Final Report Summary - MATIND (Large scale innovative pro-poor programs focused on reducing maternal mortality in India: a proposal for impact evaluation)
Executive Summary:
This project comprises an evaluation of the performance of the demand side financing programs to reduce maternal mortality. The evaluation is comprehensive and takes into account a number of domains – access, utilization, human resources, quality of care, participation of the private sector, user perceptions, policy dimensions as well an economic analysis.
High maternal mortality in India is a serious public health challenge. Demand side financing interventions have emerged as a strategy to promote access to emergency obstetric care. Two such state run programs, Janani Suraksha Yojana (JSY) and Chiranjeevi Yojana (CY), were designed and implemented to reduce financial access barriers that preclude women from obtaining emergency obstetric care. The MATIND project operates in two Indian provinces, Madhya Pradesh and Gujarat, where these programs are operationalized.
JSY, a conditional cash transfer, awards money directly to a woman who delivers in a public health facility. This is studied in the Madhya Pradesh province. CY, a voucher based program, empanels private obstetricians in the Gujarat province, who are reimbursed by the government to perform deliveries of socioeconomically disadvantaged women. The programs have been in operation for the last ten years.
This project is a comprehensive evaluation for two innovative demand side financing programs that aims to study: (i) trends in program uptake since inception, (ii) the influence of the program on institutional delivery rates, maternal and neonatal outcomes, (iii) explore service providers’ experiences, (iv) emergency transportation, (vi) quality of care administered, (vii) experiences of the beneficiaries and non-beneficiaries, (viii) the role of community health workers and (ix) the cost effectiveness of the programs.
Primary data was collected in each district at three levels: 1. state health departments (policy level) 2. obstetric care facilities, and 3.recently delivered mothers in the community. A combination of qualitative and quantitative methods was used including: 1. facility and community-based surveys, 2. structured observations, 3. record reviews, 4. case vignettes, and 5. in-depth interviews.
In summary, the JSY has resulted in steep increases in the proportion of facility based deliveries in the study area but also in the country. However our analysis was unable to detect any influence of program utilization with reduction in maternal mortality. 90% of institutional deliveries in the study districts occurred under the program. However the quality of care under the program was less than optimal. There were serious deficiencies in the skills of front line birth attendants, and the ability to provide comprehensive emergency obstetric care services like cesarean and blood transfusion was extremely limited. The program ran a sub arm in partnership with the private sector, however as the state paid private obstetricians differentially for vaginal and cesarean births, it resulted in disproportionate cesareans and the state had to close the partnership. Women under the program still experienced out of pocket costs, but these were not related to technical care, rather to informal payments paid at facilities.
The CY program was more targeted to vulnerable women, however though it has had upto a million births occurring under the program, the proportion of eligble women who become program beneficiaries is low (15%). This is largely because of a lack of knowledge, an inability to get the documentation prepared and because the partnership has seen participation of private obstetricians change with time. There was a better quality of care provided under the CY program. Birth attendants worked under the close supervision of private obstetricians who were very aware of their reputations, cesareans were easily available. However although the program was intended to be cashless at the point of care for the beneficiary, this was not always the case, as private obstetricians were uncertain if they would be reimbursed by the state as the paper work was burdensome.
The role of Accredited Social health activist was explored in the programs. The cost effectiveness of each program was studied.
The overall evaluation has important lesson for the design and implementation of large scale demand side financing programs, particularly in the design and management of state led public private partnerships to achieve universal health coverage.
Project Context and Objectives:
MATIND studied two state run programs, Janani Suraksha Yojana (JSY) and Chiranjeevi Yojana (CY), designed to reduce financial access barriers that preclude women from obtaining emergency obstetric care in India. MATIND assessed the following: (1) trends in programs uptake since programs inception, (2) the influence of the programs on institutional delivery rates, maternal and neonatal outcomes, (3) service providers’ experiences, (4) emergency transportation systems for obstetric referrals, (5) characteristics of the facilities participating in the programs, (6) quality of care administered, (7) experiences of the beneficiaries and non-beneficiaries, (8) the role of community health workers, and (9) the degree of subsidy to users because of the programs.
Why was MATIND conducted?
Maternal mortality is a 21st century problem essentially only for the poor, and one virtually eliminated for people with the means and status to access health care. Such a marker of global inequity is shocking and is an indication of wider development issues targeted in some of the other MDGs, especially poverty, education, and gender. A link between poverty and maternal health has been clear for more than a century, and is supported by extensive evidence from literature. Even though tax based public healthcare provision has been known to be one of the best alternatives to promote equity in access, it has been difficult to ensure that funding is distributed to the most needy, particularly the poor living in non-urban areas. This difficulty is even more acute in low income countries, where insurance and capital markets are less well developed.
Demand side financing directs purchasing power to the target group, to enable them to purchase specific services[42] and can take a number of different forms (including vouchers, conditional cash transfers). Interventions to reduce demand side barriers are justified to rectify market failures, and in the pursuance of equity. Such programs on a large scale have first been used to pursue reduction in maternal mortality in the Indian context.
Maternal mortality in India is a serious public health challenge. Demand-side financing interventions have emerged as a strategy to promote women’s access to emergency obstetric care (EmOC) in an attempt to reduce maternal deaths. Two state-run programs, JSY and CY, have been implemented to increase women´s access to EmOC: (1) JSY awards money directly to a woman who delivers in a public health facility, and (2) CY empanels private obstetricians in Gujarat province, who are reimbursed (a fixed package for 100 deliveries including C-section and complications) by the government to perform deliveries for poor and tribal women. In spite of JSY and CY magnitude and the amount of resources allocated, no comprehensive evaluations of these programs have been conducted.
Where was MATIND conducted?
The study took place in two provinces, Madhya Pradesh and Gujarat (Figure 1) where JSY and CY were studied, respectively. In each province we purposefully selected three districts with different socioeconomic, demographic, and health profiles. Madhya Pradesh, located in the central part of India, has a population of 72 million. The state is divided into 51 administrative districts, each with a population of 1–2 million. Three districts, Shahdol, Panna and Ujjain, were selected for the study. Gujarat has a population of 60 million and 26 administrative districts. Each district has on the average a population of more than two million. Three heterogeneous districts from the western, central and eastern belts of the state, Sabarkantha, Surendranagar and Dahod, were purposefully selected for the study. These districts had different sub-population compositions i.e. varying proportions of the eligibility population (scheduled castes and tribes and women living below the poverty line) for the CY program.
What study designs did MATIND use?
Primary data was collected in each district at three levels: 1. state health departments, 2. obstetric care facilities, and 3.recently delivered mothers in the community. A combination of qualitative and quantitative methods was used including: 1. facility and community-based surveys, 2. structured observations, 3. record reviews, 4. case vignettes, and 5. in-depth interviews.
The levels of this evaluation:
(i) Firstly, at the provincial or policy/administration level: These include document reviews or program documents, key person interviews, a study of budget outlays and expenditures, fund disbursement mechanisms, administration of the respective program and accreditation procedures for private providers. These were done at the level of the Departments of Health in both provincial capitals in the study provinces and pertain to the program as in its totality across the study province.
(ii) Secondly at the level of the districts: Each province is divided into administrative units called districts (population of a district 1-1.5 million on average).Each district has its own district health administration to oversee the functioning of the health system within its boundaries.
(iii) Finally at the level of the village, eligible beneficiaries will be identified (identified by possession of a BPL card). Users will be compared with a comparison group of eligible non users using matching. Users and non-users within and between each program were compared with regard to background characteristics, outcomes and other indicators
Overall Aim: The overall aim of this proposal is to develop methodology and apply this to assess the impact of and compare two large scale programs for the financing and provision of maternal healthcare in India, in terms of a) health outcomes (maternal and perinatal), b) the influence of the program and health service and social environment structure on outcomes, c) extended benefits of the program, d) private health sector contribution to outcomes and e) financial sustainability. The two programs will also be indirectly compared to rural health insurance in China.
Specific Objectives:
At the provincial (policy) level
i. To study and compare trends reported by each program over time since inception, in terms of: uptake; institutional EmOC participation; private sector participation; and recorded maternal and neonatal outcomes (including institutional delivery, caesarean section rates, complicated delivery and mortality within the programs).
ii. To explore policymaker views on the implementation of programs at the policy level (ie roles of the ANM (Auxiliary Nurse Midwife) and ASHA (Accredited Social Health Activist), human resources for maternal healthcare, NRHM as a platform for maternal health care delivery, private participation in the JSY program)
iii. To study the two emergency obstetric transportation models utilized in each of the programs with regards to ease of access, functionality, utilization and cost of the service. Further, to explore among a group of mothers their experience of gaining access to and using the transportation.
At the level of the public and private EmoC facilities in selected study districts:
iv. To assess and compare the coverage of each program in terms of provision of EmOC and utilization by target mothers and equity in geographic access.
v. To describe and compare the institutions (public and private separately) providing EmOC under the respective programs and the quality of care offered by performing a detailed facility survey, record reviews and observations in the participating institutions.
vi. To assess the influence of each program on the number of institutional deliveries taking place in these facilities among the target group through a time series analysis of the proportion of program births out of all births over a three year period.
vii. To study and compare delivery outcomes under the programs separately for private and public facilities in terms of: Caesarean section proportion, complicated delivery proportion, average lengths of hospital stay, maternal and perinatal mortality between program mothers and non participating women otherwise delivering at the EmOC facilities (public and private sector).
viii. To explore and contrast in the study districts, the overall experiences of EmOC providers participating in the respective programs, including particular motives for participation, human resource implications,, public/private interaction and willingness (and prerequisites) for continued long term involvement.
ix. To study the eligible non participating private sector EmOC providers (in CY) in the study districts, compare these with participating private institutions and explore reasons for non participation.
At the level of mothers in the community (village level):
x. To estimate maternal complication rates in the community with a population based survey.
xi. To compare eligible non participant mothers with participant mothers under the respective programs in terms of: their characteristics, their receipt of maternal health care, the direct and indirect costs of this care, their reasons for participation/non participation and their reasons for their choice of facility.
xii. To study and compare the receipt of comprehensive maternal healthcare by women participating in each program in terms of: the rate of registration of pregnancy; receipt and source of full antenatal care; place of delivery; type of delivery; duration of hospitalization (if applicable); transportation for delivery; post natal care and receipt of incentives (JSY).
xiii. To study and compare the broader influence of each program on wider health outcomes (measured by infant survival, and infant weight).
Overall:
xiv. To study financial sustainability for each of the programs by comparing costs of the two financing systems (though a cost utility analysis and a multivariate analysis, including sensitivity analysis) and by analyzing cost trends both from financing agency perspective and from the perspective of mothers.
xv. To compare the findings on the two programs in India with findings from studies on the impact of NCMS on maternal care in China.
xvi. To disseminate the results of this assessment widely among stakeholders through workshops, academic scientific publications and appropriate national and international forums.
Project Results:
Introduction:
The MATIND project comprised the evaluation and comparison of the impact (including cost effectiveness) of two large scale health programs targeted towards reducing maternal mortality in India. Both the proposed programs were innovative, a first of their kind, in that there have not been similar demand side programs with an explicit focus on maternal healthcare (which in it itself is a new focus area for a large scale innovative program). Also both these programs were large scale programs, implemented by governments over large regions and not pilot experiments. No such large scale programs have been evaluated before though there have been repeated calls in scientific literature for the effectiveness of such programs to be studied.
Also maternal mortality continues to remain a problem for the poor largely because of access barriers to emergency obstetric care. In 2015, it remains one of the MDGs to remain unachieved, though progress has certainly been made. The pending agenda around maternal mortality is expressed in the sustainable development goal to reduce maternal mortality to <70 globally. India still contributes 70% of maternal deaths in south Asia and to 19% of maternal deaths globally.
Demand side financing as a mechanism to allow poor mothers to access institutional EmoC has never been implemented before on a large scale. The proposal has been a first in terms of assessing the two different demand side financing mechanisms employed in large scale programs. Further, both these programs are targeted toward Millennium development goal 5 (progress towards which has not been on track), to reduce maternal mortality.
The private health sector plays an important role in the provision of healthcare in many low income settings, their contribution as part of a large program to achieving a public good (safe delivery in this case) has not been assessed before. Also there is a current push for innovative mechanisms towards the achievement of universal health coverage in low middle income settings, often in partnership with the private sector. This evaluation has thrown light on this aspect.
Summary of the programs studied and objectives of the project:
MATIND studied two state run programs, Janani Suraksha Yojana (JSY) and Chiranjeevi Yojana (CY), designed to reduce financial access barriers that preclude women from obtaining emergency obstetric care in India. MATIND assessed the following: (1) trends in programs uptake since programs inception, (2) the influence of the programs on institutional delivery rates, maternal and neonatal outcomes, (3) service providers’ experiences, (4) emergency transportation systems for obstetric referrals, (5) characteristics of the facilities participating in the programs, (6) quality of care administered, (7) experiences of the beneficiaries and non-beneficiaries, (8) the role of community health workers, and (9) the degree of subsidy to users because of the programs.
The main S&T results are presented at the three levels at which the programs were studied:
(i) The policy/ state level: policy makers, program desginers and other stakeholder influencing maternal health at a policy level were included. Secondary data on the program from national and subnational surveys was also used.
(ii) The facility level: this includes the results of studies that were conducted at the level of facilities that participated in the respective programs
(iii) The community level: user and community level studies on the two programs
The results are presented for the JSY program first and then for the CY program.
The JSY program
Policy level
Has there been a relationship between JSY program utilization and reduction in maternal mortality?
India has had one of the fastest declines in Maternal Mortality Ratio (MMR); however the current number of maternal deaths is still unacceptably high. There has been limited success with supply–side interventions under the Reproductive and Child Health program (RCH-1 1999–2004) in raising the proportions of skilled attendance at births. Within this context, the Indian government launched a nationwide cash transfer program, Janani Suraksha Yojana (JSY) in 2005. The JSY aims to reduce maternal and neonatal mortality through the promotion of institutional births by providing cash incentives to women who give birth in a health institution. But has there been an association between JSY uptake and MMR reduction?
We looked into this question through two separate analyses conducted in 1) Madhya Pradesh and 2) nine low performing states of India, somewhat similar to Madhya Pradesh. The studies describe the uptake of the JSY, the rise in institutional delivery between 2005 and 2010 and estimated the change in district MMR over the same period. Further, for Madhya Pradesh, we estimated the impact of JSY on the reduction in maternal mortality by examining the association between JSY inputs and the MMR estimates. We reference data from a wide range of sources and synthesize these using modern statistical modeling techniques. We also describe trends in uptake of institutional births after the initiation of the JSY and studied the association between institutional birth proportions and the MMR in the nine states
In Madhya Pradesh the proportion of all institutional deliveries increased from 22% in 2005 to 56% in 2010 province wide. The proportion of JSY-supported institutional deliveries rose from 14% (2005) to 80% (2010). MMR declines in the districts varied from 2 to 35% over this period. For the state as a whole, MMR declined by 12% between 2005 and 2010. Despite the marked increase in JSY-supported delivery, we did not detect a significant association between JSY-supported delivery proportions and changes in MMR in the districts. Changes in maternal mortality and institutional deliveries (both JSY and non-JSY) in the districts of Madhya Pradesh between 2005 and 2010 Analysis of nine low performing states showed that proportion institutional births increased from a pre-program average of 20% to 49% in 5 years. The proportion of institutional births had a small negative correlation with the district MMR. The multivariable regression model did not establish significant association between institutional birth proportions and MMR [CI:-0.10 0.68].
Our study confirms previous report that the proportion of institutional delivery has climbed steeply in Madhya Pradesh and other low performing states in India. However, we were unable to detect a significant association between the rising proportion of institutional births and reduction of MMR both in Madhya Pradesh and other low-performing states.
Qualitative Enquiry: To explore the perceptions of JSY program managers, implementers and non-governmental stakeholders on the JSY program specifically on it’s appropriateness in the given context, effect on maternal mortality, implementation challenges and sustainability of the program and it’s achievements.
We conducted semi structured interviews with public health experts working on maternal health issues and health officials in the state of Madhya Pradesh. The experts were selected purposively and included those well known for their contribution to the maternal health field in India. They work in diverse capacities such as academic researchers, members of national committees for health program design and evaluation representing civil society organizations and UN agencies. The health officials included those working at key positions in the state health department and those working at sub-district level who are presently responsible for implementation of the JSY. Health officials were from Madhya Pradesh, a less developed central Indian state with poorer health indicators than national levels. Thus a total of 11 interviews were conducted with maximum variation sampling keeping relationship to the JSY program as the key sampling principal.
Explorations from our interviews are presented under the following themes:
Analysis of central problem – low institutional delivery: is this a problem with people or with system? : The JSY program aimed to address the problem of low utilization in facility care for childbirth. There were contrasting views among the respondents regarding the reason for low utilization. Most health officials perceived the problem as reluctance of women for facility delivery and hence found JSY an appropriate intervention to attract women to facilities, while non-governmental experts viewed it as oversimplification of complex maternal health issues. Some experts challenged the underlying assumptions of the program; one argued that ‘blaming people’ for low utilization is ‘not an honest analysis of the problem’, since there is a trust deficit between the health system and people which is the reason for low uptake.
Underlying models of development and health improvement: This theme presents the range of views about the notion of demand generation. Two models of health improvement were implicitly or explicitly drawn on by participants in interpreting the concept of demand generation for facility birth. Some participants viewed non-utilization of health facility for childbirth is a problem of individual or family behaviour, and hence approved the need of a stimulus (such as a financial incentive) to change behaviour towards institutional births- aligning with the behaviourist model of development. In contrast, some participants viewed health system and social structures as the problem and emphasized the importance of an enabling environment where people are empowered to take rational decisions and health systems are in place to meet the demand; aligning with an empowerment model of development. Another expert suggested that demand generation through monetary incentives could increase the risk of peoples’ dependency on the incentive to access the services.
With regard to the sustainability of funding from government to this large scale program, some respondents saw it as inherently unsustainable in the long term.
Quality of care: One official and several experts attributed the problem to poor quality of care and agreed on the need to improve quality of care for reduction in MMR. Several experts expressed the view that there had been a decline in quality of care offered at health facilities due to the increased load of institutional deliveries, driven by the JSY, on the weak health system. They pointed out that there were limited efforts to improve the supply side capacity and felt that MMR will decrease faster only if quality of care improves. Both official and experts highlighted systemic problems in service delivery during the JSY program, such as bed shortages, inadequate supplies and human resource issues including under-staffing, lack of competencies among staff, and poor attitudes and behavior of staff.
Respondents had opposing views on the rationale of a cash transfer to promote institutional delivery. However there was consensus on the problem of a weak supply side that could undermine the goal of maternal mortality reduction by the program.
Has the program resulted in reduced inequalities in institutional delivery and maternal mortality reduction across income groups?
The proportion of women giving birth in health institutions has increased sharply in India since the introduction of cash incentive program, Janani Suraksha Yojana (JSY) in 2005. JSY was intended to benefit disadvantaged population who had poor access to institutional care for childbirth who bore the brunt of maternal deaths. Hence, it is important to look at the increase in institutional deliveries following the implementation of JSY from an equity perspective.
We analyzed data on utilization of institutional delivery and maternal mortality ratios from nine Indian states, including Madhya Pradesh. We used area-based socioeconomic measures (ABSM) for analysis of inequalities. The concentration curve, concentration index (CI), Slope Index and Relative Index of Inequality were used to analyze the data.
Among the poorest districts, institutional deliveries increased from an average of 16% to 45%, while among those in the richest districts, the increase was from an average of 40% to 69%. Although inequality in access to institutional delivery care persists, the degree of inequality during the JSY period (2010) was lower than in the period before JSY (2004-06).
Differences in male literacy contributed 30% to overall inequality in institutional delivery. Emergency Obstetric care availability (% of C-sections in public facilities in each district was used as a proxy measure) contributed 20%. The proportion of poorest households in district contributed 18%.
The richest division experienced 97 fewer maternal deaths per hundred thousand live births than did the poorest division during 2007-09, and 135 fewer maternal deaths during 2010. This reveals increased inequality as a consequence of a lower decline in MMR in poorer divisions than in richer ones. Reduction in the MMR during JSY (2007-2009 to 2010) was estimated as being four times higher in richest division quintile than in the poorest one.
Our analysis confirms that although inequality in access to institutional delivery persists, it has been reduced since the JSY program began. The presence of higher maternal mortality with slower pace of decline in the poorest area as well as inequalities in the availability of EmOC facilities during the cash incentive program suggest that the cash incentive alone is not sufficient to achieve equity in maternal health outcomes. Rather, the cash incentive program needs to be supported by the provision of quality health care services including EmOC and improved targeting of disadvantaged populations for the cash incentive program could be considered.
How did public private partnerships under the JSY work?:
The department of health in Madhya Pradesh involved the private sector in JSY program in 2006. The partnership was called the Janani Sahyogi Yojana (JShY or maternal support scheme). Under the JShY partnership, private accredited obstetricians were paid by the state to perform the deliveries of women below poverty line. Between 2007 and 2011, 35000 births occurred in the private sector under the JShY.
We explored the motives of private obstetricians in Madhya Pradesh for joining the JShY PPP, their experiences within the PPP, the benefits and disadvantages of being part of the PPP; their interactions with the state as part of the PPP and motives for withdrawal among those who withdrew from the scheme. We also interviewed officials overseeing the implementation of the JShY at the state department of health to present their views on the JShY and the difficulties encountered. This study throws light on the dynamics of a PPP for obstetric care from the perspective of private sector obstetricians and state managers of the PPP. The findings of the study will inform the design, feasibility, prerequisites and potential pitfalls for such a partnership for obstetric care in this or other similar low income settings.
