Skoll World Forum Interviews: Spotlight on Digital Health Innovations in India

Host:
Welcome to Role Models for Change, a series of conversations with social entrepreneurs and other innovators working on the front lines of some of the world’s most pressing problems.

Peter Yeung:
Hello, Ruchit, thank you very much for joining us today at the Skoll World Forum.

Ruchit Nagar:
Thanks for having me.

Peter Yeung:
And so, to begin with, can you just introduce yourself and tell us the basics of the work that you’re doing?

Ruchit Nagar:
Absolutely. So, my name is Ruchit, I’m a resident physician training in internal medicine in pediatrics at Yale and I’m also the CEO and co-founder of Khushi Baby. Khushi Baby is a technical partner to departments of health in India and what we do is we build digital health solutions to track primary healthcare in rural communities and at scale.

Peter Yeung:
And I suppose, before we get into digging into things, could you explain what exactly does Khushi Baby mean? I understand the translation is, if I’m not mistaken, is something like Happiness Baby but can you explain what exactly that means and why you’ve chosen that name?

Ruchit Nagar:
Yeah, so it’s a long story. We started off looking at child immunization and, working in India, we were looking to have a solution that could help promote health and happiness amongst kids, amongst babies so Khushi Baby is the thought that came out. But if you look at our journey as an organization, we started with child vaccinations but then we moved to maternal and child health, then, during COVID, we moved into infectious disease, primary healthcare surveillance. So, really, it’s become this perspective that it takes a village to raise a child and raise a child healthy for life. So, really, now we’re looking at all of primary healthcare, khushi health or khushi family if you’ll say and we’re going back to also where we started.
When we started, we were looking at how do we track child vaccinations using a mobile health application, now we’re just doing it at a much bigger scale where we have some 50,000 children that are suspected for not ever receiving a vaccine, how do we get vaccines to them on that level.

Peter Yeung:
Right. And I suppose you are … Obviously, the bulk of the work focusing on Rajasthan in India and I just wonder can you explain … Obviously, you’ve explained what you’re working towards but can you highlight and explain what exactly are some of those issues on the ground and some of the problems that you are tackling?

Ruchit Nagar:
Yeah. So, we work in India’s largest state, Rajasthan, which is home to 80 million people. It’s got vastly different terrains from desert to valleys and we work across the entire state with around 70,000 health workers. We see all types of challenges and what we are trying to do is we’re trying to improve the health system. We’re trying to overcome things like data transparency, data accountability, fair health worker remuneration and timely health worker remuneration and put all of the data back together in a way that it’s actually fed back to the communities. So, we’re seeing that the health system is now moving to being more proactive towards surveillance, using digital tools to get information from what’s happening on the ground and now the next step is how do we actually put those different pieces together, harmonize that data and then make it relevant for the community health workers.

Peter Yeung:
And why do you think that that last aspect that you’re pointing out, just feeding back, as you say, to the communities, why is that important to you and what are the potential benefits of doing that?

Ruchit Nagar:
Before, the culture has always been about reporting data for the sake of reporting. How do we just measure where there are higher areas of disease burden, where are there higher areas of health delivery being actually delivered effectively? But now it’s more about how can we learn from what the data is coming in and then respond to it. And not only just respond, but be more proactive. And there are multiple ways that we can do that whether it’s sending out a messaging campaign for public health, whether it has to do with oncoming heat waves or health awareness for maternal health activities. It has to do with deploying local teams for outbreaks, for example, deploying infrastructure and equipment for areas that are more vulnerable.
There are a variety of different ways that we can actually now start to respond to the data and that’s what communities are actually looking for because they want to feel that, hey, if I’m collecting this data for my village and I’m the community health worker, how is that actually translating into some action on the ground. It’s not just me trying to report a number for the sake of showing that the health system is doing a good job but it should be for me to show the truth to get support that I need so, that way, we can actually improve the health on the ground.

Peter Yeung:
Right, yeah. I think there’s a common frustration of just data for the sake of data without actually taking those next steps of the vision and putting it to use. I just wonder, can you, I suppose, take us through how you developed that baseline? Obviously, you said that the first of actually getting those systems in place to be able to collect the data, what kinds of processes did you have to go through and then also, I suppose, looking ahead when you said building on those community networks, how did you then also go about developing those?

