Back to latest

Using data analysis to spot gaps in access to maternal and child health services

Indian rural pregnant women wearing medical masks due to illness, dizziness. Healthy pregnancy concept.

Researchers at CivicDataLab used procurement data, health data and geospatial data to review public health spending in Assam, India. They developed a method to verify if the money is reaching districts where mothers and babies need it most. Their initial findings suggest it isn’t, as fewer maternal and child health tenders are published for districts with high maternal and infant mortality rates, but more complete data would allow them to draw more accurate conclusions.

Analyzing public procurement data can reveal a lot about a government’s priorities and performance,  and which citizens benefit from public spending. When the data is published promptly and accurately, it can even be used to predict and re-assign resources to be distributed more effectively and equitably. 

In India, researchers from CivicDataLab are using data on health infrastructure procurement to understand how investments in maternal and child health relate to health outcomes in the state of Assam. By overlaying data on health spending with actual health indicators, they developed a method for monitoring which initiatives by the state are working, which geographical areas need more assistance and the different kinds of assistance required. 

Assam has one of the highest rates of infant and maternal mortality in India. Around 31 in 1000 babies die in the northeastern state before their first birthday and of the children who do survive, almost 35% of them show signs of malnutrition. 

As most of Assam’s residents rely on the public health system (rather than private care), investments in health infrastructure can go a long way towards improving the well-being of the state’s women and children. CivicDataLab created a website dedicated to exploring Assam’s contracts, sourcing the tender data from the government’s central procurement portal. They prepared the data for analysis, structuring them in a user-friendly, machine-readable format called the Open Contracting Data Standard.

The data showed that 165 tenders for maternal and child health have been published over the last five years, with a total awarded value of over INR 360 million (US$4.7 million), or about 5% of all health related tenders published in the state. It’s worth noting that there may be other tenders unaccounted for, as the source data from the government portal represents only about 20% of the state’s contracts, according to World Bank figures (MAPS Assessment report for Assam, not published online).

Figure 1: Year on year data on the government entities publishing tender information

The top contracting agency was the National Health Mission (NHM), which awarded 60% of maternal and child health tenders, followed by the Health and Family Welfare Department and Public Works Department.

The researchers also documented all the central and state-sponsored funding schemes related to advancing maternal and child health in Assam (see table below).

Scheme NameCentral/StateLaunch YearDetails
Universal Immunization ProgramCentre1985This program distributes vaccines to children free of cost for over 12 preventable diseases. 
Accredited Social Health Activist (ASHA)Centre2005The ASHAs act as a bridge between the community and health care providers, especially the more marginalized and hard to reach communities. They are health educators and promoters in the community
Janani Shishu Suraksha Karyakram (JSSK)Centre2011The initiative entitles all pregnant women delivering in public health institutions to absolutely free and no expense delivery, including caesarean section and including free drugs and consumables, free diagnostics, free blood wherever required, and free diet for 3 days during normal delivery and 7 days for C-section. This initiative also provides for free transport from home to institution, between facilities in case of a referral and drop back home. 
Comprehensive Abortion Care (CAC)Centre2010This initiative provides abortion related facilities for women, including access to safe abortion, post-abortion care, family planning through training of doctors, nurses and ASHAs for appropriate service delivery.
Congenital Heart Disease (CHD) ProgramState2010A scheme that provides free treatment of children with congenital heart defects in Narayana Hrudayalaya Hospital Bengaluru and Kolkata. According to the scheme, the government bears all costs of transport, medical, stay, food etc. 
Janani Suraksha Yojana (JSY)Centre2005The scheme provides cash assistance to mothers who have delivered in a hospital to increase the number of institutional births
MamataState2005The Mamata program seeks to reduce infant mortality rate and maternal mortality rate by incentivising parents to stay post-delivery at the hospital for 48 hours after which a mamata kit being distributed at discharge
MamoniState2009This scheme provides cash assistance for pregnant women toward nutrition as an incentive for antenatal check ups. 
Sneha SparshaState2013This scheme provides financial assistance for children who require specialized treatment for serious ailments 
Pradhan Mantri Surakshit Matritva Abhiyan (PMSMA)Centre2016This scheme offers fixed day antenatal services to women every month
LaQshya ProgramCentre2017The quality improvement initiative in labour room & maternity operating theaters aimed at improving quality of care for mothers and newborn during intrapartum and immediate postpartum period.
Rashtriya Bal Swasthya Karyakram (RBSK)Centre2013This scheme was created to provide referral support to children detected with health conditions during health screening. 
Operation SmileState2009This scheme offered free surgeries for children cleft palate and lip. 