Private obstetricians who had participated or currently participate in the program were identified based on a listing provided by the department of health in Madhya Pradesh. In each district, private obstetric facilities were purposively selected to represent for profit and charitable hospitals, hospital of different sizes and facilities that were purely obstetric and those that were part of larger multispecialty health service providers.
Participants in the JShY public private partnership program believed the partnership contributed to raising the opportunities for vulnerable women to give birth safely in a setting where they would have access to EmOC. They perceived that this would contribute to a reduction in maternal mortality. Participation was often because of altruism or because private obstetricians saw it as a way to build practices or have more experience. However at the same time there were a number of deterrents to participation, some of which were severe enough for some participants to withdraw from the partnership. One of these was the clustering of high risk cases, who the private obstetricians had never seen before, who often came with severe complications. The other issue raised was that linkages to access blood required had not been made. This was particularly important given the need for blood among potential program beneficiaries who were often are anemic or have complications like post-partum hemorrhage. Besides this clustering of cases, private sector participants reported negative experiences with the paper work and obtaining reimbursements from the government. The extremely high rate of cesarean sections has been of concern to the government who believed that the structure of the reimbursement (with higher rates of reimbursement for cesareans) drove the extraordinarily high rates. However the private obstetricians countered this saying that because of their facilities being specialist ones with full time obstetricians and operating facilities, they tended to have more referred complicated cases which drove up rates. They were not supportive of the governments new reimbursement structure which had an embedded disincentive for cesareans (paid a fixed rate for births 100 women regardless of type of delivery). The new structure caused some of the private sector partners to withdraw. Other concerns include misuse of the scheme by people who did not really deserve but had the necessary paper work. On the whole the private partnership under the JSY was rather small and now stands suspended because of these differences.
Quantitatively, we looked at births in the JShY program from secondary data. In JSaY in MP (program with differential payments for caesarean and vaginal delivery) the CS rate increased over four years, from 26.6% to 40.7% in the participating private hospitals. This trend is all the more surprising, given that the background rate for caesarean deliveries in Madhya Pradesh, where the JSaY was implemented, was only 4.9%. This resulted in the government withdrawing the program and then reintroducing it in as a bundled payment similar to the CY. However private provider participation in the program has waned since.
Facility level
The level of Emergency obstetric care under the JSY program
India has implemented a successful national cash transfer program, the Janani Suraksha Yojana (JSY), to incentivize women to deliver in facilities. It has resulted in a steep rise in facility delivery across the country. As increased utilization of facilities for childbirth is the focus of the JSY, the functionality of the facilities is critical to improved outcomes. We studied the availability and level of provision of emergency obstetric care (EmOC) signal functions in facilities implementing the JSY program in three districts of Madhya Pradesh (MP) state, central India. These are measured against World health report benchmarks. We also study the contribution of private sector facilities to the provision of EmOC in these districts.
A cross-sectional survey of all healthcare facilities that had done at least one delivery a month was conducted between February 2012 and April 2013. EmOC signal functions performed in each facility were recorded, as were human resource data and delivery numbers in each facility
A total of 152 facilities were surveyed (118 program facilities). 86% of childbirths occurred at program facilities, two thirds of which occurred at less-than-BEmOC facilities. All these less-than-BEmOCs facilities performed less than 4 basic signal functions. None of the less-than-CEmOC facilities performed all BEmOC functions. Program facilities provided few EmOC signal functions besides parenteral antibiotic or oxytocic administration. CEmOC provision was predominantly non-program (private): only one of six CEmOC facilities was in the program (public sector). Only 13% of all qualified obstetricians practiced at program facilities. Lack of supplies and insufficient training were the two most common reasons cited for not performing signal functions at program facilities.
Given the high uptake, the JSY program has an opportunity to contribute to reduction in maternal and perinatal mortality in this setting. However, for more program impact on outcomes there needs to be a significant improvement in the provision for EmOC. Among other improvements, a serious focus on training of skilled birth attendants to build competency to enable EmOC provision is necessary.
Quality of care in the institutions in the JSY program:
The JSY program has been successful in raising the proportion of facility births from 30% in 2007 to 73% in 2012. However, decline in maternal mortality ratio has followed a secular trend over the past decade. Poor quality of care (QoC) is a potential reason for this paradoxical situation. Despite the success of the JSY in raising utilization, little is known about the actual QoC provided under the program despite initiatives to improve this. We present here results from four empirical sub studies on aspects of quality of obstetric care in the JSY program in Madhya Pradesh (MP) province conducted during 2012-13 as part of the MATIND project.
Quantitative and qualitative studies were conducted to assess QoC at JSY facilities in three MP districts. Data collection included cross-sectional surveys, participant observation, retrospective case record review, and spatial analysis of transfer time between referring facility and hospitals of those other who died during study period.
Emergency Obstetric Care (EmOC) signal functions at JSY facilities: 86% of institutional births during study period occurred at JSY facilities. Out of those, 66% were in facilities that provided less than the basic EmOC. JSY facilities provided few emergency obstetric care signal functions (EmOC) besides parenteral antibiotic or oxytocic administration. Staff competence under the JSY program
Intrapartum care & staff competences need improvement: Observations revealed that care given during the labor and delivery under the JSY program did not meet standards for skilled birth attendance (SBA): practices were poor, care was not always respectful, and the work environment was not conducive for SBA.A study showed that frontline delivery room nurses (n=233) had poor competence in providing first line management for two common obstetric complications: hemorrhage and eclampsia. In general, the staff obtained low scores (median score= 5/20) on case scenarios to study provision of first line EmOC for common complications. Only 14%, 58%, & 20% were competent at initial assessment, diagnosis, and treatment respectively. The proportion of competent respondents for the initial management of eclampsia (Ecmp), antepartum hemorrhage (APH), and postpartum hemorrhage (PPH) was low. Deficiencies in training of nurse midwives, clinical supervision, and poor practice environment were possible reasons for poor staff performance. In addition, shortages of staff, increased work load since the JSY, and deficiencies in the infrastructure/supplies were reported.
Poor referral patterns and incomplete records: A study of obstetric case referral patterns showed that secondary obstetric care facilities made most referrals and 40% were for conditions expected to be treated at this level. High number of maternal deaths despite spatial access to EmOC i.e. referral from facilities within the desired two hour transfer time, indicate poor quality of referral services. Records reviewed were largely incomplete; indicating poor documentation of care - the proportion of records documenting key clinical parameters was low e.g. 14% documented postnatal blood pressure, 35% documented fetal heart rate.
QoC under the JSY program needs focused attention. This current QoC has restricted the translation of coverage gains from JSY into improved health outcomes. Improvement in the quality of nursing and midwifery education is a crucial step to create a technically competent workforce rather than merely qualified personnel. Adoption of training methods that build the confidence and competence to provide life-saving care is important for the effective functioning of nurse midwives in peripheral facilities. Encourage processes that improve QoC like maintenance of case records. Supportive supervision during training, as well as on the job, could be an important step to bridge the current gaps in competence.
Staff competencies under the JSY program
Access to emergency obstetric care by competent staff can reduce maternal mortality. India has launched the Janani Suraksha Yojana (JSY) conditional cash transfer program to promote institutional births. During implementation of the JSY, India witnessed a steep increase in the proportion of institutional deliveries-from 40% in 2004 to 73% in 2012. However, maternal mortality reduction follows a secular trend. Competent management of complications, when women deliver in facilities under the JSY, is essential for reduction in maternal mortality and therefore to a successful program outcome. We investigate, using clinical vignettes, whether birth attendants at institutions under the program are competent at providing appropriate care for obstetric complications.
A facility based cross-sectional study was conducted in three districts of Madhya Pradesh (MP) province. Written case vignettes for two obstetric complications, hemorrhage and eclampsia, were administered to 233 birth attendant nurses at 73 JSY facilities. Their competence at (a) initial assessment, (b) diagnosis, and (c) making decisions on appropriate first-line care for these complications was scored.
The mean emergency obstetric care (EmOC) competence score was 5.4 (median = 5) on a total score of 20, and 75% of participants scored below 35% of the maximum score. The overall score, although poor, was marginally higher in respondents with Skilled Birth Attendant (SBA) training, those with general nursing and midwifery qualifications, those at higher facility levels, and those conducting >30 deliveries a month. In all, 14% of respondents were competent at assessment, 58% were competent at making a correct clinical diagnosis, and 20% were competent at providing first-line care.
Birth attendants in the JSY facilities have low competence at EmOC provision. Hence, births in the JSY program cannot be considered to have access to competent EmOC. Urgent efforts are required to effectively increase the competence of birth attendants at managing obstetric complications in order to translate large gains in coverage of institutional delivery services under JSY into reductions in maternal mortality in Madhya Pradesh, India.
Case record maintenance under the JSY program:
To gain insight into the quality of care in facilities implementing the Janani Suraksha Yojana (JSY) cash transfer program in Madhya Pradesh, India, by reviewing the level of documentation in the clinical records of women who delivered.
The present retrospective, descriptive study reviewed case records of women who delivered at 73 primary, secondary, and tertiary level facilities in three districts of Madhya Pradesh between 2012 and 2013. Twenty elements of care were assessed encompassing clinical history and admission details, care during delivery and postnatal period, and discharge details.
A total of 1239 records were reviewed. The extent of documentation varied among the elements assessed—e.g. 24 (1.9%) records documented advice at discharge, 171 (13.8%) documented postnatal blood pressure, 437 (35.3%) documented fetal heart rate, and 1220 (98.5%) documented admission date. The extent of documentation was better at higher level facilities.
The quality of clinical documentation in the JSY program was found to be unacceptably poor in Madhya Pradesh. Improving staff skills and practices in clinical documentation and record keeping will be required to enable clinical processes to be assessed and quality of care to be improved.
Referrals services under the JSY program:
In 2009 the state government of Madhya Pradesh, India launched an emergency obstetric transportation service, Janani Express Yojana (JEY), to support the cash transfer program that promotes institutional delivery. JEY, a large scale public private partnership, lowers geographical access barriers to facility based care. The state contracts and pays private agencies to provide emergency transportation at no cost to the user. The objective was to study (a) the utilization of JEY among women delivering in health facilities, (b) factors associated with usage, (c) the timeliness of the service.
A cross sectional facility based study was conducted in facilities that carried out > ten deliveries a month. Researchers who spent five days in each facility administered a questionnaire to all women who gave birth there to elicit socio-demographic characteristics and transport related details.
35% of women utilised JEY to reach a facility, however utilization varied between study districts. Uptake was highest among women from rural areas (44%), scheduled tribes (55%), and poorly educated women (40%). Living in rural areas and belonging to scheduled tribes were significant predictors for JEY usage. Almost 1/3 of JEY users (n = 104) experienced a transport related delay.
The JEY service model complements the cash transfer program by providing transport to a facility to give birth. A study of the distribution of utilization in population subgroups suggests the intervention was successful in reaching the most vulnerable population, promoting equity in access. While 1/3 of women utilized the service and it saved them money; 30% experienced significant transport related delays in reaching a facility, which is comparable to women using public transportation. Further research is needed to understand why utilization is low, to explore if there is a need for service expansion at the community level and to improve the overall time efficiency of JEY.
Availability of blood under the JSY program:
In order to study the availability, distribution, and functioning of blood transfusion services for obstetric care in Madhya Pradesh secondary data regarding location and service outputs of blood banks and blood storage systems in the public and private sector in MP was obtained from concerned government departments at the state. Facility locations were mapped on Geographical Information System (GIS). GIS Analysis techniques were used to study the availability and geographic distribution of blood banks/storage units.
A qualitative study was performed to understand provider’s perspective, facilitators, and barriers to provision of blood availability. In total, 20 interviews were conducted in Ujjain and Shahdol districts. We interviewed senior officials handling managerial work in the blood bank at districts, senior and junior obstetricians, medical officer at block level to explore opinions and experiences about the provision and use of blood services in general and specifically for obstetrics cases in Madhya Pradesh.
In total list of 145 blood banks/storage units were obtained. Out of these Private (57), Civil and Other/Govt. Hospital (17), Society/Charitable Trust (14), District Hospital (44), Medical Collage/Research Centre (13). The distribution of blood allowed access to most villages/ towns in each district if a buffer of 50 KM around blood banks was considered.
Study participants highlighted various challenges for providing effective blood bank services in the districts. The process of getting a license from FDA was felt cumbersome due to several conditions. The system to refer maternal cases was disused and many times it was observed that those needing blood came late to facilities that could perform life-saving interventions like blood transfusions due to an ineffective referral system. Blood bank managers also raised the issue of limited human resources and non-functioning blood storage units at periphery facilities that overloaded the district blood banks. Many expressed the opinion that the number of blood storage units should be increased, mainly where the load of patients was highest and where more deliveries were happening. Issues of high anemia rates among women were also raised. Awareness among patients regarding the blood replacement was also highlighted during the interviews. The issue of a ‘professional donor’ (i.e. someone who donated blood on a regular basis for money) was also raised during the interviews.
At community level
In the community level, phase a survey of women in sampled villages was done to ascertain JSY uptake at community level. Of the 2607 women surveyed, 1995 were JSY beneficiaries, i.e 76% of births occurred under the progam. Among women who delivered at home (9%) the main reason given for home deliveries was that the baby came unexpectedly and quickly (n=312, 52%). Other reasons included; planned to have a home delivery (n=97, 16%), transportation related issues (n=95, 16%), no one to accompany them to the hospital (n=58, 10%) and other (n=37, 6%). Only one woman replied she could not afford to deliver in a health facility. 15% of women delivered in private sector facilities. 96% of births under the JSY program were vaginal deliveries, while only 4% were cesareans. However the performance of cesarean sections was much higher among women who delivered in facilities outside the program – 35%.
Complications treated under the program: Among the entire study sample, there was a 9% complication rate. This rate did not significantly differ between JSY beneficiaries and non-beneficiaries. Women had a median of one complication, and this was also similar in both groups. The prevalence of specific complications was similar in both groups with the exception of malpresentation. This is most likely the case for the differences in ‘Time until placenta came out’ and ‘Placenta manually removed’ as well. However, we believe the high number of caesarean sections in the non-beneficiary group is driving this difference. Pre-eclampsia (4.7%) and post-partum hemorrhage (2.2%) were the most common complications among both JSY beneficiaries and non-beneficiaries.
Out of pocket expenditure among women who delivered under the program and those who did not: Nearly women who delivered under the JSY program had some out of pocket expense. Women who delivered under JSY program had a higher median, IQR OOPE ($8, 3-18) compared to at home ($6, 2-13). Among JSY beneficiaries, poorest women had twice the net gain ($20) versus the wealthiest ($10) post cash transfer. Informal payments (64%) and food/baby items (77%) were the two most common sources of OOPE. OOPE made among JSY beneficiaries was pro-poor; poorer women made proportionally less expenditures compared to wealthier women. In an adjusted model, delivering in a JSY public facility led to increased odds of incurring expenditures (OR: 1.58 95% CI: 1.11-2.25) but at the same time to a 16% (95% CI: 0.73-0.96) decrease in the amount paid compared to home deliveries. Wealth was not a predictor of having OOPE, but it was an indicator of how much the women would pay. Among women who delivered in a private facilty the median out of pocket expense was nearly 20 times that of JSY mothers.
Participation and non- participation in the JSY program: views from the community
One-fifth of all global maternal deaths occur in India. Further, only 39% of Indian women delivered in a health facility in 2005-06. It was within this context in 2005, that the Indian Government implemented the Janani Suraksha Yojana program that incentivizes poor women to give birth in a health facility by providing them with a cash transfer upon discharge. JSY has helped raised institutional delivery to 74% in the eight years since its implementation. Despite the success of the JSY in raising institutional delivery proportions, the large number of beneficiaries, and the cost of the program, there have been few qualitative studies exploring why women participate or not in the program. The objective was to explore reasons why women choose to participate or not in the JSY program.
In March 2013, we conducted 24 individual in-depth interviews with women who delivered the previous 12 months in two districts of Madhya Pradesh, India. Qualitative framework analysis was used to analyze the data.
Our research found that most women participate in the JSY program because of a shift in the social norm towards delivering in an institution. Drivers of the shift include social pressure from the ASHA and dai to deliver in a health facility, and a growing individual perception of the importance for ‘safe’ and ‘easy’ delivery. Whilst the incentive was an important influence on many women’s behavior, others did not perceive it as an important consideration in their decision to deliver in a health facility. Many women reported procedural difficulties to receive the benefit. Retaining the cash incentive was also an issue due to out-of-pocket expenditures incurred at the facility. Non-participation was often unintentional and caused by personal circumstances, poor geographic access or driven by a perception of poor quality of care administered by program facilities. Persistent issues with quality of care and out-of-pocket expenditures have not received sufficient attention under the program.
The Chiranjeevi program:
Policy level:
The contribution of the CY to institutional delivery in Gujarat:
Many low-middle income countries have focused on improving access to and quality of obstetric care, as part of promoting a facility based intra-partum care strategy to reduce maternal mortality. The state of Gujarat in India, implements a facility based intra-partum care program through its large for-profit private obstetric sector, under a state-led public-private-partnership, the Chiranjeevi Yojana (CY), under which the state pays accredited private obstetricians to perform deliveries for poor/tribal women. We examine CY performance, its contribution to overall trends in institutional deliveries in Gujarat over the last decade and its effect on private and public sector deliveries there.
District level institutional delivery data (public, private, CY), national surveys, poverty estimates, census data were used. Institutional delivery trends in Gujarat 2000–2010 are presented; including contributions of different sectors and CY. Piece-wise regression was used to study the influence of the CY program on public and private sector institutional delivery.
Institutional delivery rose from 40.7% (2001) to 89.3% (2010), driven by sharp increases in private sector deliveries. Public sector and CY contributed 25–29% and 13–16% respectively of all deliveries each year. In 2007, 860 of 2000 private obstetricians participated in CY. Since 2007, >600,000 CY deliveries occurred i.e. one-third of births in the target population. Caesareans under CY were 6%, higher than the 2% reported among poor women by the DLHS survey just before CY. CY did not influence the already rising proportion of private sector deliveries in Gujarat.
Under the CY, there have been >600,000 beneficiaries. While caesarean proportions are higher under CY than before, it is uncertain if all beneficiaries who require sections receive these. Other issues to explore include quality of care, provider attrition and the relatively low coverage.
The CY program: policy maker perceptions on program and implementation challenges experienced in the CY public private partnership
We did interviews with key informants in the state government, non governmental stakeholders and the obstetric gynecology professional body in the state. We explored the perceptions of stakeholders on the implementation Chiranjeevi Yojana, and the implementation challenges experienced in the CY public private partnership
Participants were selected using purposive, non-probability sampling among key figures in government, professional membership organizations and experts involved in the conceptualization and implementation of Chiranjeevi Yojana. Participants represented government stakeholders, members of obstetric and gynecological professional organizations, and experts involved in the design and implementation of CY. All interviews were conducted in the state capital or in their districts.
Efforts were made to obtain interviews with key informants with a diverse range of perspectives. This enabled us to identify what perspectives and experiences around Chiranjeevi Yojana were common to the informants, as well as areas of difference. Snowball sampling was used to contact informants likely to provide insight into research questions.
There was a divergence of views between government and non government stakeholders on the place of the CY in the long term vision of health services. The government officials saw CY designed as temporary public private partnership till public sector EmOC services are adequate. However other stakeholders felt that if public health services had failed to deliver adequate infrastructure and resources to provide women safe hospital childbirth services in the last six decades, it might suggest that the program was likely to continue to be needed in the long term
A program with strengths and weaknesses: Respondents felt that program had strengths -CY has made more choice available to poor women in rural areas, it seems to have reduced out of pocket costs and has reduced mortality and morbidity compared to other options.
However the program was also challenged by implementation weaknesses: On the other hand, there were problems with low uptake, low coverage among the poorest, lack of interdepartmental coordination (particularly around the issuance of BPL (below poverty line) cards, low awareness about the scheme was also cited as a reason for poor uptake. Problems with documentation around the BPL card were also reported as being responsible for low uptake. Also though the intention of the program was to provide free intra-partum care, beneficiaries reportedly had to make some payments, as both public and private sector respondents asserted. A public sector government official actually justified private obstetricians taking some payment from CY beneficiaries as there were long delays in receiving reimbursement from the government! Another difficulty was the attrition of private obstetricians participating over time. A common difficulty faced by the government department in implementing the program was that policy was often driven by individual leadership. Thus when the leadership changed, the policy also changed. This was experienced with regard to waxing and waning commitment to the CY program.
A number of stakeholder commented on the poor monitoring of the CY program by the state. There was also progressive mistrust of the state by the private sector in the context of CY. In general, private obstetricians felt that the government made key decisions regarding the partnership on its own. Private providers were upset over having to provide services to undeserving clients, simply because they had a BPL card. They felt the state had failed to identify poor persons correctly, because of which they were being exploited. They also resented delays in the receipt of the reimbursements.
In the long run, stakeholders perceived that there could be some difficulties with long term sustainability of the CY, not because of financial reasons but because of changes in policy stemming from frequent transfers of key decision makers. The state government perception was that the program would end once the state was strong enough, though district level officials perceived the contrary.
In conclusion, there seemed to be uncertainty on the long term future of the CY program for a number of reasons but largely because support to the program was subject to the views of the incumbent policy makers. This support waxed and waned at periodic intervals. There were issues with the functioning of the partnership, they were lack of oversight from the government, difficult in smooth functioning at the interface of the public private partnerships, possibly because of this being a first large scale partnership. Though it was well designed, there was inadequate focus on the implementation issues, putting in place the required processes well before the program started and a lack of experience on the part of both partners with dealing with a partnership.
