Introduction

Maternal and child health remains a critical public health issue around the world, having a profound impact on societal growth and well-being. Despite significant progress in reducing maternal and child mortality rates across the world over the past few decades, considerable challenges still remains, particularly in low- and middle-income countries (LMICs)1,2. Globally, approximately 223 mothers per 100,000 live births died during and following pregnancy and childbirth in 2020, with the vast majority (95%) in LMICs3. On the other side, about 4.9 million children under-five years old died in the year 20224.

India is accountable for the second-highest number of maternal deaths globally and contributed to the world’s highest under-five deaths in 20155,6. Envisioning the Sustainable Development Goals (SDGs) set by the United Nations in 2015 calling for major reductions in maternal, neonatal and child mortality by 2030, India has steadily advanced on the track to achieve this goal ahead of time with its policies for women’s health and well-being7. Maternal Mortality Ratio (MMR) has declined from 254 per 100,000 live births in 2004–06 to 97 per 100,000 live births by 2018–208, while that of under-five declined from 69 per 1000 live births in 2008 to 32 per 1000 live births in 20209,10.

Empowered Action Group (EAG) states in India have the highest maternal and Under-5 mortality rate (U-5MR) sharing the major burden of total deaths11,12,13. Government of India (GOI) in 2001 identified eight socioeconomically disadvantaged states, known as the Empowered Action Group (EAG) namely Bihar, Chhattisgarh, Jharkhand, Madhya Pradesh, Odisha, Rajasthan, Uttarakhand, and Uttar Pradesh. India has been proactive in launching and implementing various initiatives to achieve these targets with concerted efforts on poor performing states and districts. One significant flagship program is the Janani Suraksha Yojana (JSY), introduced in 2005 under the National Rural Health Mission (NRHM), aimed at reducing maternal and neonatal mortality14. JSY provides cash incentives to pregnant women if they opt for institutional deliveries, i.e. childbirth in healthcare institutions. The program targets women from low-income families, particularly those in rural areas. Accredited Social Health Activists (ASHAs) in India, who play an intermediary role between the government and beneficiaries, are also paid incentives under this scheme to motivate women to avail antenatal services, choose institutional deliveries, and obtain postnatal services15. With the government spending over $200 million a year between 2008 and 09 and 2009–10, the JSY was one of the world’s largest cash transfer programs, serving almost 7.7 million people annually until 2012–1316,17,18,19.

Numerous global and regional initiatives have been launched to address maternal and child health issues, with a focus on LMICs. Conditional Cash Transfer (CCT) schemes, a demand-side program have emerged as an effective strategy to encourage institutional deliveries by providing financial incentives to pregnant women, contingent on their delivering in healthcare institutions20. Studies have demonstrated the effectiveness of CCT programs in enhancing maternal health outcomes19,21,22. Additionally, there are a few studies from LMICs that indicate these programs can also improve neonatal health outcomes23. Studies conducted in India at the national, state, local, and individual levels have also examined the welfare benefits and impact of the JSY on mother and child health care, health-seeking behavior and mortality outcomes17,19,23,24,25,26,27,28,29,30. Institutional deliveries in India have risen sharply from 47% in 2007-08 to over 88.6% in 2019-2131. Besides, the JSY has also resulted in reduction of out-of-pocket expenditure for delivery care32,33, in turn reducing the catastrophic health expenditure burden for new parents, most importantly for those from socioeconomically disadvantage backgrounds34.

While JSY primarily focuses on increasing institutional deliveries and improved healthcare service utilization, the period immediately following childbirth is equally critical for the health and well-being of both the mother and the newborn. However, little is known about the effect of JSY on postnatal services in the vulnerable population in EAG states. Our study attempts to explore whether JSY can also have an impact on four key neonatal parameters namely- skin-to-skin contacts of mother and newborn immediately after birth, early initiation of breastfeeding, zero dose immunization, and postnatal checkup within 2 months of birth. Although JSY does not directly target these specific neonatal parameters, we hypothesize that women who avail JSY services can benefit from improved access to maternal and newborn healthcare services, enhanced knowledge, and a more positive attitude towards healthcare-seeking behaviors and postnatal care. This indirect influence has potentially led to improved neonatal outcomes among JSY beneficiaries.