Ruchit Nagar:
Yeah, so it certainly has been a journey. Our approach has been to start with doing a digital health census and empowering the community health worker to start with the census of her village. So, these community health workers known as ASHAs are locally elected by their village, they’re responsible for all kinds of health promotion activities. Every day they visit 10 houses. So, now they have a smartphone that they own for the first time and they’re doing a registration of everyone, they’re asking about social determinants of health and they’re asking about different health risk factors and health outcomes that people are experiencing. This gives us a baseline for the first time.
Previously, baselines were estimated. There would be a census that took place 10 years ago, they would add a formula and then estimate what the population might be or what the disease burden might be but that would sometimes mismatch with the actual situation on the ground. So, when people would set targets, when the officials would set targets, those targets wouldn’t actually match the reality that health workers were experiencing. So, now we’re doing this bottom-up approach.
So, it starts with the census and then, from that census, we have modules within our solution that allow for longitudinal follow-up. So, based on who you are, you might be in the reproductive and child health bucket, you might be in the non-communicable disease bucket, you might be in the tuberculosis or infectious disease surveillance bucket and different people will fall into these different streams where they can be tracked longitudinally. Not only by their community health worker, but by the local nurse, the local doctor who all need to coordinate care to make sure that that person is healthy and well. So, that’s been our approach and to actually apply that approach and work within the government has been another challenge but we’ve also gone step by step.
We initially focused on, okay, how do we make one module work well, how do we make maternal and child health work well. Then during the COVID pandemic, we were faced with a question, do we want to continue to focus on our area of expertise or do we want try to pivot and cross-apply our learnings in digital health to deal with this new pandemic. We decided to take the second approach and focus on primary healthcare more broadly. And slowly but surely, we started to go to different health programs, whether it was the Infectious Disease Surveillance Department, whether it was the Non-Communicable Disease department, the department that’s looking after vector-borne disease and then soon others started to join on. Because, during COVID, we were able to do the first digital health census of the state back in 2020 and now these other programs we’re seeing that we can use that data and piggyback off that process to then do their own screening campaigns and their own public health interventions.
So, step-by-step, using COVID as a catalyst of sorts, we were able to build up a digital health census and then start to add different health programs that could leverage that data for their own longitudinal programs.

Peter Yeung:
Right, yeah. And I think, obviously, the pandemic has been a catalyst for a lot of different projects out there. And I wonder as well then what exactly did you find that was easy applyable but then also, I suppose, in terms of the difficulties that you’d also taken on by taking that wider remit and, obviously, there’s a lot more to deal with and, as you say, different challenges along the way but I’d just be interested if you could talk to that one.

Ruchit Nagar:
Right, that’s such a great question. I think there are a couple of different approaches. You can go deep, be good in your vertical, you can try to go broad. What we think needs to happen is you need to go diagonally, you need to be both deep and broad at the same time. Because, at the end of the day, the health worker is not just working on maternal and child health or tuberculosis or mosquito-borne disease, she’s working at all of those at the same time. And it’s not useful for her to juggle between five different paper registers and three different applications, it should be one place where she can understand this is the health of my community, these are the houses I’m going to visit, these are the priorities that I’m going to focus on. So, that has always been our approach is one of convergence.
Now, on the other hand, there have been established organizations that have domain expertise in specific verticals and they are looking to build on that expertise and promote solutions that may be more verticalized. The issue is that, when they’re promoting these solutions maybe at the central level or even at the state level, there is a disconnect with the reality of what the health workers are experiencing. And as a result, unfortunately, health workers have had to juggle between five different applications for example. They’ve not been able to realize the full potential of an integrated solution like the one that we are proposing. So, these are the ongoing challenges that we face.
And what we are trying to do now is we’re really trying to bring everyone under one umbrella and move forward together that, whether a specific development partner or government program wants a specific data set, is there a way that we can use this integrated solution to get them the data that they need while, at the same, time not burdening the community health worker with multiple different applications where they have to do redundant activity. How can we actually minimize the 20 hours of work that the community health worker spends on data reporting and effectively channel that towards things like training, things like health promotion, counselling and activities within the village? And there are 1.3 million community health workers across India taking care of close to a billion people.
So, 20 hours per health worker really adds up very quickly on a per month basis and that’s where we think digital health has a huge promise. