFA supplements for at least 100 days (36–48%), receiving comprehensive health check-ups at least once during antenatal visits (78–88%), institutional delivery (86–96%), SBA (90–96%) and post-partum check-up within two days of delivery (73–80%) increased significantly, indicators like first antenatal check-up in the first trimester (96–94%), four or more antenatal check-ups (77–73%) and two doses of TT injection (92–88%) decreased or remained unchanged during the inter-survey period.
Inequality in Utilisation of MHCSs
The findings presented in Table 3 show that inequality in the utilisation of MHCS had declined between 2005–2006 and 2015–2016 for almost all indicators, regardless of the measure of inequality that we used. The only exception was the proportion of women who had consumed IFA supplements for at least 100 days during pregnancy, which recorded an increase in inequality as measured by indicators of absolute inequality, but a small decline was observed when measured by indicators of relative inequality.
Measures of Absolute and Relative Inequality in Utilisation of Maternal Health Care Services Between 2005–2006 and 2015–2016.
Source: Author’s calculation.
Despite declines in inequality during the inter-survey period, it is still significantly high, suggesting that women who belonged to the wealthiest quintile had greater coverage than women who belonged to the poorest wealth quintile. The most equitable indicator was the receipt of two doses of TT injection according to all four measures of inequality in 2015–2016. This is followed by SBA, institutional delivery and post-partum checkup within two days of delivery as per relative measures of inequality. The least equitable indicators included consumption of IFA supplements for at least 100 days (according to relative measures of inequality), receipt of four or more antenatal check-ups and receipt of comprehensive health check-ups at least once during antenatal visits (as per all four measures of inequality) and receipt of first check-up in the first trimester.
Concentration curves for economic inequalities in utilisation of MHCS in 2005–2006 and 2015–2016. (A) Concentration curve for TT2 injections in 2005–2006 and 2015–2016. (B) Concentration curve for institutional delivery in 2005–2006 and 2015–2016. (C) Concentration curve for iron and folic acid consumption for at least 100 days in 2005–2006 and 2015–2016. (D) Concentration curve for more than four ANC visits in 2005–2006 and 2015–2016.
The concentration curves presented in Figure 1 show that inequality declined for all the selected indicators during 2006–2016, although in varying magnitudes. The pro-rich bias in the utilisation of MHCS decreased promisingly for institutional delivery and TT2 injections as concentration curves move closer to the line of equality in 2015–2016 compared to 2005–2006. In contrast, the consumption of IFA tablets and at least four ANC visits are still low among the poor as concentration curves remain significantly away from line of equality during 2015–2016, even 10 years after the implementation of NRHM.
Discussion
Our study extends the limited literature on coverage of utilisation of MHCS in the post-NRHM era in India among poor vis-à-vis the rich and corresponding inequality. This study provides a comprehensive analysis on change in the pattern of MHCS utilisation and inequality across eight different indicators covering continuum of care. Existing evidence suggested that utilisation of MHCS increased mainly among the non-poor until 2006 (Mohanty & Pathak 2009) while we showed that utilisation of MHCS increased especially among the poor during 2006–2016. However, increase in utilisation varied widely across different indicators. Institutional delivery being the major focus of reforms under NRHM observed highest increase in coverage among the poorest during the past decade. On the contrary, use of ANC services progressed slowly among poor which is in line with many countries in Asia and Africa (Boerma et al., 2018). For instance, we find that the consumption of IFA supplements, at least four ANC checkups and ANC visit during first trimester increased only by 8%, 13% and 15%, respectively. These percentages are only moderately higher than those reported in Pathak et al. (2010), which estimated that more than four antenatal checkups increased only by 0.2% among the poorest women during 1992–2006.
Corresponding inequality in MHCS utilisation decreased over the past decade although it persists. The magnitude of decline in inequality varies widely across ANC, child delivery and PNC indicators. High equity has been achieved in the child delivery and post-partum care indicators while inequality is still high for ANC services. For example, institutional delivery experienced the highest decline in inequality while consumption of IFA supplements during pregnancy observed the least decline. Wealthier women still access higher level of ANC services.
A meta synthesis of qualitative studies from African, South and Southeast Asian countries identified different factors influencing the low use of ANC services among poor women (Finlayson & Downe, 2013). For example, lack of understanding of ANC benefits, considering pregnancy a normal life event—seeking ANC only when sick, preference for traditional healers and medicines especially in rural areas, high cost-burden including cost of transport, opportunity cost of wages foregone, etc, inadequate infrastructure and lack of medicines in health care facility centers. Low quality of care is another important concern for low use of ANC care, patients face lack of respect, empathy and compassion from health care providers (Ardey & Ardey, 2015; Narichiti, 2013). Historical evidence in the Indian context even noted that ASHAs, the link between NRHM and the communities, lacked training and knowledge and were poorly paid (Bajpai & Dholakia, 2011; Saprii et al., 2015), hampering their efficiency in delivering the required services. Education of women has also been found to be an important determinant of ANC, skilled birth assistance and PNC (Jat et al., 2011; Kesterton et al., 2010).
Conclusion
The study contributes to the scant literature on equity analysis in utilisation of MNCH services at the national level after the implementation of the NRHM/NHM in India. Our findings highlight the need for accelerated efforts that are strategically targeted at the poor. Complementary programs such as improvement in women’s education, creating economic opportunities for women along with direct intervention on maternal care need to be focused upon since the interplay of all these factors lead to inequality. In addition, interventions related to improvement of the quality of care, increasing awareness, incentivising comprehensive ANC seeking, increasing the number of frontline workers and lower-level facilities in areas most in need, to extend the continuum of care and the provision of ongoing mentoring to health care providers can be tested at scale.
Footnotes
Acknowledgements
The author is thankful to Dr K. G. Santhya and Prof F. Ram for their insightful comments.
Declaration of Conflicting Interests
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by MacArthur Foundation (G-109245-0) under the aegis of “Policy research and advocacy for strategic investment in maternal, newborn and child health in India”.