Thus, the objective of this study is to unearth the effectiveness of JSY in providing comprehensive postnatal care, involving skin-to-skin contacts, early initiation of breastfeeding, zero dose immunization and enhance postnatal checkup within 2 months of birth, in the EAG states of India.

Methods

The study used recent round of nationally representative sample survey in India named as National Family Health Survey (NFHS; fifth round − 2019-21). NFHS is a large-scale, multi-round survey conducted on representative samples from all states and union territories of India. It gathers comprehensive data on fertility, mortality, reproductive health, maternal and child health, adolescent health, healthcare utilization, high-risk behaviors, and domestic violence. For this study, the data was obtained from the Demographic and Health Survey (DHS) data program portal following prior registration and permission. The dataset is publicly available and contains no identifiable information, thereby posing no ethical implications35. As presented in Fig. 1, a series of inclusion criteria were followed before selecting the final sample for analysis. Of a total of 395,403 births in the last five years preceding the survey, only recent births (implies to the most recent childbirth in terms of parity in last five years preceding the survey) in EAG states were selected (n = 86,059). For deeper comprehension we have excluded home delivery and cesarean section deliveries leaving a total of 62,415 women included for final analysis. Caesarean births were excluded as women are less likely to initiate immediate skin-to-skin contact or breastfeeding36, and they routinely receive postnatal check-ups during extended hospital stay, which could bias results. Additionally, home births were also excluded since JSY benefits are applicable only for facility-based deliveries.

Fig. 1
Fig. 1
Full size image

Schematic representation of sample covered for the analysis using data from NFHS, 2019-21.

The analysis comprised samples from the EAG states. These states are the poorest performers across multiple health indicators, which comprise almost 46% of India’s population and account for 61% of those living below the poverty line. To expedite the reduction of maternal and newborn mortality rates in India, the GOI have strategically prioritized and categorized the states as EAG i.e., Empowered Action Group (EAG) and Non EAG States37,38.

Variables

Outcome variables

For this analysis, authors have considered four key outcome variables such as comprehensive postnatal care namely skin-to-skin contacts of mother and newborn immediately after birth (categorized as Yes/No), early initiation of breastfeeding (categorized as Yes/No), zero dose immunization (including birth dose of Oral Polio, BCG and Hepatitis B vaccination) (categorized as Yes only if all vaccination were given, otherwise - No)and postnatal checkup within 2 months of birth (categorized as Yes/No). All the outcome variables were coded into binary variables. These outcome variables are essential for ensuring newborn survival and long-term health.

Treatment/exposure variables

We have considered mothers who have received cash assistance from JSY as our treatment (exposed) group and others as control (not exposed) group. This information is collected for the most recent births from women.

Matching variables

A total of 10 variables were selected based on literature review as matching variables. These matching variables comprises of age of women (15–25/26–49), education of women (No education/ Primary/ Secondary/ Higher), sex of head of household (Male/ Female), caste (Schedule Caste (SC)/ Schedule tribe(ST)/ Other Backward caste(OBC)/ Other), religion (Hindu/ Other), wealth status (Poorest/ Poorer/ Middle/ Richer/ Richest), place of residence (Urban/ Rural), exposure to media (Not at all/ Less/Once a week), seeking medical help is a problem (Low/ Moderate/ High), place of delivery (Private/ Public).

Considering these as key covariates, subsequent to our PSM analysis (Supplementary file − 1), we have also analyzed them using logistic regression analysis.

Statistical analysis

All statistical analyses were conducted using STATA version 17.0 software. Descriptive statistics, along with bivariate analysis, were utilized to examine the distribution of sample characteristics. Primarily the prevalence of each outcome variable was analyzed by the selected EAG States. Geographical representation of the overall prevalence of women receiving JSY assistance amongst those who have normal institutional delivery was also analyzed and map was generated by the authors using QGIS39. Odds ratios were first assessed to identify associations between JSY and key neonatal outcomes.

Later matching estimates of JSY on selected neonatal outcome variables among women who have received/not received JSY Services in EAG States in India is presented using the three key parameters under PSM analysis. Sampling errors and t-test results are also reported. The study also presents balance plots and love plots which are essential for assessing the effectiveness of the matching process or to validate if matched samples are comparable. Balance plot compares the distribution of covariates before and after matching, ensuring that the treatment and control groups are similar in their observed characteristics. On the other hand, love plots graphically display the standardized mean differences of covariates between the groups, highlighting improvements in balance achieved through matching.

