From my first day on the job in the Indian Army, I was given the responsibility to lead 1,000 men, take charge of large fleets of vehicles, and move tonnes of ammunition. So doing things at scale came easily to me. Seven years ago, when I shifted to the social sector, I realised that what the army took for granted, the million-plus NGOs operating in our country were struggling with. There are only a handful of names, like Pratham and Magic Bus, that operate at scale today.
So what is scale? Despite a lot of talk about it, there is no standard definition. Loosely in India, any programme that’s done at a state level or crosses 100,000 beneficiaries is referred to as being “at scale”.
I turned to Ashok Alexander to understand his take on scaling public health interventions.
An alumnus of St Stephen’s College, Delhi, and IIM Ahmedabad, Alexander’s career spans 34 years, from being a Director at McKinsey to heading the Bill & Melinda Gates Foundation India office. It was while at Gates and spearheading Avahan, India’s largest HIV prevention programme, that Alexander began visualising an organisation, the key proposition of which would be delivery at scale.
A decade later, he established the Antara Foundation with an aim to improve maternal and child health outcomes in India. Only, Antara and Alexander are doing it, not with the help of an army of human resource, but with data and technology.
What have been your biggest learnings in running large-scale programmes?
When Avahan became the largest HIV prevention programme in two-and-a-half years, one was forced to acknowledge that there was something that made it go to scale so quickly. It taught me that scaling up a social programme needs a three-pronged approach: supply, demand and an enabling environment, backed by the effective use of data, very similar to focus areas in any business.
What does focusing on the supply side mean?
Antara Foundation’s work in Rajasthan focuses on significantly improving maternal and child health outcomes for the state. The responsibility of delivering on maternal and child health on the ground in India rests with what we have started calling the triple As: the ANM (the Auxiliary Nurse Midwife, serving a population of 5,000; 1 per two villages), the ASHA (Accredited Social Health Activist, serving a population of 1,000; 1 per village), and the AWW (Anganwadi Worker, serving a population 1,000; 1 per village). Simply put, there are 2.5 frontline health workers per village in India, with the mandate to deliver on maternal and child health. They are backed by the health structures (primary health centres, hospitals, health policies). On the other side are the beneficiaries: pregnant and lactating mothers, and children from 0-6 years.
The key in public health delivery, we learned, is to optimise the interactions between the beneficiaries and the frontline workers.
Once that happens, delivery at any scale becomes possible. The problem in India is that the triple As don’t collaborate. Belonging to different ministries — the Ministry of Health and Family Welfare (ANM, ASHA); and the Ministry of Women and Child Development (AWW) — there are very few incentives to collaborate, to share data or plan together.
How does data help?
One of the primary roles of an AWW is growth monitoring of all children in her village, between the ages of zero and six. She routinely weighs each child in her Anganwadi Centre, classifies them into normal, moderately underweight or severely underweight, recording these in her Anganwadi register. The ASHA worker, however, follows a different protocol for growth monitoring. She does house-to-house visits and classifies children using MUAC tapes, which measure upper arm circumference. She notes this data in a different register. So a child, let’s call her Madhu, might be classified as severely underweight by the AWW, but not by the ASHA. The ANM, ideally, should have got this information about Madhu from the ASHA and AWW, based on which she should plan to meet Madhu and administer the necessary treatment. But this is where everything falls through the cracks. The three As don’t talk, data filed by all three is in different registers, the registers are copious, and data is either incomplete or lost.
How is Antara trying to overcome this?
Antara has devised a tablet solution, an app, something that the CM of Rajasthan will be announcing in a few weeks, that brings together the data of the threes As on a common platform. So now, what the AWW enters will be available for the ASHA and then for the ANM to see in real time on her tablet. It not only has numbers, but also visual data, such as growth charts, making it easy to read and take action.
What was the process leading up to the development of the software?
One critical step before we developed the app was that of rationalising data. We undertook a six-month exercise across two districts (approximately 2,500 villages in Rajasthan), where we looked through each register that the ANMs were filling, column by column. What we found was that these copious registers had a fair amount of redundancy, duplicate information, and in some cases, just a complicated layout. We then rationalised these registers by reducing the number of columns (by almost 15-20% of the original 544 columns) and consolidating data on a single page or register, which earlier may have been spread over 5. Today, these rationalised registers have already been adopted state-wide in Rajasthan, and our hope is that they can be scaled nationally.
Looking at the future, where do you see technology’s potential in the area of public health?
Big data has a lot of predictive power. It would be ideal if we could have applications that could go through large amounts of relevant maternal and child health indicators and predict where the next hotbed is. This can enable early preventive interventions.
Rupinder Kaur is an ex-army officer, currently working on the implementation and monitoring of large-scale social programmes