Facility level:
Level of EmOC functioning in the the CY program in Gujarat:
The level of EmOC functioning of program facilities is comparable to eligible non-program facilities and much higher than public sector facilities. 35% of CY facilities functioned at CEmOC levels, meaning they had the ability to manage life-threatening complications. 25% of non-program facilities functioned at this level (and <15% in the public sector). All CY program facilities were staffed with fulltime obstetricians and gynaecologists (vs. 63% in non program facilities), the facilities had a high proportion of nursing staff, both trained and untrained. The program facilities though lacked in house access to ancillary services like pharmacy, urgent ultrasounds and laboratory tests. These needed to be done in separate facilities outside the main facility. Such services were more frequently available in the non-program facilities, possibly because many of these were larger hospitals. However simple but key lab investigations like haemoglobin, typing and cross matching could be done at program facilities. A large proportion of CY program facilities also had key equipment to manage complications like vacuum extractors, outlet forceps as well as equipment for routine labour like mucus extractors. Program facilities had on site emergency drugs like magnesium sulphate, oxytocin, this was less available in non-program facilities. 50% of program facilities had no emergency vehicle for transportation; 25% had some private arrangement for transport. Quality control measures were similar to non-program facilities
The kinds of facilities that participated in the CY program partnership:
Private CEmOC ‘facilities which participated in the CY program were newer, smaller, conducted a lower proportion of deliveries by caesarean sections and were owned by obstetricians with less than 5 years of experience compared to private CEmOC facilities which did not participate in the CY program. One-fourth of private providers located in district capitals and half of those in small block towns participated in the CY program
Emergency Transport in Gujarat, supporting the CY rogram
Due to the unpredictable nature of maternal complications effective transportation services are essential to ensure all women obtain timely medical care. In 2007, 525 emergency transportation ambulances, termed the 108 service were implemented across the state of Gujarat under a public private partnership. This emergency transportation supports the CY program users. This service provides free 24/7 Emergency Response; assisting 41,057 deliveries since inception. This study aimed to establish the characteristics of the emergency transportation service users; the mode of transportation among vulnerable women not utilizing the emergency transportation service & quantify the proportion of women arriving at the health facility <2hours after deciding to leave home.
All facilities conducting >10 deliveries per month in three districts of Gujarat were visited for a five day period between July 2012 and May 2013. Post-‐natally administered questionnaires obtained socio-‐economic and transport characteristics of all women; women residing in rural areas (n=1056) were included in this study.
Privately hired vehicles (54.0%) and 108 emergency transportation service (25.4%) were the most prevalent modes of transportation. 108 emergency transportation service use was higher among women who planned a government facility delivery (53.3%), had low education (31.9%), were poor (31.7%), travelled less than 20km (30.6%) and had no specified delivery plan (30.5%). 68.2% of poor women did not use the 108 service with 24.6% of these travelling to government facilities by hired transportation. 94% of women arrived <2 hours after deciding to leave home and 76.4% <1 hour.
Current transportation networks and road infrastructure in Gujarat enable the majority of women to reach health facilities in less than one hour. The high usage of the 108 emergency transportation service among low educated and poor mothers demonstrate its’ success in strengthening access to maternal care among vulnerable women. However, coverage among poor women is not comprehensive and many continue to utilize hired transportation.
Outcomes in CY facilities and non CY facilities:
We surveyed 901 women giving births in high load facilities (all facilities in our study districts that did >30 deliveries in the last 3 months). A third gave birth in the public sector, the remaining two third delivered in private sector facilties. Half of these women delivered in CY facilities (1/3 of the overall sample) but only half became program beneficiaries (1/6 of all eligible).
Eligible women who gave birth in private non CY facilities had the highest Cesarean rates of 20%. In CY facilities cesarean rates were lower (8%, similar among beneficiaries and non beneficiaries, though the number of Cesareans was too small to make conclusions with certainty). The public sector had the lower proportion of cesareans at 5%. Women stayed in the hospital after a vaginal delivery for 12-18 hours. Complication were equally seen in all three facility groups.
Quality of care:
We used exit interviews where the WHO indicators to assess positive (and some neutral/harmful pratices) that the eligible women were subjected to in the facility. There were no differences in the practices that CY beneficiaries and non beneficiaries in CY facilities were subject to. However women who delivered in non CY facilities were more likely to receive interventions not deemed necessary like intravenous fluids or enemas. Women who delivered in a public facility had the least interventions indicative of positive, neutral or negative care.
Access to blood under the CY program: This study was undertaken in the study districts. A mixed method approach was adopted. Data collected included secondary data from 31 blood banks on different input and output indicators. 26 Interviews with stakeholders such as blood bank managers, pathologists and beneficiaries.
Preliminary analysis shows district level variation in level of access and cost of blood transfusion. Use of blood transfusion services for maternal health varied in 3 study districts with highest proportions in Sabarkantha (28%) and lowest in Surendranagar (15%). Median cost of blood varied between 460 and 940 Indian Rupee in study districts. Cost of blood units also varied across type of blood banks- government, private and private not for profit despite price fixing by National AIDS Control Society. Qualitative interviews with blood bank managers and beneficiaries revealed that low levels of voluntary blood donation is an issue with peripheral blood banks. Changed norms of blood banking make it more difficult for private and smaller blood banks to conduct voluntary blood donation drives. Skilled dedicated human resources for blood banking services is the major challenge in study districts (rural areas) but not for urban areas. In Dahod mothers faced maximum challenge in getting blood and had to pay even after replacement. Professional donors are reality in the rural areas. Waiting time for getting blood bag was about 2 hours after reaching the blood bank. Mothers found expense of blood transfusion very high. The process of obtaining licence for blood bank and storage is a lengthy and complicated process with many agencies involved. The regulations have made blood safe but scarce in rural areas as many blood banks do not have the permission to conduct donation camps. Blood bank managers, especially private blood banks, felt that infrastructure and human resources requirements are at times unreasonable. Majority of blood storage centres are defunct as mother blood banks do not have enough blood units. The process of licencing of blood banks and renewal is tedious and lengthy process. No analysis of data at the district or state level for monitoring and improving access to blood for mothers has been made.
Motives for participation and non participation in the program by private obstetricians:
In India a lack of access to emergency obstetric care contributes to maternal deaths. In 2005 Gujarat state launched a public-private partnership (PPP) programme, Chiranjeevi Yojana (CY), under which the state pays accredited private obstetricians a fixed fee for providing free intrapartum care to poor and tribal women. A million women have delivered under CY so far. The participation of private obstetricians in the partnership is central to the programme's effectiveness. We explored with private obstetricians the reasons and experiences that influenced their decisions to participate in the CY programme.
In this qualitative study we interviewed 24 purposefully selected private obstetricians in Gujarat. We explored their views on the scheme, the reasons and experiences leading up to decisions to participate, not participate or withdraw from the CY, as well as their opinions about the scheme's impact. We analysed data using the Framework approach.
Participants expressed a tension between doing public good and making a profit. Bureaucratic procedures and perceptions of programme misuse seemed to influence providers to withdraw from the programme or not participate at all. Providers feared that participating in CY would lower the status of their practices and some were deterred by the likelihood of more clinically difficult cases among eligible CY beneficiaries. Some providers resented taking on what they saw as a state responsibility to provide safe maternity services to poor women. Younger obstetricians in the process of establishing private practices, and those in more remote, 'less competitive' areas, were more willing to participate in CY. Some doctors had reservations over the quality of care that doctors could provide given the financial constraints of the scheme.
While some private obstetricians willingly participate in CY and are satisfied with its functioning, a larger number shared concerns about participation. Operational difficulties and a trust deficit between the public and private health sectors affect retention of private providers in the scheme. Further refinement of the scheme, in consultation with private partners, and trust building initiatives could strengthen the programme. These findings offer lessons to those developing public-private partnerships to widen access to health services for underprivileged groups.
Does the CY increase geographic access to obstetric services?
The high rate of maternal mortality in India is of grave concern. Poor rural Indian women are most vulnerable to preventable maternal deaths primarily because they have limited availability of affordable emergency obstetric care (EmOC) within reasonable geographic proximity. Scarcity of obstetricians in the public sector combined with financial barriers to accessing private sector obstetrician services preclude this underserved population from availing lifesaving functions of comprehensive EmOC such as C-section. In order to overcome this limitation, Government of Gujarat initiated a unique public–private partnership program called Chiranjeevi Yojana (CY) in 2005. The program envisaged leveraging private sector providers to increase availability and thereby accessibility of EmOC care for vulnerable sections of society. Under CY, private sector providers render obstetric care services to poor women at no cost to patients. This study examines the contribution of the CY in improving availability of CEmOC services between 2006 and 2012 in three districts of Gujarat, India.
Primary data on facility locations, EmOC functionality, and obstetric bed availability were collected in the years 2012 and 2013 in three study districts. Secondary data from Census 2001 and 2011 were used along with required geographic information from Topo sheets and Google Earth maps. ArcGIS version 10 was used to analyze the availability of services using two-step floating catchment area (2SFCA) method.
Our analysis suggests that the availability of CEmOC services within reasonable travel distance has greatly improved in all three study districts as a result of CY. We also show that the declining participation of the private sector did not result in an increase in distance to the nearest facility, but the extent of availability of providers for several villages was reduced. Spatial and temporal analyses in this paper provide a comprehensive understanding of trends in the availability of EmOC services within reasonable travel distance.
Community level
In the community phase on the ground, of eligible women who gave birth in the study villages; 28% gave birth under the program. 21% of women delivered in public sector facilties, 33% delivered in the private sector (a little less than half in CY eligible facilities), 15% delivered at home. We explored why women who gave birth in eligible facilities were not given the benefits of the program; reasons included lack of documentation (2/3), doctor deciding not to provide benefits (1/3).
We studied delivery outcomes among this group. The results from the community based survey gave us different results than the did the earlier facility based study – this is likely because the denominator here includes all women (including home deliveries) and that we obtained a much larger pool of eligible women, resulting in larger sub groups. This allowed us to make more robust conclusions. Overall in the entire cohort there was a 7% cesarean section rate. Among the eligible women who gave birth at a CY facility, there was a stark difference among CY beneficiaries and non beneficiaries – the former group had a cesarean rate of 4.6% while the latter had a cesarean section rate of 13.4%. This would indicate that it is possible that private obstetricians move women who needs cesareans out of the program but within their own facilities. The proportion of cesarean sections among CY beneficiaries (4.6%) was very similar to that among eligible women who delivered in public sector facilities (4.5%). Among women who delivered in the private facilities (on CY), the rate of cesarean (15.6%) was similar to those non CY beneficiaries in CY facilities.
There was also a difference detected at community level as to where eligible women chose to deliver. A significantly higher propotion of primiparous women chose to give birth in private sector facilities than did in public sector ones. Four in five women who delivered at home were multiparous. Among the entire study sample, there was a 17.8% complication rate; 15.7% among CY beneficiaries and 18.6% among non-beneficiaries. This difference was not significant.
The main differences between CY beneficiaries and non-beneficiaries were education, income, caste and delivery type. CY beneficiaries tended to having higher education (median 7 years) compared to non-beneficiaries (median 5 years). However, they tended to be poorer (5000 INR compared to 6000 INR). A higher proportion of CY beneficiaries were from scheduled tribe (55.8%).
There was also a difference in the out of pocket expenses that the women experienced. On the whole women in the cohort had an average of 3037 INR as delivery related expenses. Among CY beneficiaries, this was lower at 1621 INR. In public sector faciliites, this was 618 INR. Among non beneficiaries, there mean OOPE 3572 INR. This trend was similar when disaggregated by vaginal and abdominal deliveries.
We also did qualitative interviews with eligible women in the community who delivered under the program to study their understanding of the program.
Women who delivered under the CY program had a largely economic understanding of what the program provided in relation to their childbirth.
‘Chiranjeevi means getting back 2500 rupees – that is the benefit. What else improvement can it bring? Now people say CY is here- government is paying money. As government is giving grant to Dr X (private doctor), he is giving money back.’ (Family member of parturient).
‘CY is that way that we give the money and then we get it back’.
Women perceived that even though they had to pay upfront, the program did help reduce economic burdens for women and their families.
‘Chiranjeevi Yojana is good treatment. People….if someone has not got money then also, if not today then tomorrow, the doctor will give them money. People will spend 3000-5000 rupees, so if they get this yojana, they will get this benefit.’
There was much importance attributed to ‘filling the form’, the key paperwork required to become a CY beneficiary. It was nearly as if for women/families, if they got the form filled right, they could be sure they would get their money back.
‘We paid but it does not matter. Money will be returned to us as we have filled the form.’
‘If the form is ready, they don’t charge you. If it not ready, then they will charge you.’
One woman reported that her husband had organized to have the form filled. She did not know who had helped him do it but that he had sought help. Another beneficiary said that the detailed paperwork was fixed before delivery.
The program also created an interesting impression on the ‘public-ization’ of private healthcare facilities because of the new program – that deliveries were free in private hospitals now (like in government hospitals).
‘Private hospitals have now become government. So private hospitals distribute money after delivery. CY is good because the plan is for a free delivery’.
However a constant theme which was reiterated was that beneficiaries did pay the private obstetrician for their deliveries upfront and then had this paid back to them once the state paid the private obstetrician. This was a common thread reported by women in the different study districts.
In another set of interviews with women and their families, a number of families had benefitted from CY and said it was a key factor in their decisions about where to give birth; they also expressed a wish to benefit from the scheme again in the future.
However, a number of barriers to using CY were reported: 1) Many facilities were no longer offering CY, which disappointed families who had previously accessed its benefits; in one case, the scheme was still advertised in the clinic but the family were still denied the scheme; 2) the bureaucracy associated with applying to benefit from the scheme was seen as onerous; 3) BPL cards were not transferrable across geographic areas, which caused particular problems for migrant workers to access the scheme; 4) families reported being charged for medicines used during care at private facilities, and not reimbursed for these costs; 5) a few doctors were reported as more eager to offer services under the national insurance scheme RSBY.
Some families in possession of a BPL card perceived that they had no need of the scheme as they could afford to pay for delivery care (and would lose social status by using a government scheme). A few families said that cost was not a key decision-making factor, they would find the money somehow and that cost would not prevent them from seeking private health care.
• CY beneficiaries seem to be paying providers in advance for their delivery. This amount is returned to the woman/ family subsequently, possibly if the woman is approved to be a CY beneficiary by the government depending on the paperwork submitted by the private obstetrician. The private obstetrician repays the woman/family on receiving the claimed reimbursement from the state
• Where families have used CY in the past and local providers have withdrawn CY services, families would like to see it reinstated
• Low awareness and uncertainty about CY may have limited its effectiveness in improving access to delivery care for the poorest.
• Bureaucratic barriers to poor families in accessing CY benefits need to be reduced
• Approaches to preventing additional charges for patients need to be developed. Due to the importance of social networks in disseminating information about facilities and schemes, such overcharging may reduce families’ willingness to use the scheme over time as well as potentially impoverishing poor families further.
• Women generally have limited or only partial input into decision-making about where to give birth. Key decision makers within families need to be reached with information about the CY scheme as well as pregnant women themselves.
Besides this there was also a study of the RHINCAV and CHIMACA projects done. These also looked at healthcare financing in the Chinese context, the CHIMACA particularly for maternal health outcomes.
Project dissemination:
Plans for dissemination, particularly among policy makers, civil society groups and researchers were clearly defined (an entire work package has been dedicated to this) at the outset. As stated, open access scientific publications were published from this project, as far as possible to allow national and international access to reports and evidence coming out of this proposal.
Besides scientific publications, the project included a close collaboration between the consortium and stakeholders including policy makers, civil society groups and researchers in the field. Policy makers and program managers were actively engaged throughout the entire evaluation process. Involvements of stakeholders in the initial methodological consultations were also important. On completion of the evaluation, widespread access to the results were ensured, through the distribution of policy briefs and the conduction of national dissemination meetings with the government of India and the state governments. Getting research into policy has been an important priority of this project. We actively involved local policy makers in the process of the evaluation research, besides making the results readily accessible to them. The process of dissemination and engagement is expected to continue even after the life of project.
At the end of the project, there were formal dissemination meetings program held in India, with participation from all partners on the project. Meetings were held at national level with the government of India, and the state governments of Madhya Pradesh and Gujarat. The findings of the project were disseminated formally at a large meeting. Participants included representatives from the Maternal Child Health Division, Government of India, Indian Council for Medical Research, academia, Federation of Obstetric and gynaecological societies, as well as other stakeholders and development partners in each state working with maternal and child health. In addition the project has presented finding at the number of international conferences, academic papers have been published. More locally dissemination of the relevant findings have been done upto grass root level with community health activists (ASHAs).
As stated above, dissemination of the project findings is expected to continue well after the project has ended
Ethical approvals for the MATIND project: The project received ethical approvals in its entirety from all five participating sites in the respective countries. Periodic updates, major changes in study design were informed to the respective ethical boards as mandated by the country rules. In addition to ethical approval, the project received legal permission for implementation from the Indian Council of Medical Research, New Delhi, India and permission from the Maternal and Child Health Division, Ministry of Health and Family Welfare of the Government of India, New Delhi.
Potential Impact:
This project comprised the evaluation and comparison of the impact (including cost effectiveness) of two large scale health programs targeted towards reducing maternal mortality. Both the proposed programs were innovative, a first of their kind, in that there have not been similar demand side programs with an explicit focus on maternal healthcare (which in it itself is a new focus area for a large scale innovative program). Also both these programs were large scale programs, implemented by governments over large regions and not pilot experiments. No such large scale programs have been evaluated before though there have been repeated calls in scientific literature for the effectiveness of such programs to be studied.
Also maternal mortality continues to remain a problem for the poor largely because of access barriers to emergency obstetric care. In 2015, it remains one of the MDGs to remain unachieved, though progress has certainly been made. The pending agenda around maternal mortality is expressed in the sustainable development goal to reduce maternal mortality to <70 globally. India still contributes 70% of maternal deaths in south Asia and to 19% of maternal deaths globally.
Demand side financing as a mechanism to allow poor mothers to access institutional EmoC has never been implemented before on a large scale. The proposal will be a first in terms of comparison of outcomes between the two different demand side financing mechanisms employed in large scale programs. Further, both these programs are targeted toward Millennium development goal 5 (progress towards which has been slow and difficult), to reduce maternal mortality.
The private health sector plays an important role in the provision of healthcare in many low income settings, their contribution as part of a large program to achieving a public good (safe delivery in this case) has not been assessed before. Also there is a current push for innovative mechanisms towards the achievement of universal health coverage in low middle income settings, often in partnership with the private sector. This evaluation has thrown light on this aspect
Socioeconomic impact and wide societal implications:
The project was focused on India, a country still contributing to 19% of global maternal deaths. Success with reducing maternal deaths is desperately necessary in India if we are to support the new sustainable development goals. Also the political priority accorded to maternal mortality by recent governments has seen the large scale implementation of innovative solutions over the last decade years which are comprehensively evaluated for the first time.
The project makes an important contribution towards looking at what works from a health system perspective with regard to maternal mortality reduction. Maternal mortality has a strong poverty and gender dimension. Maternal mortality has been described as ‘‘a 21st century problem essentially only for the poor, and one virtually eliminated for people with the resources to access health services”. Even, within poor countries, the poorest women are as much as six times more likely to die during childbirth as richer women. Inequalities in the use of maternal health services mirror those in maternal health outcomes. While there are many dimensions to disadvantage (including race, ethnicity, age, urban/rural status, etc.) and while the causes of inequity vary among and within countries, it is however clear that women from the wealthiest groups consistently make the most use of skilled care during pregnancy, delivery, and the postpartum period and have the best chance to survive childbirth. Indeed, analyses of inequalities in use of health services have shown that rich-poor ratios are often greater when it comes to professional delivery assistance than for other basic health services. Not only are wealth inequalities greater for use of maternity care, but little progress has been achieved in reducing these particular inequalities, particularly in sub-Saharan Africa and South Asia.
The project through the evaluation of these programs for maternal mortality reduction indicates issues in program design and implementation that can influence this goal. Issues from the ground at the level of facilities implementing the program, user and community perspectives have been captured. The findings from this project have important lessons for low middle income governments aiming to start programs aimed at maternal mortality reduction.
From a health systems point of view, the project has studied a number of aspects related to demand side financing programs. Since the project began the literature around demand side financing in the area of maternal health care has widened. However much of the reporting has been from small programs in small areas run with donor support in low income settings. These programs studied under MATIND are different given their scale and that they are completely tax funded by the government. The program empirically demonstrates the prerequisites that need to be in place prior to a successful demand side finding program being implemented and the implications of these not being in place. There has been an urgent need to know what works programmatically in the area of maternal mortality in the context of low income settings. The MATIND project has spoken to this need.
The project also generates knowledge on public private sector partnerships for public health outcomes in the global south. This is particularly relevant given that the direction forward in many low middle income countries to achieve universal health coverage is through partnerships with the private sector. The MATIND project delves deeply into partnership issues and has explored key enabling factors as well as barriers to successful partnerships using these two programs as examples.
Dissemination:
Plans for dissemination, particularly among policy makers, civil society groups and researchers were clearly defined (an entire work package has been dedicated to this) at the outset. As stated, open access scientific publications were published from this project, as far as possible to allow national and international access to reports and evidence coming out of this proposal.
Besides scientific publications, the project included a close collaboration between the consortium and stakeholders including policy makers, civil society groups and researchers in the field. Policy makers and program managers were actively engaged throughout the entire evaluation process. Involvements of stakeholders in the initial methodological consultations were also important. On completion of the evaluation, widespread access to the results were ensured, through the distribution of policy briefs and the conduction of national dissemination meetings with the government of India and the state governments. Getting research into policy has been an important priority of this project. We actively involved local policy makers in the process of the evaluation research, besides making the results readily accessible to them. The process of dissemination and engagement is expected to continue even after the life of project.
A brief overview of dissemination strategies and events is presented below:
All European, Indian and Chinese partners met and interacted at the official MATIND kick off meeting in August 2011 as well as the 2nd annual meeting in December of 2013 and the final consortium meeting in August 2015. Besides the kick-off meeting, there has been a number of cross partner collaboration meetings:
Partners Involved Where When Objective
1,3,6 Ujjain, Madhya Pradesh and Ahmedabad, Gujarat May 22nd to June 4th 2011 To discuss project planning, selection of districts, secondary data variables, phase wise implementation plan, process involved to develop the forms, SOPs, and logistics. Initial exploratory field visits were also performed.
1,3,4,5,6 Ahmedabad, Gujarat August 24th – 27th 2011 The official MATIND kick-off meeting
1,3,4,6 Ujjain, Madhya Pradesh August 31st-September 3rd 2011 To discuss the development of the protocol and qualitative topic guide.
1,5 Hangzhou, China November 2011 To discuss the role of Partner 5 in the project
3,4,6 Ujjain, Madhya Pradesh November 12th-14th 2011 To discuss study instruments for the facility based portion of WP2 and to do field visits
3,6 Ujjain, Madhya Pradesh December 23rd-25th 2011 To discuss the logistical implementation of gathering GIS coordinates for the study districts.
1,3,6 Ujjain, Madhya Pradesh January 20th-25th 2012 To discuss the roll-out plan for the Shajapur pilot, plan for main study, logistics, GIS plan, red cap data entry software
1,4 Liverpool, UK April 3rd 2012 To discuss WP 4 – private providers participation in the programs
1,3 Stockholm, Sweden May 5th-10th 2012 MATIND project discussion; plan development, mainly qualitative studies and data Analysis of pilot data, Phase 2 discussion
1,3,6 Ahmedabad, Gujarat May17th -27th2012 To review the initial results and implementation of Facility survey
1,3 Stockholm, Sweden June 11th, 2012 To discuss the cost effectiveness studies
1,3 Stockholm, Sweden June 13th 2012 To discuss the community based studies
1,4 Liverpool, UK June 22nd 2012 Qualitative portion of the MATIND project
1,3 Stockholm, Sweden July 3rd- 7th2012 Data entry and analysis of data from Shahdol district
3,6 Ahmedbad, Gujarat August 23rd- 25th 2012 GIS related discussions
4,6 Ahmedbad, Gujarat August 2012 To discuss the qualitative work in Gujarat
1,3,4,6 Stockholm, Sweden September 14th 2012 To discuss facility and community based studies
1,3,6 Stockholm, Sweden September 17th- 20th 2012 To discuss community based studies and the cost effectiveness studies
1,3,6 Madhya Pradesh and
Gujarat, India 22nd May – 4th June 2011 To discuss project planning, selection of districts, secondary data variables, phase wise implementation plan, process involved to develop the forms, SOPs, and logistics. Initial exploratory field visits were also performed.
1 & 4 Liverpool, UK 21-¬‐22 June 2012 To discuss the progression of the qualitative work in the
DOW
1, 3 & 6 Stockholm, Sweden
10th Sept – 30th Oct 2012 MATIND Project Meeting to discuss economic studies & Manuscript writing
1, 3 & 6
Ujjain, Madhya Pradesh 30th Nov – 4th Dec 2012
Discussion on WP 2 completion and planning for WP 6
1, 3 & 6
Ujjain, Madhya Pradesh
& Udaipur, Rajathstan
28th Feb – 17th March
2013
Finalization of WP 6 studies, data collection and database construction
1, 3 & 6
Ahmedabad, Gujarat 24th – 26th March 2013
Results sharing from WP 2
1 & 3
Stockholm, Sweden 12th Apr – 1st May 2013 Project meeting, discussion of methodology for community based studies (WP 6/7), manuscript writing
1 & 4
Liverpool, UK 12th-¬‐13th Feb 2013 To discuss data collected on women’s experience and reasons for participating in the JSY program (WP 6 obj v)
1, 3, 4, & 6
Stockholm, Sweden
29 April to 1 May 2013
Discuss analysis of observational study data on quality of care in MP (WP 2 iii) including developing coding framework and charting. Also to discuss the analysis of qualitative interview data on women’s perceptions of quality of care in Gujarat (WP 2 iii). Partner 4 administered
training in qualitative analysis of interview data, framework analysis
1, 3 & 4
Stockholm, Sweden
19-¬‐20 May 2013
Discuss and present studies in WP 2 from MP.
1 & 4 Liverpool, UK 3-¬‐6 June 2013
To discuss coding framework on women’s experience and reasons for participating in the JSY program (WP 6 obj v)
1 & 4 Liverpool, UK 10-¬‐11 July 2013 To discuss descriptive analysis for the study on experience and reasons for participating in the JSY program (WP 6 obj v
3 & 6
Ujjain, Madhya Pradesh
24th – 27th July 2013
Discuss database management for community based studies
(WP 6/7)
1 & 5 Stockholm, Sweden 18-20 September 2013 Progress on RHINCAV/CHIMACA reviews. Analysis of the NCMS data
1, 4 & 6
Stockholm, Sweden
27th September 2013
To discuss technical and logistical issues regarding qualitative data collection, management and analysis, including management of transcription and translation; to discuss concept and process for study of emergency transport (WP 2iii)
4 & 6
Stockholm, Sweden
29th Sept – 2nd October
2013
Discuss the development of CY policymaker interviews (WP
1 ii) including protocol development, training in qualitative analysis of interview data.
3 & 4
Madhya Pradesh, India
24th-¬‐30th November 2013 To discuss further analysis of observational study data (WP
2 iii) including interpretation and developing a paper for publication
All partners Khajuraho, Madhya
Pradesh 2nd-¬‐6th December 2013 2nd Annual Partner Meeting
1 & 3
Ujjain, Madhya Pradesh 4th – 8th March 2014 Project team meeting, publication plan, discussion of blood bank study design (WP2) and qualitative work (WP 4/5)
4 & 6 Gujarat, India 17th–18th March 2014 To discuss further analysis of qualitative interview data on women’s perceptions of quality of care in CY facilities (Gujarat) (WP 2 iii), and policy maker perceptions of CY
3 & 4 Madhya Pradesh, India 19-¬‐21 March 2014 To discuss further development of qualitative study of perceptions and experiences of ASHAs (WP 5 ii), including training RA and reviewing pilot interview data, and to discuss analysis of qualitative study of private providers in JSY 2 (WP 4) and study of policy makers perceptions of JSY (MP) (WP 1 iv).
All partners Delhi, Ahmedabad, Gujarat, Bhopal, Madhya Pradesh 4th – 9th August 2015 Dissemination meeting to the Central and local policy makers/Governments
Project results have been presented at the following conferences:
Conference
Name Date of Conference Abstract
Name Authors Status
13th World Congress on Public Health 23-27 April 2012 Is MMR Really Declining – Challenges in Monitoring Ng, M., Levin-Rector, A., Diwan, V., Sabde Y., De Costa, A. Presented as poster (Abstract #124)
13th World Congress on Public Health 23-27 April 2012 India's JSY Cash Transfer Program for Maternal Health: Who Participates and Who Does Not – a Report From Ujjain District Sidney, K., El-Khatib, Z., Diwan, V., De Costa, A. Presented as poster (Abstract #120)
Second Global Symposium on Health Systems Research (HSR)
Beijing, China 31 Oct –
3 Nov 2012 'Improving access to maternity services: an overview of cash transfer and voucher schemes in South Asia' Jehan, K., Sidney, K., Smith, H. and de Costa, A. To present as poster
Second Global Symposium on Health Systems Research (HSR)
Beijing, China 31 Oct –
3 Nov 2012 Need to strengthen BEmOCs? – A GIS based study on the utilization of maternal health service in Madhya Pradesh, India. Sabde, Y., Randive, B., Chaturvedi, S., Singh, M., Sidney, K., De Costa, A., Diwan, V. To present as poster
Global Maternal Health Conference 2013: Quality of Care 15-¬‐17 Jan
2013 Janani Express: Utilization of an Emergency Maternal Transportation Service in Ujjain District, Madhya Pradesh India Kristi Sidney, Bharat Randive, Ayesha De Costa, Vishal Diwan Poster
Global Maternal Health Conference 2013: Quality of Care 15-¬‐17 Jan
2013 India’s conditional cash transfer programme to promote institutional births: What has been the effect on maternal mortality? Bharat Randive, Vishal Diwan, Ayesha De Costa Oral
Global Maternal Health Conference 2013: Quality of Care 15-¬‐17 Jan
2013 What has been the effect of JSY on maternal mortality? Studies from Madhya Pradesh, India Ng M, Misra A, Diwan V, Levin, Rector A, De Costa A Oral
Global Maternal Health Conference 2013: Quality of Care 15-¬‐17 Jan
2013 Too close to die: A study of effectiveness of referral system for emergency obstetric care using spatial analysis in Madhya Pradesh, India Sarika Chaturvedi, Bharat Randive, Manish Singh, Ayesha De Costa, Vishal Diwan Oral
Global Maternal Health Conference 2013: Quality of Care 15-¬‐17 Jan
2013 Does SBA training improve quality of care?: Studying the link between SBA training and provider practices under the JSY cash transfer programme in Madhya Pradesh, India Sarika Chaturvedi, Bharat Randive, Yogesh Sabde, Kristi Sidney, Ayesha De Costa Poster
Global Maternal Health Conference 2013: Quality of Care 15-¬‐17 Jan
2013 Which mothers travel to reach a facility under the cash transfer program? A network analysis from Madhya Pradesh, India Yogesh Sabde, Sarika Chaturvedi, Zia Ur Rehman, Sayyad Ali, Manish Singh, Kristi Sidney Poster
Global Health Metrics & Evaluations Conference 17-¬‐19 June
2013 Impact of a public private performance based financing partnership on C-¬‐section proportions: Cross-¬‐sectional study from a district in Gujarat, India Kranti Vora, Parvathy Raman, Dileep Mavalankar, Veena Iyer, Kristi Sidney, Rajesh Mehta Poster
Global Health Metrics & Evaluations Conference 17-¬‐19 June
2013 Effective coverage of institutional deliveries under the JSY program in high maternal mortality provinces of India Bharat Randive, Sarika Chaturvedi, Vishal Diwan, Ayesha De Costa Oral
Global Health Metrics & Evaluations Conference 17-¬‐19 June
2013 Initial results on the impact of Chiranjeevi Yojana: a public–private partnership programme for maternal health in Gujarat, India Marie Ng, Parvathy Shanker-¬‐ Raman, Rajesh Mehta, Ayesha DeCosta, Dileep Mavalankar Poster
3rd NETWORK MEETING with regard to EPI-¬‐4 project, Ahmedabad 27th June
2013 Inequity in access to maternal health care:under JSY in Madhya Pradesh. Yogesh Sabde, Vishal Diwan, Ayesha De Costa Oral
5th International Conference on healthGIS 2013 21st- 23rd August
2013 Public private performance based financing program for institutional delivery – Chiranjeevi Yojana: Which private facilities participate? Which eligible women participate? A spatial analysis from Sabarkantha district of Gujarat, India Veena Iyer, Ayesha De Costa, Dileep Mavalankar Oral
5th International Conference on health GIS 2013 21st- 23rd August
2013 An analysis to study access to emergency transport vehicles under a scheme called “Janani Express Yojana (JEY)” in two districts of MP, India using spatial approach. Yogesh D Sabde, Sarika D Chaturvedi, Kristi Sidney, Bharat B Randive, Ayesha De Costa, Vishal Diwan Oral
Joint State Conference of IPHA-¬‐IAPSM, Pune India 25th ‐26th Feb
2014 Better pregnancy outcomes need more than bringing women to hospitals: A call to improve quality of institutional birthsobstetric referral services in India’s JSY cash transfer program for Sarika Chaturvedi, Bharat Randive, Vishal Diwan, Ayesha De Costa Oral
National Seminar on maternal neonatal health and safe abortion: opportunity and challenges held at Hyderabad April 21-27 2014 Availability and distribution of safe abortion services in rural areas: a facility assessment study in Madhya Pradesh, India Sarika Chaturvedi, Bharat Randive, Vishal Diwan, Ayesha De Costa Oral
Global Health System (HSR) Conference, Captown South Africa- 29.09.14 to 05.10.14 Which mothers bypassed nearer facility under the cash transfer program? - A network analysis from Madhya Pradesh, India Yogesh D Sabde, Sarika D Chaturvedi, Bharat B Randive, Ayesha De Costa, Vishal Diwan Poster
Global Health System (HSR) Conference, Captown South Africa- 29.09.14 to 05.10.14 An equity analysis of institutional delivery uptake and maternal mortality reduction in context of cash transfer program (JSY): results from 9 states of India Randive, B., San Sebastian, M., De Costa, A., Lindholm, L Oral
11th World Congress in Health Economics- Milan July 11-16 2015 Out of pocket payments made during stay at facilities by women eligible to receive cash transfers to give birth in facilities under India’s JSY program: findings from Madhya Pradesh Sidney K, Salazar M, Marrone G, Diwan V, DeCosta A, Lindholm L. Oral
11th World Congress in Health Economics- Milan July 11-16 2015 India’s JSY cash transfer program to promote facility births: Increased access to facility based care but not to quality care Raven J, Chaturvedi S , Diwan V, De Costa A Oral
11th World Congress in Health Economics- Milan July 11-16 2015 The Chiranjeevi Yojana, a public-private-partnership for obstetric services in Gujarat, India: Who are the private participating obstetric service providers? Iyer V, Sidney K, Malavankar D, DeCosta A Oral
The following articles have been published in international peer-reviewed journals:
1. Sidney K, Diwan V, El-Khatib Z, De Costa A. India's JSY cash transfer program for maternal health: who participates and who doesn't--a report from Ujjain district. Reprod Health. 2012 Jan 24;9:2.
2. Jehan K, Sidney K, Smith H, de Costa A.Improving access to maternity services: an overview of cash transfer and voucher schemes in South Asia. Reprod Health Matters. 2012 Jun;20(39):142-54.
3. Sidney K, De Costa A, Diwan V, Mavalankar DV, Smith H. An evaluation of two large scale demand side financing programs for maternal health in India: the MATIND study protocol. BMC Public Health. 2012 Aug 27;12(1):699.
4. 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. 2013. 8(6): e67452. doi:10.1371/journal.pone.0067452
5. De Costa A, Vora K, Ryan K, , Raman P, Santacatterina M, Mavalankar D. The state-¬‐led large scale public private partnership 'Chiranjeevi program' to increase access to institutional delivery among poor women in Gujarat, India: How has it done? What can we learn? PLOS ONE. 2014. 9(5): e95704. doi:10.1371/journal.pone.0095704
6. Ng M, Misra A, Diwan V, Agnani M, Levin-Rector5 A, De Costa A. An assessment of the impact of the JSY cash transfer program on maternal mortality reduction in Madhya Pradesh, India. Glob Health Action 2014, 7: 24939 – (http://dx.doi.org/10.3402/gha.v7.24939(öffnet in neuem Fenster))
7. 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. Social Science and Medicine, 2014, 123: 1-6 http://dx.doi.org/10.1016/j.socscimed.2014.10.042(öffnet in neuem Fenster)
8. Sabde Y, De Costa A and Diwan V. A spatial analysis to study access to emergency obstetric transport services under the public private “Janani Express Yojana” program in two districts of Madhya Pradesh, India. Reproductive Health 2014, 11:57. http://www.reproductive-health-journal.com/content/11/1/57(öffnet in neuem Fenster)
9. Chaturvedi S, Randive B, Diwan V, De Costa A (2014) Quality of Obstetric Referral Services in India’s JSY Cash Transfer Programme for Institutional Births: A Study from Madhya Pradesh Province. PLoS ONE 9(5): e96773. doi:10.1371/journal.pone.0096773
10. Sidney K, Ryan K, Diwan V, De Costa A. Utilization of a State Run Public Private Emergency Transportation Service Exclusively for Childbirth: The Janani (Maternal) Express Program in Madhya Pradesh, India. PLoS ONE 2014, 9(5): e96287. doi:10.1371/journal.pone.0096287
11. Chaturvedi et al. Competence of birth attendants at providing emergency obstetric care under India’s JSY conditional cash transfer program for institutional delivery: an assessment using case vignettes in Madhya Pradesh province. BMC Pregnancy and Childbirth. 2014. 14:174.
12. Lennart Bogg, Vishal Diwan, Kranti S Vora and Ayesha De Costa. Impact of alternative maternal demand-side financial support programs in India on the caesarean section rates – indications of supplier-included demand. Maternal and Child Health Journal. 2015. DOI 10.1007/s10995-015-1810-2
13. Chaturvedi S, Upadhyay S, De Costa A, et al. Implementation of the partograph in India’s JSY cash transfer programme for facility births: a mixed methods study in Madhya Pradesh province. BMJ Open 2015;5:e006211. doi:10.1136/bmjopen-2014- 006211
14. Ganguly PS, Jehan K, De Costa A et al. Considerations of private sector obstetricians on participation in the state led “Chiranjeevi Yojana” scheme to promote institutional delivery in Gujarat, India: a qualitative study. BMC Pregnancy and Childbirth. 2014. 14:352. (http://www.ncbi.nlm.nih.gov/pubmed/24787692(öffnet in neuem Fenster))
15. Vora K, Yasobant S, Mavalankar D. Predictors of Availing Maternal Health Schemes: A community based study in Gujarat, India. (2014) Indian Journal of Community Health. 2014:26 (2); 174-180.
16. Chaturvedi S, Ali S, Randive B, Sabde Y, Diwan V, De Costa A. Availability and distribution of safe abortion services in rural areas: A facility assessment study in Madhya Pradesh, India. Global Health Action. 2015, 8: 26346 - http://dx.doi.org/10.3402/gha.v8.26346(öffnet in neuem Fenster)
17. Yasobant, S., Vora, K.S. Hughes, C., Upadhyay, A. and Mavalankar, D.V. (2015) Geovisualization: A Newer GIS Technology for Implementation Research in Health. Journal of Geographic Information System, 7, 20-28. (http://dx.doi.org/10.4236/jgis.2015.71002 )
18. Chaturvedi S, De Costa A, Raven J: Does the Janani Suraksha Yojana cash transfer program to promote facility births in India ensure skilled birth attendance? : A qualitative study of intrapartum care in Madhya Pradesh. Global Health Action. 2015 Jul 7;8:27427. doi: 10.3402/gha.v8.27427
19. Vora, K.S. Yasobant, S., Sengupta, R., De Costa, A., Upadhyay, A. and Mavalankar, D.V. Options for optimal coverage of free C-section services for poor mothers in Indian state of Gujarat: Location allocation analysis using GIS. PLoS ONE. 2015.
20. Chaturvedi S, Randive B, Raven J, Diwan V, DeCosta A. Assessment of the quality of clinical documentation in India’s JSY cash transfer program for facility births in Madhya Pradesh. IJGO. 2015. DOI: http://dx.doi.org/10.1016/j.ijgo.2015.07.016(öffnet in neuem Fenster)
Under review:
1. Vora, K.S. Yasobant, S., Patel, A., Upadhyay, A. and Mavalankar, D.V. Has Chiranjeevi Yojana changed the geographic availability of free Comprehensive Emergency Obstetric Care Services in Gujarat, India? BMC Health Services. 2015.
2. Sidney K, Salazar M, Marrone G, Diwan V, DeCosta A, Lindholm L. Out-of-pocket expenditures for childbirth in the context of the Janani Suraksha Yojana (JSY) cash transfer program to promote facility births: Who pays and how much? Studies from Madhya Pradesh, India. Health Policy Planning. Manuscript.
3. Sidney K, Tolhurst R, Jehan K, Diwan V, DeCosta A: The cash transfer is important but all women anyway go to hospital for childbirth nowadays’ - a qualitative exploration of why women participate in a cash transfer program to promote institutional deliveries in Madhya Pradesh, India. BMC Pregnancy and Childbirth. Manuscript.
4. Sidney K, Iyer V, Vora K, Mavalankar D, De Costa A. State-wide program to promote institutional delivery Gujarat, India: Who participates and the degree of financial subsidy provided by the Chiranjeevi Yojana Program. Journal of Health, Population and Nutrition. Manuscript.
5. Altman RS, Sidney K, De Costa A, Vora K, Salazar M. Is institutional childbirth associated with lower neonatal mortality risk in Gujarat India: A cohort study in a vulnerable population in the context of the Chiranjeevi Yojana (CY) program. Manuscript.
6. Iyer V, Sidney K, Mehta R, Mavalankar D, De Costa A. Emergency Obstetric Care availability and provision under a public-private partnership in three district of Gujarat, India- the Chiranjeevi Yojana, who are the private partners? Lessons for Universal Health Coverage: cross-sectional facility survey. Manuscript.
7. Sabde Y, Diwan V, Randive B, Chaturvedi S, Sidney K, Salazar M, DeCosta A. The availability of Emergency Obstetric Care under the JSY cash transfer program in Madhya Pradesh, India. BMC Pregnancy and Childbirth. Manuscript.
8. Sabde Y, Chaturvedi S, Randive B, Sidney K, DeCosta A, Salazar M, Diwan V. Bypassing health care facilities for childbirth in Madhya Pradesh (MP), India: A multilevel study. Manuscript.