The study further conducted a comprehensive analysis of the identified matching variables (considering them as potential covariates) using statistical methods (Supplementary file 1). This analysis allowed an in-depth exploration of matching variables in relation to the selected outcome variables, providing a robust understanding of their associations and potential impacts.

Results

Background characteristics

A total of 62,415 women who had normal delivery in a healthcare facility were included in our analysis. Of these 33,623 women (53.9%) received JSY services in the EAG States. A skewness in the percentage of women receiving JSY assistance can be seen from Fig. 2. Proportion of women receiving JSY is higher than the national average in the states of Odisha (77.1%), Ladakh (72.3%), Jammu & Kashmir (58.8%) and Madhya Pradesh (58.5%). Amongst these four states, two fall in the category of EAG whereas other EAG states i.e., Bihar (47.7%), Uttar Pradesh (49.9%), Uttarakhand (42.8%), Chhattisgarh (52.9%) and Rajasthan (50.2%) and Jharkhand (36.1%) still fall in low performing category i.e., below national average.

Fig. 2
Fig. 2
Full size image

Spatial distribution presenting percentage of women receiving JSY assistance amongst those who have normal institutional delivery, India by EAG/Non EAG States 2019-21 (Created by Authors using QGIS).

Descriptive analysis

Further analysis of each selected outcome variable was compared by mothers who received the JSY benefit and those who did not (Table 1). It is interesting to note that almost all the service uptake variables were better amongst those women who received JSY services. Besides, this difference existed across all the EAG States for all the selected outcome variables.

Table 1 Percentage of selected outcome variables amongst those who have received v/s did not receive JSY services in selected EAG States in India, 2019-21.
Full size table

Single and multiple binary logistic regression analysis

Further, a comprehensive analysis of the background characteristics of the matching variables considering them as essential covariates were undertaken with all the selected outcome variables (Supplementary file-1, Table S1).

Women who have received JSY services (Supplementary file- 1, Table S2) have significantly higher odds of skin to skin contact [1.45 (95% CI 1.38, 1.52)], breastfeeding [1.34 (95% CI 1.30, 1.39)], having all post-natal checkup within 2 months [1.53 (95% CI 1.48, 1.58)] and receiving zero dose immunization as per National Immunization Schedule (NIS) [1.72 (95% CI 1.65, 1.79)] than those women who have not received JSY services. This result remained similar when adjusted for all the matching variables. Women who have received JSY services when adjusted for other variables have significantly higher odds of putting child on chest [1.2 (95% CI 1.14, 1.27)], breast feeding [1.27 (95% CI 1.22, 1.31)], having all post-natal checkup within 2 months [1.50 (95% CI 1.45, 1.55)] and receiving zero dose immunization as per NIS [1.47 (95% CI 1.41, 1.54)].

Verification of the PSM analysis

Figure 3 presents the standardized % bias across all covariates before and after matching. It is evident that the mean differences were high across almost all the variables before matching, but such bias was significantly reduced after matching. i.e. covariates were found sufficiently balanced.

Fig. 3
Fig. 3
Full size image

Love plot indicating standardized % bias across all covariates in matched and unmatched sample.

Figure 4 represents the distribution of propensity scores explaining the quality of the matching process. The figure indicates that, post-matching, the distributions for the treated and control groups became nearly identical, indicating unbiasedness in the estimated treatment effects.

Fig. 4
Fig. 4
Full size image

Balance graph indicating the quality of the matching process.

Table 2 presents the findings from main PSM analysis and estimated the effect of JSY on Selected neonatal outcome. These results are interpreted with the important consideration that they are derived after controlling for socio-economic covariates.

ATT is defined as the Average Treatment Effect on the Treated, ATU as the Average Treatment Effect on the Untreated, and ATE as the Average Treatment Effect. An ATE indicates the difference in the selected outcomes for women who received JSY assistance versus those who did not. The highest average treatment effect of 11% was noted for newborn’s postnatal check-up within 2 months. This indicates that on an average newborn’s postnatal checkup is 11% higher for women who received JSY compared to those who did not receive JSY. Other three outcomes namely skin-to-skin contacts of mother and newborn, early initiation of breastfeeding, and zero dose immunization reported an ATE of 0.02, 0.05, and 0.08 respectively. This implies that that on an average skin-to-skin contacts of mother and newborn, early initiation of breastfeeding, and zero dose immunization is 2%, 5% and 8% higher for women who received JSY compared to those who did not receive JSY respectively. Amongst 4 selected outcome variables- skin-to-skin contact yielded no significant difference between the JSY versus no JSY group.

The first outcome variable i.e., skin-to-skin contacts of mother and newborn yielded a not significant result (t-stat – 0.850).

Table 2 Matching estimates of JSY on selected neonatal outcome variables among women those who have received/not received JSY services in EAG States in India, 2019-21.
Full size table

Similarly, for the second outcome variable i.e., breastfeeding within 1 h, the average treatment effect on treated values for treated and control groups were 0.41 and 0.39, respectively. It implies that 41% of women who received JSY initiate breastfeeding within an hour while breastfeeding in the absence of JSY assistance, 39% of women would initiate it within 1 h of birth. Value of ATU indicates that 7% more women who did not receive JSY assistance, would have breastfed the child within 1 h of delivery if they had received the JSY assistance.

Newborn’s postnatal check-up was another outcome variable that shows the average treatment effect on treated values for treated and control groups were 0.51 and 0.42, respectively. It indicates 51% of newborns of the women who receive JSY assistance, have post-natal check-ups within 2 months, whereas it would be 42% in the absence of JSY. The value of ATU indicates that an additional 12% of the women who did not receive JSY assistance would have undergone newborn postnatal checkups if they had received the JSY assistance.

Another essential variable analyzed was whether newborns received all zero-dose vaccinations as per NIS. The result indicates that the ATE on treated values for treated and control groups were 0.75 and 0.64, respectively. This indicates 75% of newborns of the women who receive JSY assistance have all zero-dose vaccination and in the absence of JSY assistance, 64% of newborns would receive all zero-dose vaccination. The value of ATU indicates that 6% more newborns of women who did not receive JSY assistance would have received all zero-dose vaccinations if their mothers had received the JSY assistance.

Discussion

The study used a large scale population level data of EAG states of India and examined the status of selected neonatal indicators amongst women who have and have not received JSY. We noted that almost all the outcomes were better amongst those women who received JSY services except skin-to-skin contact of mother and newborn. This difference existed across all the EAG States individually for all the selected outcome variables, implying that JSY successfully promotes better practices and health metrics. The uniform benefit across all EAG suggests that the program’s impact is substantial and can be a valuable model for improving healthcare practices and policies.

These findings are in sync with previous study examining the unintended effect of JSY on key parameters like contraception usage17. CCT programs have a proven track record of enhancing maternal and child health outcomes, especially in low- and middle-income countries22,40. Their origins can be traced back to the 1990s with Mexico’s Progresa program41. On that premise, JSY in India was launched by the Government of India in 2005 with an objective to reduce maternal and neonatal mortality by promoting institutional deliveries among pregnant women, particularly those from low-income and disadvantaged backgrounds. The scheme provides cash assistance to pregnant women, which varies depending on the state and whether it is categorized as high or low performing. In EAG (low performing states) states, the scheme covers all pregnant women, which was considered for this analysis42. Remarkably, three states among the total eight states have the high maternal and Under-5 mortality rate (U-5MR)11,12,13.

For breastfeeding within one hour of birth, findings reveal that 39% of women without JSY assistance would initiate breastfeeding, whereas additional 7% women might have also initiated breastfeeding within one hour if they would have received JSY assistance. This indicates that while JSY has a noticeable impact on encouraging timely breastfeeding, a significant proportion of women would breastfeed early even without the program’s intervention. Unlike skin-to-skin contact, this practice also gets influenced by cultural practices and family support43,44. Cultural beliefs, such as misconceptions about colostrum and gender-related practices, may discourage early initiation of breastfeeding (EIBF), while nulliparous mothers—due to younger age and limited knowledge—represent an important target group for interventions45. Involvement through JSY assistance ensures that women are in contact with the health system, particularly with frontline workers who play a major role in facilitating these practices.

In line with other published literature17, our study noted that 42% of newborns would receive all recommended postnatal check-ups within two months even without JSY support and there would be 12% increase in the likelihood of completing these check-ups among those who did not receive the support, if they would receive JSY assistance. On the other side the fourth outcome i.e., zero-dose vaccinations also suggests that 64% of newborns would receive all zero-dose vaccinations as per the National Immunization Schedule (NIS) without JSY support and JSY could lead to an additional 6% increase in vaccination rates among those who did not receive the assistance. In line with other published literature where CCT improves neonatal health outcomes23,24, our finding also highlights JSY’s effectiveness in promoting postnatal checkups and early vaccination, which is vital for preventing diseases in newborns. These could be majorly facilitated by the ASHAs i.e., the Accredited Social Health Activists, who play a crucial role in educating and encouraging mothers about the importance of postnatal care and immunization. They facilitate access to healthcare services, ensuring timely check-ups and vaccinations. Without JSY, cultural practices and family support result in significant baseline adherence, but JSY’s structured support and the proactive involvement of ASHAs significantly boost compliance, enhancing maternal and neonatal health outcomes46.

Studies have shown a reduction in perinatal and neonatal mortality rates and increased institutional deliveries due to CCT19,47 which can be attributed to an increase in overall healthcare access. Some studies suggest that CCT programs can improve neonatal health outcomes23,48. However, the key services for newborns like early breastfeeding, skin-to-skin contact, postnatal checkups, and zero-dose vaccination that have a significant impact on reducing adverse neonatal outcomes have not, to the best of our knowledge, been studied in relation to CCT programs like JSY. Where this being one of our key strengths for this study, the rigor of the statistical technique used is another strength that we have. We employed PSM to assess the impact of JSY services on four key neonatal parameters. This statistical method ensures that study participants in both groups are similar based on all relevant characteristics, minimizing selection bias. This method is often employed to see treatment/exposure effects in cross-sectional, observational, or non-experimental data when randomized clinical trials are not feasible or not being undertaken. It ensures that the groups are identical except for the treatment variable, leading to a less biased estimate of treatment/exposure effects49. Apart from all this, the use of the most recent round of data (minimizing recall bias) from a nationally representative large sample survey are key strengths.

There are certain limitations in our analysis. We relied on secondary data that may be associated with false reporting of certain indicators, which may introduce some systematic error, equally affecting both the groups (random misclassification). Thereby, a likelihood of the measures of effect getting diluted. Additionally, NFHS data may be subject to recall bias due to retrospective reporting in past 5 years preceding the survey. To minimize this, we have only included the recent delivery information. Moreover, the findings may not be generalizable beyond the EAG states.

Additionally, study reported that JSY uptake varies across EAG states, with some states like Odisha showing higher coverage compared to Bihar or Jharkhand. This heterogeneity likely reflects differences in state-level program implementation, health system capacity, and community awareness or engagement with maternal health services. Although limited literature is available on these inter-state variations, we recommend conducting a qualitative study to explore the reasons behind differential JSY uptake across states.

Beyond the main goal of JSY, we have attempted to examine the broader benefits of JSY-induced facility-based deliveries. The variables chosen were those that improve overall neonatal health outcomes. Cultural barriers, variability in healthcare facility practices, and the program’s primary focus on increasing institutional deliveries without specific incentives for these key activities can limit its effectiveness in promoting these critical neonatal health practices. Despite these limitations we can assess if JSY indirectly influences these essential practices, identify gaps, and develop strategies to overcome barriers.

Conclusion and recommendations

We can infer that the impact of JSY extends beyond institutional delivery into the postnatal period in the poorer states of India. It underscores the point that institutional deliveries per se may not influence behaviors that may affect neonatal survival. The contact and interaction of beneficiaries with health system as evident from utilization of JSY services during antenatal period seems to have lasting impact. Better coordination and counseling may increase the effect size further. Additionally, monitoring and leveraging community networks can ensure consistent support and follow-up care for mothers. Addressing these areas can optimize JSY, ensuring healthier futures for mothers and their children in India. Additionally, linking ASHA incentives to breastfeeding counseling and incorporating neonatal indicators into JSY monitoring could potentially strengthen overall program implementation and outcome at large.