Table 1: List of schemes introduced in Assam toward bettering maternal and child health

There are more than 13 schemes at the central and state level, but procurement in this category was concentrated among the following schemes (see Figure 2 below):

Other notable procurements included tenders for creating maternal and child health wings within district hospitals, building labor rooms and maternity wards in districts with large minority populations, and constructing IVF centers. 

Figure 2: This figure shows the yearly distribution of schemes under which tenders were published. 

Geospatial Analysis

CivicDataLab used geospatial analysis to compare infant mortality rates (IMR) in the various districts of Assam with tenders related to maternal and child health in the same areas.

Infant mortality rates vary wildly across Assam, as Figure 3 shows. In districts like Cachar, Hailakandi, Dima Hasao, and Karimganj, almost twice as many infants die before their first birthday than in districts like Kamrup, Chirang, Baksa.

Figure 3: Map of Assam with infant mortality rates (Note: the IMR data used here are projected data for 2017 since district-wise information on this particular health indicator is not accessible after 2014-15)

A quick look at the geographical distribution of the maternal and child health tenders across the state in Figure 4 shows that areas like Kamrup, Goalpara, Golaghat, Cachar, Karimganj have a lot of tenders published for maternal and child health while others such as Dima Hasao, Karbi Anglong, Udalguri have fewer maternal and child health tenders published. 

Figure 4: Map of Assam with maternal and child health (MCH) tenders published

Overlaying this information in Figure 5 below, helps identify areas that suffer from high infant mortality rates and low government support for maternal and child health i.e. from procurement spend. The central Assam region (consisting of Karbi Anglong, Dima Hasao, Cachar), areas within the Barak valley region (Hailakandi), as well as the upper Assam region (Sivasagar) are highlighted as the areas that need most assistance/attention and with the highest mortality rates. While areas in the lower Assam region including Goalpara, Kamrup (Rural and Metropolitan), Nalbari, Barpeta have better health indicators and a larger concentration of maternal and child health tenders being contracted for those areas. 

Figure 5: Map of Assam overlaying infant mortality rates (IMR) and maternal and child health (MCH) tenders, where dark blue shading indicates districts with a low IMR and dark red with a high IMR. Light green circles indicate a low MCH tender value and dark green a high MCH tender value.

Figure 6: A Composite Map  showing the distribution of the IMR and tender values in Assam. The dark red is the district with high IMR and low tender value while the dark green are districts with low IMR and high tender value. 

Similar analyses can be done overlaying composite health quality indices for children with tender information. Or the density of health infrastructure across districts can also be overlaid with the density of health center tenders  to get a deeper understanding of which areas are constantly overlooked for infrastructure. The population density of each district is another variable that could be factored into the analysis, if such data is available.

These types of analyses can be used to identify and re-assign resources and to understand the holes in the governments’ current resource management strategy. 

Recommendations for government agencies

To improve the accuracy of this analysis, and therefore distribute health resources more effectively and efficiently, the researchers propose the following recommendations: 

  1. The publishing of tenders needs to be accurate and complete.  Currently only around 20% of the actual contracting procedures for Assam are being published. This means there are at least 70,000 tenders from the last five years that have not been published (according to the World Bank’s MAPS Assessment report for Assam). This information and more rigorous publishing practices will help improve the accuracy of calculations and guide a better policy response. 
  2. The publishing of tenders needs to be more timely. For around 30% of the total tenders, the contract award information was published. But the average time from contract award to publishing award information was 250 days (whereas the Finance Department requires this contract award information to be uploaded within 30 days of the award). While the duration shortened over the last five years, the contract award data holds useful information about how much different regions are spending on different tenders and whom these contracts are being awarded to. Being able to analyze this data promptly will help save the government money by setting more competitive prices. 
  3. The publishing of tenders needs to be more standardized. Governments should publish standardized, machine-readable and open data, following for example the international best practice schema Open Contracting Data Standard. To improve the analysis discussed in this blog, specific data to be uniformly published across tenders should include:
  1. Geospatial information of where the works or service will be delivered 
  2. Schemes or initiatives under which the tender is being published
  3. Uniform categories for repeat services and products, for example: MCH wing, upgradation of centers, construction of primary health care centers, covid-19 test kits 

These quick fixes will generate a more useful dataset to analyze and improve the public health of this state and the country. 

Related Stories