List of Websites:
www.matind.eu
Ayesha Mariette De Costa MD PhD
Assoc Professor – Global health
Dept of Public Health Sciences
Karolinska Insitutet
Stockholm S 171 77
0046852483336
ayesha.de.costa@ki.se
This project comprises an evaluation of the performance of the demand side financing programs to reduce maternal mortality. The evaluation is comprehensive and takes into account a number of domains – access, utilization, human resources, quality of care, participation of the private sector, user perceptions, policy dimensions as well an economic analysis.
High maternal mortality in India is a serious public health challenge. Demand side financing interventions have emerged as a strategy to promote access to emergency obstetric care. Two such state run programs, Janani Suraksha Yojana (JSY) and Chiranjeevi Yojana (CY), were designed and implemented to reduce financial access barriers that preclude women from obtaining emergency obstetric care. The MATIND project operates in two Indian provinces, Madhya Pradesh and Gujarat, where these programs are operationalized.
JSY, a conditional cash transfer, awards money directly to a woman who delivers in a public health facility. This is studied in the Madhya Pradesh province. CY, a voucher based program, empanels private obstetricians in the Gujarat province, who are reimbursed by the government to perform deliveries of socioeconomically disadvantaged women. The programs have been in operation for the last ten years.
This project is a comprehensive evaluation for two innovative demand side financing programs that aims to study: (i) trends in program uptake since inception, (ii) the influence of the program on institutional delivery rates, maternal and neonatal outcomes, (iii) explore service providers’ experiences, (iv) emergency transportation, (vi) quality of care administered, (vii) experiences of the beneficiaries and non-beneficiaries, (viii) the role of community health workers and (ix) the cost effectiveness of the programs.
Primary data was collected in each district at three levels: 1. state health departments (policy level) 2. obstetric care facilities, and 3.recently delivered mothers in the community. A combination of qualitative and quantitative methods was used including: 1. facility and community-based surveys, 2. structured observations, 3. record reviews, 4. case vignettes, and 5. in-depth interviews.
In summary, the JSY has resulted in steep increases in the proportion of facility based deliveries in the study area but also in the country. However our analysis was unable to detect any influence of program utilization with reduction in maternal mortality. 90% of institutional deliveries in the study districts occurred under the program. However the quality of care under the program was less than optimal. There were serious deficiencies in the skills of front line birth attendants, and the ability to provide comprehensive emergency obstetric care services like cesarean and blood transfusion was extremely limited. The program ran a sub arm in partnership with the private sector, however as the state paid private obstetricians differentially for vaginal and cesarean births, it resulted in disproportionate cesareans and the state had to close the partnership. Women under the program still experienced out of pocket costs, but these were not related to technical care, rather to informal payments paid at facilities.
The CY program was more targeted to vulnerable women, however though it has had upto a million births occurring under the program, the proportion of eligble women who become program beneficiaries is low (15%). This is largely because of a lack of knowledge, an inability to get the documentation prepared and because the partnership has seen participation of private obstetricians change with time. There was a better quality of care provided under the CY program. Birth attendants worked under the close supervision of private obstetricians who were very aware of their reputations, cesareans were easily available. However although the program was intended to be cashless at the point of care for the beneficiary, this was not always the case, as private obstetricians were uncertain if they would be reimbursed by the state as the paper work was burdensome.
The role of Accredited Social health activist was explored in the programs. The cost effectiveness of each program was studied.
The overall evaluation has important lesson for the design and implementation of large scale demand side financing programs, particularly in the design and management of state led public private partnerships to achieve universal health coverage.
Project Context and Objectives:
MATIND studied two state run programs, Janani Suraksha Yojana (JSY) and Chiranjeevi Yojana (CY), designed to reduce financial access barriers that preclude women from obtaining emergency obstetric care in India. MATIND assessed the following: (1) trends in programs uptake since programs inception, (2) the influence of the programs on institutional delivery rates, maternal and neonatal outcomes, (3) service providers’ experiences, (4) emergency transportation systems for obstetric referrals, (5) characteristics of the facilities participating in the programs, (6) quality of care administered, (7) experiences of the beneficiaries and non-beneficiaries, (8) the role of community health workers, and (9) the degree of subsidy to users because of the programs.
Why was MATIND conducted?
Maternal mortality is a 21st century problem essentially only for the poor, and one virtually eliminated for people with the means and status to access health care. Such a marker of global inequity is shocking and is an indication of wider development issues targeted in some of the other MDGs, especially poverty, education, and gender. A link between poverty and maternal health has been clear for more than a century, and is supported by extensive evidence from literature. Even though tax based public healthcare provision has been known to be one of the best alternatives to promote equity in access, it has been difficult to ensure that funding is distributed to the most needy, particularly the poor living in non-urban areas. This difficulty is even more acute in low income countries, where insurance and capital markets are less well developed.
Demand side financing directs purchasing power to the target group, to enable them to purchase specific services[42] and can take a number of different forms (including vouchers, conditional cash transfers). Interventions to reduce demand side barriers are justified to rectify market failures, and in the pursuance of equity. Such programs on a large scale have first been used to pursue reduction in maternal mortality in the Indian context.
Maternal mortality in India is a serious public health challenge. Demand-side financing interventions have emerged as a strategy to promote women’s access to emergency obstetric care (EmOC) in an attempt to reduce maternal deaths. Two state-run programs, JSY and CY, have been implemented to increase women´s access to EmOC: (1) JSY awards money directly to a woman who delivers in a public health facility, and (2) CY empanels private obstetricians in Gujarat province, who are reimbursed (a fixed package for 100 deliveries including C-section and complications) by the government to perform deliveries for poor and tribal women. In spite of JSY and CY magnitude and the amount of resources allocated, no comprehensive evaluations of these programs have been conducted.
Where was MATIND conducted?
The study took place in two provinces, Madhya Pradesh and Gujarat (Figure 1) where JSY and CY were studied, respectively. In each province we purposefully selected three districts with different socioeconomic, demographic, and health profiles. Madhya Pradesh, located in the central part of India, has a population of 72 million. The state is divided into 51 administrative districts, each with a population of 1–2 million. Three districts, Shahdol, Panna and Ujjain, were selected for the study. Gujarat has a population of 60 million and 26 administrative districts. Each district has on the average a population of more than two million. Three heterogeneous districts from the western, central and eastern belts of the state, Sabarkantha, Surendranagar and Dahod, were purposefully selected for the study. These districts had different sub-population compositions i.e. varying proportions of the eligibility population (scheduled castes and tribes and women living below the poverty line) for the CY program.
What study designs did MATIND use?
Primary data was collected in each district at three levels: 1. state health departments, 2. obstetric care facilities, and 3.recently delivered mothers in the community. A combination of qualitative and quantitative methods was used including: 1. facility and community-based surveys, 2. structured observations, 3. record reviews, 4. case vignettes, and 5. in-depth interviews.
The levels of this evaluation:
(i) Firstly, at the provincial or policy/administration level: These include document reviews or program documents, key person interviews, a study of budget outlays and expenditures, fund disbursement mechanisms, administration of the respective program and accreditation procedures for private providers. These were done at the level of the Departments of Health in both provincial capitals in the study provinces and pertain to the program as in its totality across the study province.
(ii) Secondly at the level of the districts: Each province is divided into administrative units called districts (population of a district 1-1.5 million on average).Each district has its own district health administration to oversee the functioning of the health system within its boundaries.
(iii) Finally at the level of the village, eligible beneficiaries will be identified (identified by possession of a BPL card). Users will be compared with a comparison group of eligible non users using matching. Users and non-users within and between each program were compared with regard to background characteristics, outcomes and other indicators
Overall Aim: The overall aim of this proposal is to develop methodology and apply this to assess the impact of and compare two large scale programs for the financing and provision of maternal healthcare in India, in terms of a) health outcomes (maternal and perinatal), b) the influence of the program and health service and social environment structure on outcomes, c) extended benefits of the program, d) private health sector contribution to outcomes and e) financial sustainability. The two programs will also be indirectly compared to rural health insurance in China.
Specific Objectives:
At the provincial (policy) level
i. To study and compare trends reported by each program over time since inception, in terms of: uptake; institutional EmOC participation; private sector participation; and recorded maternal and neonatal outcomes (including institutional delivery, caesarean section rates, complicated delivery and mortality within the programs).
ii. To explore policymaker views on the implementation of programs at the policy level (ie roles of the ANM (Auxiliary Nurse Midwife) and ASHA (Accredited Social Health Activist), human resources for maternal healthcare, NRHM as a platform for maternal health care delivery, private participation in the JSY program)
iii. To study the two emergency obstetric transportation models utilized in each of the programs with regards to ease of access, functionality, utilization and cost of the service. Further, to explore among a group of mothers their experience of gaining access to and using the transportation.
At the level of the public and private EmoC facilities in selected study districts:
iv. To assess and compare the coverage of each program in terms of provision of EmOC and utilization by target mothers and equity in geographic access.
v. To describe and compare the institutions (public and private separately) providing EmOC under the respective programs and the quality of care offered by performing a detailed facility survey, record reviews and observations in the participating institutions.
vi. To assess the influence of each program on the number of institutional deliveries taking place in these facilities among the target group through a time series analysis of the proportion of program births out of all births over a three year period.
vii. To study and compare delivery outcomes under the programs separately for private and public facilities in terms of: Caesarean section proportion, complicated delivery proportion, average lengths of hospital stay, maternal and perinatal mortality between program mothers and non participating women otherwise delivering at the EmOC facilities (public and private sector).
viii. To explore and contrast in the study districts, the overall experiences of EmOC providers participating in the respective programs, including particular motives for participation, human resource implications,, public/private interaction and willingness (and prerequisites) for continued long term involvement.
ix. To study the eligible non participating private sector EmOC providers (in CY) in the study districts, compare these with participating private institutions and explore reasons for non participation.
At the level of mothers in the community (village level):
x. To estimate maternal complication rates in the community with a population based survey.
xi. To compare eligible non participant mothers with participant mothers under the respective programs in terms of: their characteristics, their receipt of maternal health care, the direct and indirect costs of this care, their reasons for participation/non participation and their reasons for their choice of facility.
xii. To study and compare the receipt of comprehensive maternal healthcare by women participating in each program in terms of: the rate of registration of pregnancy; receipt and source of full antenatal care; place of delivery; type of delivery; duration of hospitalization (if applicable); transportation for delivery; post natal care and receipt of incentives (JSY).
xiii. To study and compare the broader influence of each program on wider health outcomes (measured by infant survival, and infant weight).
Overall:
xiv. To study financial sustainability for each of the programs by comparing costs of the two financing systems (though a cost utility analysis and a multivariate analysis, including sensitivity analysis) and by analyzing cost trends both from financing agency perspective and from the perspective of mothers.
xv. To compare the findings on the two programs in India with findings from studies on the impact of NCMS on maternal care in China.
xvi. To disseminate the results of this assessment widely among stakeholders through workshops, academic scientific publications and appropriate national and international forums.
Project Results:
Introduction:
The MATIND project comprised the evaluation and comparison of the impact (including cost effectiveness) of two large scale health programs targeted towards reducing maternal mortality in India. Both the proposed programs were innovative, a first of their kind, in that there have not been similar demand side programs with an explicit focus on maternal healthcare (which in it itself is a new focus area for a large scale innovative program). Also both these programs were large scale programs, implemented by governments over large regions and not pilot experiments. No such large scale programs have been evaluated before though there have been repeated calls in scientific literature for the effectiveness of such programs to be studied.
Also maternal mortality continues to remain a problem for the poor largely because of access barriers to emergency obstetric care. In 2015, it remains one of the MDGs to remain unachieved, though progress has certainly been made. The pending agenda around maternal mortality is expressed in the sustainable development goal to reduce maternal mortality to <70 globally. India still contributes 70% of maternal deaths in south Asia and to 19% of maternal deaths globally.
Demand side financing as a mechanism to allow poor mothers to access institutional EmoC has never been implemented before on a large scale. The proposal has been a first in terms of assessing the two different demand side financing mechanisms employed in large scale programs. Further, both these programs are targeted toward Millennium development goal 5 (progress towards which has not been on track), to reduce maternal mortality.
The private health sector plays an important role in the provision of healthcare in many low income settings, their contribution as part of a large program to achieving a public good (safe delivery in this case) has not been assessed before. Also there is a current push for innovative mechanisms towards the achievement of universal health coverage in low middle income settings, often in partnership with the private sector. This evaluation has thrown light on this aspect.
Summary of the programs studied and objectives of the project:
MATIND studied two state run programs, Janani Suraksha Yojana (JSY) and Chiranjeevi Yojana (CY), designed to reduce financial access barriers that preclude women from obtaining emergency obstetric care in India. MATIND assessed the following: (1) trends in programs uptake since programs inception, (2) the influence of the programs on institutional delivery rates, maternal and neonatal outcomes, (3) service providers’ experiences, (4) emergency transportation systems for obstetric referrals, (5) characteristics of the facilities participating in the programs, (6) quality of care administered, (7) experiences of the beneficiaries and non-beneficiaries, (8) the role of community health workers, and (9) the degree of subsidy to users because of the programs.
The main S&T results are presented at the three levels at which the programs were studied:
(i) The policy/ state level: policy makers, program desginers and other stakeholder influencing maternal health at a policy level were included. Secondary data on the program from national and subnational surveys was also used.
(ii) The facility level: this includes the results of studies that were conducted at the level of facilities that participated in the respective programs
(iii) The community level: user and community level studies on the two programs
The results are presented for the JSY program first and then for the CY program.
The JSY program
Policy level
Has there been a relationship between JSY program utilization and reduction in maternal mortality?
India has had one of the fastest declines in Maternal Mortality Ratio (MMR); however the current number of maternal deaths is still unacceptably high. There has been limited success with supply–side interventions under the Reproductive and Child Health program (RCH-1 1999–2004) in raising the proportions of skilled attendance at births. Within this context, the Indian government launched a nationwide cash transfer program, Janani Suraksha Yojana (JSY) in 2005. The JSY aims to reduce maternal and neonatal mortality through the promotion of institutional births by providing cash incentives to women who give birth in a health institution. But has there been an association between JSY uptake and MMR reduction?
We looked into this question through two separate analyses conducted in 1) Madhya Pradesh and 2) nine low performing states of India, somewhat similar to Madhya Pradesh. The studies describe the uptake of the JSY, the rise in institutional delivery between 2005 and 2010 and estimated the change in district MMR over the same period. Further, for Madhya Pradesh, we estimated the impact of JSY on the reduction in maternal mortality by examining the association between JSY inputs and the MMR estimates. We reference data from a wide range of sources and synthesize these using modern statistical modeling techniques. We also describe trends in uptake of institutional births after the initiation of the JSY and studied the association between institutional birth proportions and the MMR in the nine states
In Madhya Pradesh the proportion of all institutional deliveries increased from 22% in 2005 to 56% in 2010 province wide. The proportion of JSY-supported institutional deliveries rose from 14% (2005) to 80% (2010). MMR declines in the districts varied from 2 to 35% over this period. For the state as a whole, MMR declined by 12% between 2005 and 2010. Despite the marked increase in JSY-supported delivery, we did not detect a significant association between JSY-supported delivery proportions and changes in MMR in the districts. Changes in maternal mortality and institutional deliveries (both JSY and non-JSY) in the districts of Madhya Pradesh between 2005 and 2010 Analysis of nine low performing states showed that proportion institutional births increased from a pre-program average of 20% to 49% in 5 years. The proportion of institutional births had a small negative correlation with the district MMR. The multivariable regression model did not establish significant association between institutional birth proportions and MMR [CI:-0.10 0.68].
Our study confirms previous report that the proportion of institutional delivery has climbed steeply in Madhya Pradesh and other low performing states in India. However, we were unable to detect a significant association between the rising proportion of institutional births and reduction of MMR both in Madhya Pradesh and other low-performing states.
Qualitative Enquiry: To explore the perceptions of JSY program managers, implementers and non-governmental stakeholders on the JSY program specifically on it’s appropriateness in the given context, effect on maternal mortality, implementation challenges and sustainability of the program and it’s achievements.
We conducted semi structured interviews with public health experts working on maternal health issues and health officials in the state of Madhya Pradesh. The experts were selected purposively and included those well known for their contribution to the maternal health field in India. They work in diverse capacities such as academic researchers, members of national committees for health program design and evaluation representing civil society organizations and UN agencies. The health officials included those working at key positions in the state health department and those working at sub-district level who are presently responsible for implementation of the JSY. Health officials were from Madhya Pradesh, a less developed central Indian state with poorer health indicators than national levels. Thus a total of 11 interviews were conducted with maximum variation sampling keeping relationship to the JSY program as the key sampling principal.
Explorations from our interviews are presented under the following themes:
Analysis of central problem – low institutional delivery: is this a problem with people or with system? : The JSY program aimed to address the problem of low utilization in facility care for childbirth. There were contrasting views among the respondents regarding the reason for low utilization. Most health officials perceived the problem as reluctance of women for facility delivery and hence found JSY an appropriate intervention to attract women to facilities, while non-governmental experts viewed it as oversimplification of complex maternal health issues. Some experts challenged the underlying assumptions of the program; one argued that ‘blaming people’ for low utilization is ‘not an honest analysis of the problem’, since there is a trust deficit between the health system and people which is the reason for low uptake.
Underlying models of development and health improvement: This theme presents the range of views about the notion of demand generation. Two models of health improvement were implicitly or explicitly drawn on by participants in interpreting the concept of demand generation for facility birth. Some participants viewed non-utilization of health facility for childbirth is a problem of individual or family behaviour, and hence approved the need of a stimulus (such as a financial incentive) to change behaviour towards institutional births- aligning with the behaviourist model of development. In contrast, some participants viewed health system and social structures as the problem and emphasized the importance of an enabling environment where people are empowered to take rational decisions and health systems are in place to meet the demand; aligning with an empowerment model of development. Another expert suggested that demand generation through monetary incentives could increase the risk of peoples’ dependency on the incentive to access the services.
With regard to the sustainability of funding from government to this large scale program, some respondents saw it as inherently unsustainable in the long term.
Quality of care: One official and several experts attributed the problem to poor quality of care and agreed on the need to improve quality of care for reduction in MMR. Several experts expressed the view that there had been a decline in quality of care offered at health facilities due to the increased load of institutional deliveries, driven by the JSY, on the weak health system. They pointed out that there were limited efforts to improve the supply side capacity and felt that MMR will decrease faster only if quality of care improves. Both official and experts highlighted systemic problems in service delivery during the JSY program, such as bed shortages, inadequate supplies and human resource issues including under-staffing, lack of competencies among staff, and poor attitudes and behavior of staff.
Respondents had opposing views on the rationale of a cash transfer to promote institutional delivery. However there was consensus on the problem of a weak supply side that could undermine the goal of maternal mortality reduction by the program.
Has the program resulted in reduced inequalities in institutional delivery and maternal mortality reduction across income groups?
The proportion of women giving birth in health institutions has increased sharply in India since the introduction of cash incentive program, Janani Suraksha Yojana (JSY) in 2005. JSY was intended to benefit disadvantaged population who had poor access to institutional care for childbirth who bore the brunt of maternal deaths. Hence, it is important to look at the increase in institutional deliveries following the implementation of JSY from an equity perspective.
We analyzed data on utilization of institutional delivery and maternal mortality ratios from nine Indian states, including Madhya Pradesh. We used area-based socioeconomic measures (ABSM) for analysis of inequalities. The concentration curve, concentration index (CI), Slope Index and Relative Index of Inequality were used to analyze the data.
Among the poorest districts, institutional deliveries increased from an average of 16% to 45%, while among those in the richest districts, the increase was from an average of 40% to 69%. Although inequality in access to institutional delivery care persists, the degree of inequality during the JSY period (2010) was lower than in the period before JSY (2004-06).
Differences in male literacy contributed 30% to overall inequality in institutional delivery. Emergency Obstetric care availability (% of C-sections in public facilities in each district was used as a proxy measure) contributed 20%. The proportion of poorest households in district contributed 18%.
The richest division experienced 97 fewer maternal deaths per hundred thousand live births than did the poorest division during 2007-09, and 135 fewer maternal deaths during 2010. This reveals increased inequality as a consequence of a lower decline in MMR in poorer divisions than in richer ones. Reduction in the MMR during JSY (2007-2009 to 2010) was estimated as being four times higher in richest division quintile than in the poorest one.
Our analysis confirms that although inequality in access to institutional delivery persists, it has been reduced since the JSY program began. The presence of higher maternal mortality with slower pace of decline in the poorest area as well as inequalities in the availability of EmOC facilities during the cash incentive program suggest that the cash incentive alone is not sufficient to achieve equity in maternal health outcomes. Rather, the cash incentive program needs to be supported by the provision of quality health care services including EmOC and improved targeting of disadvantaged populations for the cash incentive program could be considered.
How did public private partnerships under the JSY work?:
The department of health in Madhya Pradesh involved the private sector in JSY program in 2006. The partnership was called the Janani Sahyogi Yojana (JShY or maternal support scheme). Under the JShY partnership, private accredited obstetricians were paid by the state to perform the deliveries of women below poverty line. Between 2007 and 2011, 35000 births occurred in the private sector under the JShY.
We explored the motives of private obstetricians in Madhya Pradesh for joining the JShY PPP, their experiences within the PPP, the benefits and disadvantages of being part of the PPP; their interactions with the state as part of the PPP and motives for withdrawal among those who withdrew from the scheme. We also interviewed officials overseeing the implementation of the JShY at the state department of health to present their views on the JShY and the difficulties encountered. This study throws light on the dynamics of a PPP for obstetric care from the perspective of private sector obstetricians and state managers of the PPP. The findings of the study will inform the design, feasibility, prerequisites and potential pitfalls for such a partnership for obstetric care in this or other similar low income settings.
Private obstetricians who had participated or currently participate in the program were identified based on a listing provided by the department of health in Madhya Pradesh. In each district, private obstetric facilities were purposively selected to represent for profit and charitable hospitals, hospital of different sizes and facilities that were purely obstetric and those that were part of larger multispecialty health service providers.
Participants in the JShY public private partnership program believed the partnership contributed to raising the opportunities for vulnerable women to give birth safely in a setting where they would have access to EmOC. They perceived that this would contribute to a reduction in maternal mortality. Participation was often because of altruism or because private obstetricians saw it as a way to build practices or have more experience. However at the same time there were a number of deterrents to participation, some of which were severe enough for some participants to withdraw from the partnership. One of these was the clustering of high risk cases, who the private obstetricians had never seen before, who often came with severe complications. The other issue raised was that linkages to access blood required had not been made. This was particularly important given the need for blood among potential program beneficiaries who were often are anemic or have complications like post-partum hemorrhage. Besides this clustering of cases, private sector participants reported negative experiences with the paper work and obtaining reimbursements from the government. The extremely high rate of cesarean sections has been of concern to the government who believed that the structure of the reimbursement (with higher rates of reimbursement for cesareans) drove the extraordinarily high rates. However the private obstetricians countered this saying that because of their facilities being specialist ones with full time obstetricians and operating facilities, they tended to have more referred complicated cases which drove up rates. They were not supportive of the governments new reimbursement structure which had an embedded disincentive for cesareans (paid a fixed rate for births 100 women regardless of type of delivery). The new structure caused some of the private sector partners to withdraw. Other concerns include misuse of the scheme by people who did not really deserve but had the necessary paper work. On the whole the private partnership under the JSY was rather small and now stands suspended because of these differences.
Quantitatively, we looked at births in the JShY program from secondary data. In JSaY in MP (program with differential payments for caesarean and vaginal delivery) the CS rate increased over four years, from 26.6% to 40.7% in the participating private hospitals. This trend is all the more surprising, given that the background rate for caesarean deliveries in Madhya Pradesh, where the JSaY was implemented, was only 4.9%. This resulted in the government withdrawing the program and then reintroducing it in as a bundled payment similar to the CY. However private provider participation in the program has waned since.
Facility level
The level of Emergency obstetric care under the JSY program
India has implemented a successful national cash transfer program, the Janani Suraksha Yojana (JSY), to incentivize women to deliver in facilities. It has resulted in a steep rise in facility delivery across the country. As increased utilization of facilities for childbirth is the focus of the JSY, the functionality of the facilities is critical to improved outcomes. We studied the availability and level of provision of emergency obstetric care (EmOC) signal functions in facilities implementing the JSY program in three districts of Madhya Pradesh (MP) state, central India. These are measured against World health report benchmarks. We also study the contribution of private sector facilities to the provision of EmOC in these districts.
A cross-sectional survey of all healthcare facilities that had done at least one delivery a month was conducted between February 2012 and April 2013. EmOC signal functions performed in each facility were recorded, as were human resource data and delivery numbers in each facility
A total of 152 facilities were surveyed (118 program facilities). 86% of childbirths occurred at program facilities, two thirds of which occurred at less-than-BEmOC facilities. All these less-than-BEmOCs facilities performed less than 4 basic signal functions. None of the less-than-CEmOC facilities performed all BEmOC functions. Program facilities provided few EmOC signal functions besides parenteral antibiotic or oxytocic administration. CEmOC provision was predominantly non-program (private): only one of six CEmOC facilities was in the program (public sector). Only 13% of all qualified obstetricians practiced at program facilities. Lack of supplies and insufficient training were the two most common reasons cited for not performing signal functions at program facilities.
Given the high uptake, the JSY program has an opportunity to contribute to reduction in maternal and perinatal mortality in this setting. However, for more program impact on outcomes there needs to be a significant improvement in the provision for EmOC. Among other improvements, a serious focus on training of skilled birth attendants to build competency to enable EmOC provision is necessary.
Quality of care in the institutions in the JSY program:
The JSY program has been successful in raising the proportion of facility births from 30% in 2007 to 73% in 2012. However, decline in maternal mortality ratio has followed a secular trend over the past decade. Poor quality of care (QoC) is a potential reason for this paradoxical situation. Despite the success of the JSY in raising utilization, little is known about the actual QoC provided under the program despite initiatives to improve this. We present here results from four empirical sub studies on aspects of quality of obstetric care in the JSY program in Madhya Pradesh (MP) province conducted during 2012-13 as part of the MATIND project.
Quantitative and qualitative studies were conducted to assess QoC at JSY facilities in three MP districts. Data collection included cross-sectional surveys, participant observation, retrospective case record review, and spatial analysis of transfer time between referring facility and hospitals of those other who died during study period.
Emergency Obstetric Care (EmOC) signal functions at JSY facilities: 86% of institutional births during study period occurred at JSY facilities. Out of those, 66% were in facilities that provided less than the basic EmOC. JSY facilities provided few emergency obstetric care signal functions (EmOC) besides parenteral antibiotic or oxytocic administration. Staff competence under the JSY program
Intrapartum care & staff competences need improvement: Observations revealed that care given during the labor and delivery under the JSY program did not meet standards for skilled birth attendance (SBA): practices were poor, care was not always respectful, and the work environment was not conducive for SBA.A study showed that frontline delivery room nurses (n=233) had poor competence in providing first line management for two common obstetric complications: hemorrhage and eclampsia. In general, the staff obtained low scores (median score= 5/20) on case scenarios to study provision of first line EmOC for common complications. Only 14%, 58%, & 20% were competent at initial assessment, diagnosis, and treatment respectively. The proportion of competent respondents for the initial management of eclampsia (Ecmp), antepartum hemorrhage (APH), and postpartum hemorrhage (PPH) was low. Deficiencies in training of nurse midwives, clinical supervision, and poor practice environment were possible reasons for poor staff performance. In addition, shortages of staff, increased work load since the JSY, and deficiencies in the infrastructure/supplies were reported.
Poor referral patterns and incomplete records: A study of obstetric case referral patterns showed that secondary obstetric care facilities made most referrals and 40% were for conditions expected to be treated at this level. High number of maternal deaths despite spatial access to EmOC i.e. referral from facilities within the desired two hour transfer time, indicate poor quality of referral services. Records reviewed were largely incomplete; indicating poor documentation of care - the proportion of records documenting key clinical parameters was low e.g. 14% documented postnatal blood pressure, 35% documented fetal heart rate.
QoC under the JSY program needs focused attention. This current QoC has restricted the translation of coverage gains from JSY into improved health outcomes. Improvement in the quality of nursing and midwifery education is a crucial step to create a technically competent workforce rather than merely qualified personnel. Adoption of training methods that build the confidence and competence to provide life-saving care is important for the effective functioning of nurse midwives in peripheral facilities. Encourage processes that improve QoC like maintenance of case records. Supportive supervision during training, as well as on the job, could be an important step to bridge the current gaps in competence.
Staff competencies under the JSY program
Access to emergency obstetric care by competent staff can reduce maternal mortality. India has launched the Janani Suraksha Yojana (JSY) conditional cash transfer program to promote institutional births. During implementation of the JSY, India witnessed a steep increase in the proportion of institutional deliveries-from 40% in 2004 to 73% in 2012. However, maternal mortality reduction follows a secular trend. Competent management of complications, when women deliver in facilities under the JSY, is essential for reduction in maternal mortality and therefore to a successful program outcome. We investigate, using clinical vignettes, whether birth attendants at institutions under the program are competent at providing appropriate care for obstetric complications.
A facility based cross-sectional study was conducted in three districts of Madhya Pradesh (MP) province. Written case vignettes for two obstetric complications, hemorrhage and eclampsia, were administered to 233 birth attendant nurses at 73 JSY facilities. Their competence at (a) initial assessment, (b) diagnosis, and (c) making decisions on appropriate first-line care for these complications was scored.
The mean emergency obstetric care (EmOC) competence score was 5.4 (median = 5) on a total score of 20, and 75% of participants scored below 35% of the maximum score. The overall score, although poor, was marginally higher in respondents with Skilled Birth Attendant (SBA) training, those with general nursing and midwifery qualifications, those at higher facility levels, and those conducting >30 deliveries a month. In all, 14% of respondents were competent at assessment, 58% were competent at making a correct clinical diagnosis, and 20% were competent at providing first-line care.
Birth attendants in the JSY facilities have low competence at EmOC provision. Hence, births in the JSY program cannot be considered to have access to competent EmOC. Urgent efforts are required to effectively increase the competence of birth attendants at managing obstetric complications in order to translate large gains in coverage of institutional delivery services under JSY into reductions in maternal mortality in Madhya Pradesh, India.
Case record maintenance under the JSY program:
To gain insight into the quality of care in facilities implementing the Janani Suraksha Yojana (JSY) cash transfer program in Madhya Pradesh, India, by reviewing the level of documentation in the clinical records of women who delivered.
The present retrospective, descriptive study reviewed case records of women who delivered at 73 primary, secondary, and tertiary level facilities in three districts of Madhya Pradesh between 2012 and 2013. Twenty elements of care were assessed encompassing clinical history and admission details, care during delivery and postnatal period, and discharge details.
A total of 1239 records were reviewed. The extent of documentation varied among the elements assessed—e.g. 24 (1.9%) records documented advice at discharge, 171 (13.8%) documented postnatal blood pressure, 437 (35.3%) documented fetal heart rate, and 1220 (98.5%) documented admission date. The extent of documentation was better at higher level facilities.
The quality of clinical documentation in the JSY program was found to be unacceptably poor in Madhya Pradesh. Improving staff skills and practices in clinical documentation and record keeping will be required to enable clinical processes to be assessed and quality of care to be improved.
Referrals services under the JSY program:
In 2009 the state government of Madhya Pradesh, India launched an emergency obstetric transportation service, Janani Express Yojana (JEY), to support the cash transfer program that promotes institutional delivery. JEY, a large scale public private partnership, lowers geographical access barriers to facility based care. The state contracts and pays private agencies to provide emergency transportation at no cost to the user. The objective was to study (a) the utilization of JEY among women delivering in health facilities, (b) factors associated with usage, (c) the timeliness of the service.
A cross sectional facility based study was conducted in facilities that carried out > ten deliveries a month. Researchers who spent five days in each facility administered a questionnaire to all women who gave birth there to elicit socio-demographic characteristics and transport related details.
35% of women utilised JEY to reach a facility, however utilization varied between study districts. Uptake was highest among women from rural areas (44%), scheduled tribes (55%), and poorly educated women (40%). Living in rural areas and belonging to scheduled tribes were significant predictors for JEY usage. Almost 1/3 of JEY users (n = 104) experienced a transport related delay.
The JEY service model complements the cash transfer program by providing transport to a facility to give birth. A study of the distribution of utilization in population subgroups suggests the intervention was successful in reaching the most vulnerable population, promoting equity in access. While 1/3 of women utilized the service and it saved them money; 30% experienced significant transport related delays in reaching a facility, which is comparable to women using public transportation. Further research is needed to understand why utilization is low, to explore if there is a need for service expansion at the community level and to improve the overall time efficiency of JEY.
Availability of blood under the JSY program:
In order to study the availability, distribution, and functioning of blood transfusion services for obstetric care in Madhya Pradesh secondary data regarding location and service outputs of blood banks and blood storage systems in the public and private sector in MP was obtained from concerned government departments at the state. Facility locations were mapped on Geographical Information System (GIS). GIS Analysis techniques were used to study the availability and geographic distribution of blood banks/storage units.
A qualitative study was performed to understand provider’s perspective, facilitators, and barriers to provision of blood availability. In total, 20 interviews were conducted in Ujjain and Shahdol districts. We interviewed senior officials handling managerial work in the blood bank at districts, senior and junior obstetricians, medical officer at block level to explore opinions and experiences about the provision and use of blood services in general and specifically for obstetrics cases in Madhya Pradesh.
In total list of 145 blood banks/storage units were obtained. Out of these Private (57), Civil and Other/Govt. Hospital (17), Society/Charitable Trust (14), District Hospital (44), Medical Collage/Research Centre (13). The distribution of blood allowed access to most villages/ towns in each district if a buffer of 50 KM around blood banks was considered.
Study participants highlighted various challenges for providing effective blood bank services in the districts. The process of getting a license from FDA was felt cumbersome due to several conditions. The system to refer maternal cases was disused and many times it was observed that those needing blood came late to facilities that could perform life-saving interventions like blood transfusions due to an ineffective referral system. Blood bank managers also raised the issue of limited human resources and non-functioning blood storage units at periphery facilities that overloaded the district blood banks. Many expressed the opinion that the number of blood storage units should be increased, mainly where the load of patients was highest and where more deliveries were happening. Issues of high anemia rates among women were also raised. Awareness among patients regarding the blood replacement was also highlighted during the interviews. The issue of a ‘professional donor’ (i.e. someone who donated blood on a regular basis for money) was also raised during the interviews.
At community level
In the community level, phase a survey of women in sampled villages was done to ascertain JSY uptake at community level. Of the 2607 women surveyed, 1995 were JSY beneficiaries, i.e 76% of births occurred under the progam. Among women who delivered at home (9%) the main reason given for home deliveries was that the baby came unexpectedly and quickly (n=312, 52%). Other reasons included; planned to have a home delivery (n=97, 16%), transportation related issues (n=95, 16%), no one to accompany them to the hospital (n=58, 10%) and other (n=37, 6%). Only one woman replied she could not afford to deliver in a health facility. 15% of women delivered in private sector facilities. 96% of births under the JSY program were vaginal deliveries, while only 4% were cesareans. However the performance of cesarean sections was much higher among women who delivered in facilities outside the program – 35%.
Complications treated under the program: Among the entire study sample, there was a 9% complication rate. This rate did not significantly differ between JSY beneficiaries and non-beneficiaries. Women had a median of one complication, and this was also similar in both groups. The prevalence of specific complications was similar in both groups with the exception of malpresentation. This is most likely the case for the differences in ‘Time until placenta came out’ and ‘Placenta manually removed’ as well. However, we believe the high number of caesarean sections in the non-beneficiary group is driving this difference. Pre-eclampsia (4.7%) and post-partum hemorrhage (2.2%) were the most common complications among both JSY beneficiaries and non-beneficiaries.
Out of pocket expenditure among women who delivered under the program and those who did not: Nearly women who delivered under the JSY program had some out of pocket expense. Women who delivered under JSY program had a higher median, IQR OOPE ($8, 3-18) compared to at home ($6, 2-13). Among JSY beneficiaries, poorest women had twice the net gain ($20) versus the wealthiest ($10) post cash transfer. Informal payments (64%) and food/baby items (77%) were the two most common sources of OOPE. OOPE made among JSY beneficiaries was pro-poor; poorer women made proportionally less expenditures compared to wealthier women. In an adjusted model, delivering in a JSY public facility led to increased odds of incurring expenditures (OR: 1.58 95% CI: 1.11-2.25) but at the same time to a 16% (95% CI: 0.73-0.96) decrease in the amount paid compared to home deliveries. Wealth was not a predictor of having OOPE, but it was an indicator of how much the women would pay. Among women who delivered in a private facilty the median out of pocket expense was nearly 20 times that of JSY mothers.
Participation and non- participation in the JSY program: views from the community
One-fifth of all global maternal deaths occur in India. Further, only 39% of Indian women delivered in a health facility in 2005-06. It was within this context in 2005, that the Indian Government implemented the Janani Suraksha Yojana program that incentivizes poor women to give birth in a health facility by providing them with a cash transfer upon discharge. JSY has helped raised institutional delivery to 74% in the eight years since its implementation. Despite the success of the JSY in raising institutional delivery proportions, the large number of beneficiaries, and the cost of the program, there have been few qualitative studies exploring why women participate or not in the program. The objective was to explore reasons why women choose to participate or not in the JSY program.
In March 2013, we conducted 24 individual in-depth interviews with women who delivered the previous 12 months in two districts of Madhya Pradesh, India. Qualitative framework analysis was used to analyze the data.
Our research found that most women participate in the JSY program because of a shift in the social norm towards delivering in an institution. Drivers of the shift include social pressure from the ASHA and dai to deliver in a health facility, and a growing individual perception of the importance for ‘safe’ and ‘easy’ delivery. Whilst the incentive was an important influence on many women’s behavior, others did not perceive it as an important consideration in their decision to deliver in a health facility. Many women reported procedural difficulties to receive the benefit. Retaining the cash incentive was also an issue due to out-of-pocket expenditures incurred at the facility. Non-participation was often unintentional and caused by personal circumstances, poor geographic access or driven by a perception of poor quality of care administered by program facilities. Persistent issues with quality of care and out-of-pocket expenditures have not received sufficient attention under the program.
The Chiranjeevi program:
Policy level:
The contribution of the CY to institutional delivery in Gujarat:
Many low-middle income countries have focused on improving access to and quality of obstetric care, as part of promoting a facility based intra-partum care strategy to reduce maternal mortality. The state of Gujarat in India, implements a facility based intra-partum care program through its large for-profit private obstetric sector, under a state-led public-private-partnership, the Chiranjeevi Yojana (CY), under which the state pays accredited private obstetricians to perform deliveries for poor/tribal women. We examine CY performance, its contribution to overall trends in institutional deliveries in Gujarat over the last decade and its effect on private and public sector deliveries there.
District level institutional delivery data (public, private, CY), national surveys, poverty estimates, census data were used. Institutional delivery trends in Gujarat 2000–2010 are presented; including contributions of different sectors and CY. Piece-wise regression was used to study the influence of the CY program on public and private sector institutional delivery.
Institutional delivery rose from 40.7% (2001) to 89.3% (2010), driven by sharp increases in private sector deliveries. Public sector and CY contributed 25–29% and 13–16% respectively of all deliveries each year. In 2007, 860 of 2000 private obstetricians participated in CY. Since 2007, >600,000 CY deliveries occurred i.e. one-third of births in the target population. Caesareans under CY were 6%, higher than the 2% reported among poor women by the DLHS survey just before CY. CY did not influence the already rising proportion of private sector deliveries in Gujarat.
Under the CY, there have been >600,000 beneficiaries. While caesarean proportions are higher under CY than before, it is uncertain if all beneficiaries who require sections receive these. Other issues to explore include quality of care, provider attrition and the relatively low coverage.
The CY program: policy maker perceptions on program and implementation challenges experienced in the CY public private partnership
We did interviews with key informants in the state government, non governmental stakeholders and the obstetric gynecology professional body in the state. We explored the perceptions of stakeholders on the implementation Chiranjeevi Yojana, and the implementation challenges experienced in the CY public private partnership
Participants were selected using purposive, non-probability sampling among key figures in government, professional membership organizations and experts involved in the conceptualization and implementation of Chiranjeevi Yojana. Participants represented government stakeholders, members of obstetric and gynecological professional organizations, and experts involved in the design and implementation of CY. All interviews were conducted in the state capital or in their districts.
Efforts were made to obtain interviews with key informants with a diverse range of perspectives. This enabled us to identify what perspectives and experiences around Chiranjeevi Yojana were common to the informants, as well as areas of difference. Snowball sampling was used to contact informants likely to provide insight into research questions.
There was a divergence of views between government and non government stakeholders on the place of the CY in the long term vision of health services. The government officials saw CY designed as temporary public private partnership till public sector EmOC services are adequate. However other stakeholders felt that if public health services had failed to deliver adequate infrastructure and resources to provide women safe hospital childbirth services in the last six decades, it might suggest that the program was likely to continue to be needed in the long term
A program with strengths and weaknesses: Respondents felt that program had strengths -CY has made more choice available to poor women in rural areas, it seems to have reduced out of pocket costs and has reduced mortality and morbidity compared to other options.
However the program was also challenged by implementation weaknesses: On the other hand, there were problems with low uptake, low coverage among the poorest, lack of interdepartmental coordination (particularly around the issuance of BPL (below poverty line) cards, low awareness about the scheme was also cited as a reason for poor uptake. Problems with documentation around the BPL card were also reported as being responsible for low uptake. Also though the intention of the program was to provide free intra-partum care, beneficiaries reportedly had to make some payments, as both public and private sector respondents asserted. A public sector government official actually justified private obstetricians taking some payment from CY beneficiaries as there were long delays in receiving reimbursement from the government! Another difficulty was the attrition of private obstetricians participating over time. A common difficulty faced by the government department in implementing the program was that policy was often driven by individual leadership. Thus when the leadership changed, the policy also changed. This was experienced with regard to waxing and waning commitment to the CY program.
A number of stakeholder commented on the poor monitoring of the CY program by the state. There was also progressive mistrust of the state by the private sector in the context of CY. In general, private obstetricians felt that the government made key decisions regarding the partnership on its own. Private providers were upset over having to provide services to undeserving clients, simply because they had a BPL card. They felt the state had failed to identify poor persons correctly, because of which they were being exploited. They also resented delays in the receipt of the reimbursements.
In the long run, stakeholders perceived that there could be some difficulties with long term sustainability of the CY, not because of financial reasons but because of changes in policy stemming from frequent transfers of key decision makers. The state government perception was that the program would end once the state was strong enough, though district level officials perceived the contrary.
In conclusion, there seemed to be uncertainty on the long term future of the CY program for a number of reasons but largely because support to the program was subject to the views of the incumbent policy makers. This support waxed and waned at periodic intervals. There were issues with the functioning of the partnership, they were lack of oversight from the government, difficult in smooth functioning at the interface of the public private partnerships, possibly because of this being a first large scale partnership. Though it was well designed, there was inadequate focus on the implementation issues, putting in place the required processes well before the program started and a lack of experience on the part of both partners with dealing with a partnership.
Facility level:
Level of EmOC functioning in the the CY program in Gujarat:
The level of EmOC functioning of program facilities is comparable to eligible non-program facilities and much higher than public sector facilities. 35% of CY facilities functioned at CEmOC levels, meaning they had the ability to manage life-threatening complications. 25% of non-program facilities functioned at this level (and <15% in the public sector). All CY program facilities were staffed with fulltime obstetricians and gynaecologists (vs. 63% in non program facilities), the facilities had a high proportion of nursing staff, both trained and untrained. The program facilities though lacked in house access to ancillary services like pharmacy, urgent ultrasounds and laboratory tests. These needed to be done in separate facilities outside the main facility. Such services were more frequently available in the non-program facilities, possibly because many of these were larger hospitals. However simple but key lab investigations like haemoglobin, typing and cross matching could be done at program facilities. A large proportion of CY program facilities also had key equipment to manage complications like vacuum extractors, outlet forceps as well as equipment for routine labour like mucus extractors. Program facilities had on site emergency drugs like magnesium sulphate, oxytocin, this was less available in non-program facilities. 50% of program facilities had no emergency vehicle for transportation; 25% had some private arrangement for transport. Quality control measures were similar to non-program facilities
The kinds of facilities that participated in the CY program partnership:
Private CEmOC ‘facilities which participated in the CY program were newer, smaller, conducted a lower proportion of deliveries by caesarean sections and were owned by obstetricians with less than 5 years of experience compared to private CEmOC facilities which did not participate in the CY program. One-fourth of private providers located in district capitals and half of those in small block towns participated in the CY program
Emergency Transport in Gujarat, supporting the CY rogram
Due to the unpredictable nature of maternal complications effective transportation services are essential to ensure all women obtain timely medical care. In 2007, 525 emergency transportation ambulances, termed the 108 service were implemented across the state of Gujarat under a public private partnership. This emergency transportation supports the CY program users. This service provides free 24/7 Emergency Response; assisting 41,057 deliveries since inception. This study aimed to establish the characteristics of the emergency transportation service users; the mode of transportation among vulnerable women not utilizing the emergency transportation service & quantify the proportion of women arriving at the health facility <2hours after deciding to leave home.
All facilities conducting >10 deliveries per month in three districts of Gujarat were visited for a five day period between July 2012 and May 2013. Post-‐natally administered questionnaires obtained socio-‐economic and transport characteristics of all women; women residing in rural areas (n=1056) were included in this study.
Privately hired vehicles (54.0%) and 108 emergency transportation service (25.4%) were the most prevalent modes of transportation. 108 emergency transportation service use was higher among women who planned a government facility delivery (53.3%), had low education (31.9%), were poor (31.7%), travelled less than 20km (30.6%) and had no specified delivery plan (30.5%). 68.2% of poor women did not use the 108 service with 24.6% of these travelling to government facilities by hired transportation. 94% of women arrived <2 hours after deciding to leave home and 76.4% <1 hour.
Current transportation networks and road infrastructure in Gujarat enable the majority of women to reach health facilities in less than one hour. The high usage of the 108 emergency transportation service among low educated and poor mothers demonstrate its’ success in strengthening access to maternal care among vulnerable women. However, coverage among poor women is not comprehensive and many continue to utilize hired transportation.
Outcomes in CY facilities and non CY facilities:
We surveyed 901 women giving births in high load facilities (all facilities in our study districts that did >30 deliveries in the last 3 months). A third gave birth in the public sector, the remaining two third delivered in private sector facilties. Half of these women delivered in CY facilities (1/3 of the overall sample) but only half became program beneficiaries (1/6 of all eligible).
Eligible women who gave birth in private non CY facilities had the highest Cesarean rates of 20%. In CY facilities cesarean rates were lower (8%, similar among beneficiaries and non beneficiaries, though the number of Cesareans was too small to make conclusions with certainty). The public sector had the lower proportion of cesareans at 5%. Women stayed in the hospital after a vaginal delivery for 12-18 hours. Complication were equally seen in all three facility groups.
Quality of care:
We used exit interviews where the WHO indicators to assess positive (and some neutral/harmful pratices) that the eligible women were subjected to in the facility. There were no differences in the practices that CY beneficiaries and non beneficiaries in CY facilities were subject to. However women who delivered in non CY facilities were more likely to receive interventions not deemed necessary like intravenous fluids or enemas. Women who delivered in a public facility had the least interventions indicative of positive, neutral or negative care.
Access to blood under the CY program: This study was undertaken in the study districts. A mixed method approach was adopted. Data collected included secondary data from 31 blood banks on different input and output indicators. 26 Interviews with stakeholders such as blood bank managers, pathologists and beneficiaries.
Preliminary analysis shows district level variation in level of access and cost of blood transfusion. Use of blood transfusion services for maternal health varied in 3 study districts with highest proportions in Sabarkantha (28%) and lowest in Surendranagar (15%). Median cost of blood varied between 460 and 940 Indian Rupee in study districts. Cost of blood units also varied across type of blood banks- government, private and private not for profit despite price fixing by National AIDS Control Society. Qualitative interviews with blood bank managers and beneficiaries revealed that low levels of voluntary blood donation is an issue with peripheral blood banks. Changed norms of blood banking make it more difficult for private and smaller blood banks to conduct voluntary blood donation drives. Skilled dedicated human resources for blood banking services is the major challenge in study districts (rural areas) but not for urban areas. In Dahod mothers faced maximum challenge in getting blood and had to pay even after replacement. Professional donors are reality in the rural areas. Waiting time for getting blood bag was about 2 hours after reaching the blood bank. Mothers found expense of blood transfusion very high. The process of obtaining licence for blood bank and storage is a lengthy and complicated process with many agencies involved. The regulations have made blood safe but scarce in rural areas as many blood banks do not have the permission to conduct donation camps. Blood bank managers, especially private blood banks, felt that infrastructure and human resources requirements are at times unreasonable. Majority of blood storage centres are defunct as mother blood banks do not have enough blood units. The process of licencing of blood banks and renewal is tedious and lengthy process. No analysis of data at the district or state level for monitoring and improving access to blood for mothers has been made.
Motives for participation and non participation in the program by private obstetricians:
In India a lack of access to emergency obstetric care contributes to maternal deaths. In 2005 Gujarat state launched a public-private partnership (PPP) programme, Chiranjeevi Yojana (CY), under which the state pays accredited private obstetricians a fixed fee for providing free intrapartum care to poor and tribal women. A million women have delivered under CY so far. The participation of private obstetricians in the partnership is central to the programme's effectiveness. We explored with private obstetricians the reasons and experiences that influenced their decisions to participate in the CY programme.
In this qualitative study we interviewed 24 purposefully selected private obstetricians in Gujarat. We explored their views on the scheme, the reasons and experiences leading up to decisions to participate, not participate or withdraw from the CY, as well as their opinions about the scheme's impact. We analysed data using the Framework approach.
Participants expressed a tension between doing public good and making a profit. Bureaucratic procedures and perceptions of programme misuse seemed to influence providers to withdraw from the programme or not participate at all. Providers feared that participating in CY would lower the status of their practices and some were deterred by the likelihood of more clinically difficult cases among eligible CY beneficiaries. Some providers resented taking on what they saw as a state responsibility to provide safe maternity services to poor women. Younger obstetricians in the process of establishing private practices, and those in more remote, 'less competitive' areas, were more willing to participate in CY. Some doctors had reservations over the quality of care that doctors could provide given the financial constraints of the scheme.
While some private obstetricians willingly participate in CY and are satisfied with its functioning, a larger number shared concerns about participation. Operational difficulties and a trust deficit between the public and private health sectors affect retention of private providers in the scheme. Further refinement of the scheme, in consultation with private partners, and trust building initiatives could strengthen the programme. These findings offer lessons to those developing public-private partnerships to widen access to health services for underprivileged groups.
Does the CY increase geographic access to obstetric services?
The high rate of maternal mortality in India is of grave concern. Poor rural Indian women are most vulnerable to preventable maternal deaths primarily because they have limited availability of affordable emergency obstetric care (EmOC) within reasonable geographic proximity. Scarcity of obstetricians in the public sector combined with financial barriers to accessing private sector obstetrician services preclude this underserved population from availing lifesaving functions of comprehensive EmOC such as C-section. In order to overcome this limitation, Government of Gujarat initiated a unique public–private partnership program called Chiranjeevi Yojana (CY) in 2005. The program envisaged leveraging private sector providers to increase availability and thereby accessibility of EmOC care for vulnerable sections of society. Under CY, private sector providers render obstetric care services to poor women at no cost to patients. This study examines the contribution of the CY in improving availability of CEmOC services between 2006 and 2012 in three districts of Gujarat, India.
Primary data on facility locations, EmOC functionality, and obstetric bed availability were collected in the years 2012 and 2013 in three study districts. Secondary data from Census 2001 and 2011 were used along with required geographic information from Topo sheets and Google Earth maps. ArcGIS version 10 was used to analyze the availability of services using two-step floating catchment area (2SFCA) method.
Our analysis suggests that the availability of CEmOC services within reasonable travel distance has greatly improved in all three study districts as a result of CY. We also show that the declining participation of the private sector did not result in an increase in distance to the nearest facility, but the extent of availability of providers for several villages was reduced. Spatial and temporal analyses in this paper provide a comprehensive understanding of trends in the availability of EmOC services within reasonable travel distance.
Community level
In the community phase on the ground, of eligible women who gave birth in the study villages; 28% gave birth under the program. 21% of women delivered in public sector facilties, 33% delivered in the private sector (a little less than half in CY eligible facilities), 15% delivered at home. We explored why women who gave birth in eligible facilities were not given the benefits of the program; reasons included lack of documentation (2/3), doctor deciding not to provide benefits (1/3).
We studied delivery outcomes among this group. The results from the community based survey gave us different results than the did the earlier facility based study – this is likely because the denominator here includes all women (including home deliveries) and that we obtained a much larger pool of eligible women, resulting in larger sub groups. This allowed us to make more robust conclusions. Overall in the entire cohort there was a 7% cesarean section rate. Among the eligible women who gave birth at a CY facility, there was a stark difference among CY beneficiaries and non beneficiaries – the former group had a cesarean rate of 4.6% while the latter had a cesarean section rate of 13.4%. This would indicate that it is possible that private obstetricians move women who needs cesareans out of the program but within their own facilities. The proportion of cesarean sections among CY beneficiaries (4.6%) was very similar to that among eligible women who delivered in public sector facilities (4.5%). Among women who delivered in the private facilities (on CY), the rate of cesarean (15.6%) was similar to those non CY beneficiaries in CY facilities.
There was also a difference detected at community level as to where eligible women chose to deliver. A significantly higher propotion of primiparous women chose to give birth in private sector facilities than did in public sector ones. Four in five women who delivered at home were multiparous. Among the entire study sample, there was a 17.8% complication rate; 15.7% among CY beneficiaries and 18.6% among non-beneficiaries. This difference was not significant.
The main differences between CY beneficiaries and non-beneficiaries were education, income, caste and delivery type. CY beneficiaries tended to having higher education (median 7 years) compared to non-beneficiaries (median 5 years). However, they tended to be poorer (5000 INR compared to 6000 INR). A higher proportion of CY beneficiaries were from scheduled tribe (55.8%).
There was also a difference in the out of pocket expenses that the women experienced. On the whole women in the cohort had an average of 3037 INR as delivery related expenses. Among CY beneficiaries, this was lower at 1621 INR. In public sector faciliites, this was 618 INR. Among non beneficiaries, there mean OOPE 3572 INR. This trend was similar when disaggregated by vaginal and abdominal deliveries.
We also did qualitative interviews with eligible women in the community who delivered under the program to study their understanding of the program.
Women who delivered under the CY program had a largely economic understanding of what the program provided in relation to their childbirth.
‘Chiranjeevi means getting back 2500 rupees – that is the benefit. What else improvement can it bring? Now people say CY is here- government is paying money. As government is giving grant to Dr X (private doctor), he is giving money back.’ (Family member of parturient).
‘CY is that way that we give the money and then we get it back’.
Women perceived that even though they had to pay upfront, the program did help reduce economic burdens for women and their families.
‘Chiranjeevi Yojana is good treatment. People….if someone has not got money then also, if not today then tomorrow, the doctor will give them money. People will spend 3000-5000 rupees, so if they get this yojana, they will get this benefit.’
There was much importance attributed to ‘filling the form’, the key paperwork required to become a CY beneficiary. It was nearly as if for women/families, if they got the form filled right, they could be sure they would get their money back.
‘We paid but it does not matter. Money will be returned to us as we have filled the form.’
‘If the form is ready, they don’t charge you. If it not ready, then they will charge you.’
One woman reported that her husband had organized to have the form filled. She did not know who had helped him do it but that he had sought help. Another beneficiary said that the detailed paperwork was fixed before delivery.
The program also created an interesting impression on the ‘public-ization’ of private healthcare facilities because of the new program – that deliveries were free in private hospitals now (like in government hospitals).
‘Private hospitals have now become government. So private hospitals distribute money after delivery. CY is good because the plan is for a free delivery’.
However a constant theme which was reiterated was that beneficiaries did pay the private obstetrician for their deliveries upfront and then had this paid back to them once the state paid the private obstetrician. This was a common thread reported by women in the different study districts.
In another set of interviews with women and their families, a number of families had benefitted from CY and said it was a key factor in their decisions about where to give birth; they also expressed a wish to benefit from the scheme again in the future.
However, a number of barriers to using CY were reported: 1) Many facilities were no longer offering CY, which disappointed families who had previously accessed its benefits; in one case, the scheme was still advertised in the clinic but the family were still denied the scheme; 2) the bureaucracy associated with applying to benefit from the scheme was seen as onerous; 3) BPL cards were not transferrable across geographic areas, which caused particular problems for migrant workers to access the scheme; 4) families reported being charged for medicines used during care at private facilities, and not reimbursed for these costs; 5) a few doctors were reported as more eager to offer services under the national insurance scheme RSBY.
Some families in possession of a BPL card perceived that they had no need of the scheme as they could afford to pay for delivery care (and would lose social status by using a government scheme). A few families said that cost was not a key decision-making factor, they would find the money somehow and that cost would not prevent them from seeking private health care.
• CY beneficiaries seem to be paying providers in advance for their delivery. This amount is returned to the woman/ family subsequently, possibly if the woman is approved to be a CY beneficiary by the government depending on the paperwork submitted by the private obstetrician. The private obstetrician repays the woman/family on receiving the claimed reimbursement from the state
• Where families have used CY in the past and local providers have withdrawn CY services, families would like to see it reinstated
• Low awareness and uncertainty about CY may have limited its effectiveness in improving access to delivery care for the poorest.
• Bureaucratic barriers to poor families in accessing CY benefits need to be reduced
• Approaches to preventing additional charges for patients need to be developed. Due to the importance of social networks in disseminating information about facilities and schemes, such overcharging may reduce families’ willingness to use the scheme over time as well as potentially impoverishing poor families further.
• Women generally have limited or only partial input into decision-making about where to give birth. Key decision makers within families need to be reached with information about the CY scheme as well as pregnant women themselves.
Besides this there was also a study of the RHINCAV and CHIMACA projects done. These also looked at healthcare financing in the Chinese context, the CHIMACA particularly for maternal health outcomes.
Project dissemination:
Plans for dissemination, particularly among policy makers, civil society groups and researchers were clearly defined (an entire work package has been dedicated to this) at the outset. As stated, open access scientific publications were published from this project, as far as possible to allow national and international access to reports and evidence coming out of this proposal.
Besides scientific publications, the project included a close collaboration between the consortium and stakeholders including policy makers, civil society groups and researchers in the field. Policy makers and program managers were actively engaged throughout the entire evaluation process. Involvements of stakeholders in the initial methodological consultations were also important. On completion of the evaluation, widespread access to the results were ensured, through the distribution of policy briefs and the conduction of national dissemination meetings with the government of India and the state governments. Getting research into policy has been an important priority of this project. We actively involved local policy makers in the process of the evaluation research, besides making the results readily accessible to them. The process of dissemination and engagement is expected to continue even after the life of project.
At the end of the project, there were formal dissemination meetings program held in India, with participation from all partners on the project. Meetings were held at national level with the government of India, and the state governments of Madhya Pradesh and Gujarat. The findings of the project were disseminated formally at a large meeting. Participants included representatives from the Maternal Child Health Division, Government of India, Indian Council for Medical Research, academia, Federation of Obstetric and gynaecological societies, as well as other stakeholders and development partners in each state working with maternal and child health. In addition the project has presented finding at the number of international conferences, academic papers have been published. More locally dissemination of the relevant findings have been done upto grass root level with community health activists (ASHAs).
As stated above, dissemination of the project findings is expected to continue well after the project has ended
Ethical approvals for the MATIND project: The project received ethical approvals in its entirety from all five participating sites in the respective countries. Periodic updates, major changes in study design were informed to the respective ethical boards as mandated by the country rules. In addition to ethical approval, the project received legal permission for implementation from the Indian Council of Medical Research, New Delhi, India and permission from the Maternal and Child Health Division, Ministry of Health and Family Welfare of the Government of India, New Delhi.
Potential Impact:
This project comprised the evaluation and comparison of the impact (including cost effectiveness) of two large scale health programs targeted towards reducing maternal mortality. Both the proposed programs were innovative, a first of their kind, in that there have not been similar demand side programs with an explicit focus on maternal healthcare (which in it itself is a new focus area for a large scale innovative program). Also both these programs were large scale programs, implemented by governments over large regions and not pilot experiments. No such large scale programs have been evaluated before though there have been repeated calls in scientific literature for the effectiveness of such programs to be studied.
Also maternal mortality continues to remain a problem for the poor largely because of access barriers to emergency obstetric care. In 2015, it remains one of the MDGs to remain unachieved, though progress has certainly been made. The pending agenda around maternal mortality is expressed in the sustainable development goal to reduce maternal mortality to <70 globally. India still contributes 70% of maternal deaths in south Asia and to 19% of maternal deaths globally.
Demand side financing as a mechanism to allow poor mothers to access institutional EmoC has never been implemented before on a large scale. The proposal will be a first in terms of comparison of outcomes between the two different demand side financing mechanisms employed in large scale programs. Further, both these programs are targeted toward Millennium development goal 5 (progress towards which has been slow and difficult), to reduce maternal mortality.
The private health sector plays an important role in the provision of healthcare in many low income settings, their contribution as part of a large program to achieving a public good (safe delivery in this case) has not been assessed before. Also there is a current push for innovative mechanisms towards the achievement of universal health coverage in low middle income settings, often in partnership with the private sector. This evaluation has thrown light on this aspect
Socioeconomic impact and wide societal implications:
The project was focused on India, a country still contributing to 19% of global maternal deaths. Success with reducing maternal deaths is desperately necessary in India if we are to support the new sustainable development goals. Also the political priority accorded to maternal mortality by recent governments has seen the large scale implementation of innovative solutions over the last decade years which are comprehensively evaluated for the first time.
The project makes an important contribution towards looking at what works from a health system perspective with regard to maternal mortality reduction. Maternal mortality has a strong poverty and gender dimension. Maternal mortality has been described as ‘‘a 21st century problem essentially only for the poor, and one virtually eliminated for people with the resources to access health services”. Even, within poor countries, the poorest women are as much as six times more likely to die during childbirth as richer women. Inequalities in the use of maternal health services mirror those in maternal health outcomes. While there are many dimensions to disadvantage (including race, ethnicity, age, urban/rural status, etc.) and while the causes of inequity vary among and within countries, it is however clear that women from the wealthiest groups consistently make the most use of skilled care during pregnancy, delivery, and the postpartum period and have the best chance to survive childbirth. Indeed, analyses of inequalities in use of health services have shown that rich-poor ratios are often greater when it comes to professional delivery assistance than for other basic health services. Not only are wealth inequalities greater for use of maternity care, but little progress has been achieved in reducing these particular inequalities, particularly in sub-Saharan Africa and South Asia.
The project through the evaluation of these programs for maternal mortality reduction indicates issues in program design and implementation that can influence this goal. Issues from the ground at the level of facilities implementing the program, user and community perspectives have been captured. The findings from this project have important lessons for low middle income governments aiming to start programs aimed at maternal mortality reduction.
From a health systems point of view, the project has studied a number of aspects related to demand side financing programs. Since the project began the literature around demand side financing in the area of maternal health care has widened. However much of the reporting has been from small programs in small areas run with donor support in low income settings. These programs studied under MATIND are different given their scale and that they are completely tax funded by the government. The program empirically demonstrates the prerequisites that need to be in place prior to a successful demand side finding program being implemented and the implications of these not being in place. There has been an urgent need to know what works programmatically in the area of maternal mortality in the context of low income settings. The MATIND project has spoken to this need.
The project also generates knowledge on public private sector partnerships for public health outcomes in the global south. This is particularly relevant given that the direction forward in many low middle income countries to achieve universal health coverage is through partnerships with the private sector. The MATIND project delves deeply into partnership issues and has explored key enabling factors as well as barriers to successful partnerships using these two programs as examples.
Dissemination:
Plans for dissemination, particularly among policy makers, civil society groups and researchers were clearly defined (an entire work package has been dedicated to this) at the outset. As stated, open access scientific publications were published from this project, as far as possible to allow national and international access to reports and evidence coming out of this proposal.
Besides scientific publications, the project included a close collaboration between the consortium and stakeholders including policy makers, civil society groups and researchers in the field. Policy makers and program managers were actively engaged throughout the entire evaluation process. Involvements of stakeholders in the initial methodological consultations were also important. On completion of the evaluation, widespread access to the results were ensured, through the distribution of policy briefs and the conduction of national dissemination meetings with the government of India and the state governments. Getting research into policy has been an important priority of this project. We actively involved local policy makers in the process of the evaluation research, besides making the results readily accessible to them. The process of dissemination and engagement is expected to continue even after the life of project.
A brief overview of dissemination strategies and events is presented below:
All European, Indian and Chinese partners met and interacted at the official MATIND kick off meeting in August 2011 as well as the 2nd annual meeting in December of 2013 and the final consortium meeting in August 2015. Besides the kick-off meeting, there has been a number of cross partner collaboration meetings:
Partners Involved Where When Objective
1,3,6 Ujjain, Madhya Pradesh and Ahmedabad, Gujarat May 22nd to June 4th 2011 To discuss project planning, selection of districts, secondary data variables, phase wise implementation plan, process involved to develop the forms, SOPs, and logistics. Initial exploratory field visits were also performed.
1,3,4,5,6 Ahmedabad, Gujarat August 24th – 27th 2011 The official MATIND kick-off meeting
1,3,4,6 Ujjain, Madhya Pradesh August 31st-September 3rd 2011 To discuss the development of the protocol and qualitative topic guide.
1,5 Hangzhou, China November 2011 To discuss the role of Partner 5 in the project
3,4,6 Ujjain, Madhya Pradesh November 12th-14th 2011 To discuss study instruments for the facility based portion of WP2 and to do field visits
3,6 Ujjain, Madhya Pradesh December 23rd-25th 2011 To discuss the logistical implementation of gathering GIS coordinates for the study districts.
1,3,6 Ujjain, Madhya Pradesh January 20th-25th 2012 To discuss the roll-out plan for the Shajapur pilot, plan for main study, logistics, GIS plan, red cap data entry software
1,4 Liverpool, UK April 3rd 2012 To discuss WP 4 – private providers participation in the programs
1,3 Stockholm, Sweden May 5th-10th 2012 MATIND project discussion; plan development, mainly qualitative studies and data Analysis of pilot data, Phase 2 discussion
1,3,6 Ahmedabad, Gujarat May17th -27th2012 To review the initial results and implementation of Facility survey
1,3 Stockholm, Sweden June 11th, 2012 To discuss the cost effectiveness studies
1,3 Stockholm, Sweden June 13th 2012 To discuss the community based studies
1,4 Liverpool, UK June 22nd 2012 Qualitative portion of the MATIND project
1,3 Stockholm, Sweden July 3rd- 7th2012 Data entry and analysis of data from Shahdol district
3,6 Ahmedbad, Gujarat August 23rd- 25th 2012 GIS related discussions
4,6 Ahmedbad, Gujarat August 2012 To discuss the qualitative work in Gujarat
1,3,4,6 Stockholm, Sweden September 14th 2012 To discuss facility and community based studies
1,3,6 Stockholm, Sweden September 17th- 20th 2012 To discuss community based studies and the cost effectiveness studies
1,3,6 Madhya Pradesh and
Gujarat, India 22nd May – 4th June 2011 To discuss project planning, selection of districts, secondary data variables, phase wise implementation plan, process involved to develop the forms, SOPs, and logistics. Initial exploratory field visits were also performed.
1 & 4 Liverpool, UK 21-¬‐22 June 2012 To discuss the progression of the qualitative work in the
DOW
1, 3 & 6 Stockholm, Sweden
10th Sept – 30th Oct 2012 MATIND Project Meeting to discuss economic studies & Manuscript writing
1, 3 & 6
Ujjain, Madhya Pradesh 30th Nov – 4th Dec 2012
Discussion on WP 2 completion and planning for WP 6
1, 3 & 6
Ujjain, Madhya Pradesh
& Udaipur, Rajathstan
28th Feb – 17th March
2013
Finalization of WP 6 studies, data collection and database construction
1, 3 & 6
Ahmedabad, Gujarat 24th – 26th March 2013
Results sharing from WP 2
1 & 3
Stockholm, Sweden 12th Apr – 1st May 2013 Project meeting, discussion of methodology for community based studies (WP 6/7), manuscript writing
1 & 4
Liverpool, UK 12th-¬‐13th Feb 2013 To discuss data collected on women’s experience and reasons for participating in the JSY program (WP 6 obj v)
1, 3, 4, & 6
Stockholm, Sweden
29 April to 1 May 2013
Discuss analysis of observational study data on quality of care in MP (WP 2 iii) including developing coding framework and charting. Also to discuss the analysis of qualitative interview data on women’s perceptions of quality of care in Gujarat (WP 2 iii). Partner 4 administered
training in qualitative analysis of interview data, framework analysis
1, 3 & 4
Stockholm, Sweden
19-¬‐20 May 2013
Discuss and present studies in WP 2 from MP.
1 & 4 Liverpool, UK 3-¬‐6 June 2013
To discuss coding framework on women’s experience and reasons for participating in the JSY program (WP 6 obj v)
1 & 4 Liverpool, UK 10-¬‐11 July 2013 To discuss descriptive analysis for the study on experience and reasons for participating in the JSY program (WP 6 obj v
3 & 6
Ujjain, Madhya Pradesh
24th – 27th July 2013
Discuss database management for community based studies
(WP 6/7)
1 & 5 Stockholm, Sweden 18-20 September 2013 Progress on RHINCAV/CHIMACA reviews. Analysis of the NCMS data
1, 4 & 6
Stockholm, Sweden
27th September 2013
To discuss technical and logistical issues regarding qualitative data collection, management and analysis, including management of transcription and translation; to discuss concept and process for study of emergency transport (WP 2iii)
4 & 6
Stockholm, Sweden
29th Sept – 2nd October
2013
Discuss the development of CY policymaker interviews (WP
1 ii) including protocol development, training in qualitative analysis of interview data.
3 & 4
Madhya Pradesh, India
24th-¬‐30th November 2013 To discuss further analysis of observational study data (WP
2 iii) including interpretation and developing a paper for publication
All partners Khajuraho, Madhya
Pradesh 2nd-¬‐6th December 2013 2nd Annual Partner Meeting
1 & 3
Ujjain, Madhya Pradesh 4th – 8th March 2014 Project team meeting, publication plan, discussion of blood bank study design (WP2) and qualitative work (WP 4/5)
4 & 6 Gujarat, India 17th–18th March 2014 To discuss further analysis of qualitative interview data on women’s perceptions of quality of care in CY facilities (Gujarat) (WP 2 iii), and policy maker perceptions of CY
3 & 4 Madhya Pradesh, India 19-¬‐21 March 2014 To discuss further development of qualitative study of perceptions and experiences of ASHAs (WP 5 ii), including training RA and reviewing pilot interview data, and to discuss analysis of qualitative study of private providers in JSY 2 (WP 4) and study of policy makers perceptions of JSY (MP) (WP 1 iv).
All partners Delhi, Ahmedabad, Gujarat, Bhopal, Madhya Pradesh 4th – 9th August 2015 Dissemination meeting to the Central and local policy makers/Governments
Project results have been presented at the following conferences:
Conference
Name Date of Conference Abstract
Name Authors Status
13th World Congress on Public Health 23-27 April 2012 Is MMR Really Declining – Challenges in Monitoring Ng, M., Levin-Rector, A., Diwan, V., Sabde Y., De Costa, A. Presented as poster (Abstract #124)
13th World Congress on Public Health 23-27 April 2012 India's JSY Cash Transfer Program for Maternal Health: Who Participates and Who Does Not – a Report From Ujjain District Sidney, K., El-Khatib, Z., Diwan, V., De Costa, A. Presented as poster (Abstract #120)
Second Global Symposium on Health Systems Research (HSR)
Beijing, China 31 Oct –
3 Nov 2012 'Improving access to maternity services: an overview of cash transfer and voucher schemes in South Asia' Jehan, K., Sidney, K., Smith, H. and de Costa, A. To present as poster
Second Global Symposium on Health Systems Research (HSR)
Beijing, China 31 Oct –
3 Nov 2012 Need to strengthen BEmOCs? – A GIS based study on the utilization of maternal health service in Madhya Pradesh, India. Sabde, Y., Randive, B., Chaturvedi, S., Singh, M., Sidney, K., De Costa, A., Diwan, V. To present as poster
Global Maternal Health Conference 2013: Quality of Care 15-¬‐17 Jan
2013 Janani Express: Utilization of an Emergency Maternal Transportation Service in Ujjain District, Madhya Pradesh India Kristi Sidney, Bharat Randive, Ayesha De Costa, Vishal Diwan Poster
Global Maternal Health Conference 2013: Quality of Care 15-¬‐17 Jan
2013 India’s conditional cash transfer programme to promote institutional births: What has been the effect on maternal mortality? Bharat Randive, Vishal Diwan, Ayesha De Costa Oral
Global Maternal Health Conference 2013: Quality of Care 15-¬‐17 Jan
2013 What has been the effect of JSY on maternal mortality? Studies from Madhya Pradesh, India Ng M, Misra A, Diwan V, Levin, Rector A, De Costa A Oral
Global Maternal Health Conference 2013: Quality of Care 15-¬‐17 Jan
2013 Too close to die: A study of effectiveness of referral system for emergency obstetric care using spatial analysis in Madhya Pradesh, India Sarika Chaturvedi, Bharat Randive, Manish Singh, Ayesha De Costa, Vishal Diwan Oral
Global Maternal Health Conference 2013: Quality of Care 15-¬‐17 Jan
2013 Does SBA training improve quality of care?: Studying the link between SBA training and provider practices under the JSY cash transfer programme in Madhya Pradesh, India Sarika Chaturvedi, Bharat Randive, Yogesh Sabde, Kristi Sidney, Ayesha De Costa Poster
Global Maternal Health Conference 2013: Quality of Care 15-¬‐17 Jan
2013 Which mothers travel to reach a facility under the cash transfer program? A network analysis from Madhya Pradesh, India Yogesh Sabde, Sarika Chaturvedi, Zia Ur Rehman, Sayyad Ali, Manish Singh, Kristi Sidney Poster
Global Health Metrics & Evaluations Conference 17-¬‐19 June
2013 Impact of a public private performance based financing partnership on C-¬‐section proportions: Cross-¬‐sectional study from a district in Gujarat, India Kranti Vora, Parvathy Raman, Dileep Mavalankar, Veena Iyer, Kristi Sidney, Rajesh Mehta Poster
Global Health Metrics & Evaluations Conference 17-¬‐19 June
2013 Effective coverage of institutional deliveries under the JSY program in high maternal mortality provinces of India Bharat Randive, Sarika Chaturvedi, Vishal Diwan, Ayesha De Costa Oral
Global Health Metrics & Evaluations Conference 17-¬‐19 June
2013 Initial results on the impact of Chiranjeevi Yojana: a public–private partnership programme for maternal health in Gujarat, India Marie Ng, Parvathy Shanker-¬‐ Raman, Rajesh Mehta, Ayesha DeCosta, Dileep Mavalankar Poster
3rd NETWORK MEETING with regard to EPI-¬‐4 project, Ahmedabad 27th June
2013 Inequity in access to maternal health care:under JSY in Madhya Pradesh. Yogesh Sabde, Vishal Diwan, Ayesha De Costa Oral
5th International Conference on healthGIS 2013 21st- 23rd August
2013 Public private performance based financing program for institutional delivery – Chiranjeevi Yojana: Which private facilities participate? Which eligible women participate? A spatial analysis from Sabarkantha district of Gujarat, India Veena Iyer, Ayesha De Costa, Dileep Mavalankar Oral
5th International Conference on health GIS 2013 21st- 23rd August
2013 An analysis to study access to emergency transport vehicles under a scheme called “Janani Express Yojana (JEY)” in two districts of MP, India using spatial approach. Yogesh D Sabde, Sarika D Chaturvedi, Kristi Sidney, Bharat B Randive, Ayesha De Costa, Vishal Diwan Oral
Joint State Conference of IPHA-¬‐IAPSM, Pune India 25th ‐26th Feb
2014 Better pregnancy outcomes need more than bringing women to hospitals: A call to improve quality of institutional birthsobstetric referral services in India’s JSY cash transfer program for Sarika Chaturvedi, Bharat Randive, Vishal Diwan, Ayesha De Costa Oral
National Seminar on maternal neonatal health and safe abortion: opportunity and challenges held at Hyderabad April 21-27 2014 Availability and distribution of safe abortion services in rural areas: a facility assessment study in Madhya Pradesh, India Sarika Chaturvedi, Bharat Randive, Vishal Diwan, Ayesha De Costa Oral
Global Health System (HSR) Conference, Captown South Africa- 29.09.14 to 05.10.14 Which mothers bypassed nearer facility under the cash transfer program? - A network analysis from Madhya Pradesh, India Yogesh D Sabde, Sarika D Chaturvedi, Bharat B Randive, Ayesha De Costa, Vishal Diwan Poster
Global Health System (HSR) Conference, Captown South Africa- 29.09.14 to 05.10.14 An equity analysis of institutional delivery uptake and maternal mortality reduction in context of cash transfer program (JSY): results from 9 states of India Randive, B., San Sebastian, M., De Costa, A., Lindholm, L Oral
11th World Congress in Health Economics- Milan July 11-16 2015 Out of pocket payments made during stay at facilities by women eligible to receive cash transfers to give birth in facilities under India’s JSY program: findings from Madhya Pradesh Sidney K, Salazar M, Marrone G, Diwan V, DeCosta A, Lindholm L. Oral
11th World Congress in Health Economics- Milan July 11-16 2015 India’s JSY cash transfer program to promote facility births: Increased access to facility based care but not to quality care Raven J, Chaturvedi S , Diwan V, De Costa A Oral
11th World Congress in Health Economics- Milan July 11-16 2015 The Chiranjeevi Yojana, a public-private-partnership for obstetric services in Gujarat, India: Who are the private participating obstetric service providers? Iyer V, Sidney K, Malavankar D, DeCosta A Oral
The following articles have been published in international peer-reviewed journals:
1. Sidney K, Diwan V, El-Khatib Z, De Costa A. India's JSY cash transfer program for maternal health: who participates and who doesn't--a report from Ujjain district. Reprod Health. 2012 Jan 24;9:2.
2. Jehan K, Sidney K, Smith H, de Costa A.Improving access to maternity services: an overview of cash transfer and voucher schemes in South Asia. Reprod Health Matters. 2012 Jun;20(39):142-54.
3. Sidney K, De Costa A, Diwan V, Mavalankar DV, Smith H. An evaluation of two large scale demand side financing programs for maternal health in India: the MATIND study protocol. BMC Public Health. 2012 Aug 27;12(1):699.
4. 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. 2013. 8(6): e67452. doi:10.1371/journal.pone.0067452
5. De Costa A, Vora K, Ryan K, , Raman P, Santacatterina M, Mavalankar D. The state-¬‐led large scale public private partnership 'Chiranjeevi program' to increase access to institutional delivery among poor women in Gujarat, India: How has it done? What can we learn? PLOS ONE. 2014. 9(5): e95704. doi:10.1371/journal.pone.0095704
6. Ng M, Misra A, Diwan V, Agnani M, Levin-Rector5 A, De Costa A. An assessment of the impact of the JSY cash transfer program on maternal mortality reduction in Madhya Pradesh, India. Glob Health Action 2014, 7: 24939 – (http://dx.doi.org/10.3402/gha.v7.24939(öffnet in neuem Fenster))
7. 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. Social Science and Medicine, 2014, 123: 1-6 http://dx.doi.org/10.1016/j.socscimed.2014.10.042(öffnet in neuem Fenster)
8. Sabde Y, De Costa A and Diwan V. A spatial analysis to study access to emergency obstetric transport services under the public private “Janani Express Yojana” program in two districts of Madhya Pradesh, India. Reproductive Health 2014, 11:57. http://www.reproductive-health-journal.com/content/11/1/57(öffnet in neuem Fenster)
9. Chaturvedi S, Randive B, Diwan V, De Costa A (2014) Quality of Obstetric Referral Services in India’s JSY Cash Transfer Programme for Institutional Births: A Study from Madhya Pradesh Province. PLoS ONE 9(5): e96773. doi:10.1371/journal.pone.0096773
10. Sidney K, Ryan K, Diwan V, De Costa A. Utilization of a State Run Public Private Emergency Transportation Service Exclusively for Childbirth: The Janani (Maternal) Express Program in Madhya Pradesh, India. PLoS ONE 2014, 9(5): e96287. doi:10.1371/journal.pone.0096287
11. Chaturvedi et al. Competence of birth attendants at providing emergency obstetric care under India’s JSY conditional cash transfer program for institutional delivery: an assessment using case vignettes in Madhya Pradesh province. BMC Pregnancy and Childbirth. 2014. 14:174.
12. Lennart Bogg, Vishal Diwan, Kranti S Vora and Ayesha De Costa. Impact of alternative maternal demand-side financial support programs in India on the caesarean section rates – indications of supplier-included demand. Maternal and Child Health Journal. 2015. DOI 10.1007/s10995-015-1810-2
13. Chaturvedi S, Upadhyay S, De Costa A, et al. Implementation of the partograph in India’s JSY cash transfer programme for facility births: a mixed methods study in Madhya Pradesh province. BMJ Open 2015;5:e006211. doi:10.1136/bmjopen-2014- 006211
14. Ganguly PS, Jehan K, De Costa A et al. Considerations of private sector obstetricians on participation in the state led “Chiranjeevi Yojana” scheme to promote institutional delivery in Gujarat, India: a qualitative study. BMC Pregnancy and Childbirth. 2014. 14:352. (http://www.ncbi.nlm.nih.gov/pubmed/24787692(öffnet in neuem Fenster))
15. Vora K, Yasobant S, Mavalankar D. Predictors of Availing Maternal Health Schemes: A community based study in Gujarat, India. (2014) Indian Journal of Community Health. 2014:26 (2); 174-180.
16. Chaturvedi S, Ali S, Randive B, Sabde Y, Diwan V, De Costa A. Availability and distribution of safe abortion services in rural areas: A facility assessment study in Madhya Pradesh, India. Global Health Action. 2015, 8: 26346 - http://dx.doi.org/10.3402/gha.v8.26346(öffnet in neuem Fenster)
17. Yasobant, S., Vora, K.S. Hughes, C., Upadhyay, A. and Mavalankar, D.V. (2015) Geovisualization: A Newer GIS Technology for Implementation Research in Health. Journal of Geographic Information System, 7, 20-28. (http://dx.doi.org/10.4236/jgis.2015.71002 )
18. Chaturvedi S, De Costa A, Raven J: Does the Janani Suraksha Yojana cash transfer program to promote facility births in India ensure skilled birth attendance? : A qualitative study of intrapartum care in Madhya Pradesh. Global Health Action. 2015 Jul 7;8:27427. doi: 10.3402/gha.v8.27427
19. Vora, K.S. Yasobant, S., Sengupta, R., De Costa, A., Upadhyay, A. and Mavalankar, D.V. Options for optimal coverage of free C-section services for poor mothers in Indian state of Gujarat: Location allocation analysis using GIS. PLoS ONE. 2015.
20. Chaturvedi S, Randive B, Raven J, Diwan V, DeCosta A. Assessment of the quality of clinical documentation in India’s JSY cash transfer program for facility births in Madhya Pradesh. IJGO. 2015. DOI: http://dx.doi.org/10.1016/j.ijgo.2015.07.016(öffnet in neuem Fenster)
Under review:
1. Vora, K.S. Yasobant, S., Patel, A., Upadhyay, A. and Mavalankar, D.V. Has Chiranjeevi Yojana changed the geographic availability of free Comprehensive Emergency Obstetric Care Services in Gujarat, India? BMC Health Services. 2015.
2. Sidney K, Salazar M, Marrone G, Diwan V, DeCosta A, Lindholm L. Out-of-pocket expenditures for childbirth in the context of the Janani Suraksha Yojana (JSY) cash transfer program to promote facility births: Who pays and how much? Studies from Madhya Pradesh, India. Health Policy Planning. Manuscript.
3. Sidney K, Tolhurst R, Jehan K, Diwan V, DeCosta A: The cash transfer is important but all women anyway go to hospital for childbirth nowadays’ - a qualitative exploration of why women participate in a cash transfer program to promote institutional deliveries in Madhya Pradesh, India. BMC Pregnancy and Childbirth. Manuscript.
4. Sidney K, Iyer V, Vora K, Mavalankar D, De Costa A. State-wide program to promote institutional delivery Gujarat, India: Who participates and the degree of financial subsidy provided by the Chiranjeevi Yojana Program. Journal of Health, Population and Nutrition. Manuscript.
5. Altman RS, Sidney K, De Costa A, Vora K, Salazar M. Is institutional childbirth associated with lower neonatal mortality risk in Gujarat India: A cohort study in a vulnerable population in the context of the Chiranjeevi Yojana (CY) program. Manuscript.
6. Iyer V, Sidney K, Mehta R, Mavalankar D, De Costa A. Emergency Obstetric Care availability and provision under a public-private partnership in three district of Gujarat, India- the Chiranjeevi Yojana, who are the private partners? Lessons for Universal Health Coverage: cross-sectional facility survey. Manuscript.
7. Sabde Y, Diwan V, Randive B, Chaturvedi S, Sidney K, Salazar M, DeCosta A. The availability of Emergency Obstetric Care under the JSY cash transfer program in Madhya Pradesh, India. BMC Pregnancy and Childbirth. Manuscript.
8. Sabde Y, Chaturvedi S, Randive B, Sidney K, DeCosta A, Salazar M, Diwan V. Bypassing health care facilities for childbirth in Madhya Pradesh (MP), India: A multilevel study. Manuscript.
List of Websites:
www.matind.eu
Ayesha Mariette De Costa MD PhD
Assoc Professor – Global health
Dept of Public Health Sciences
Karolinska Insitutet
Stockholm S 171 77
0046852483336
ayesha.de.costa@ki.se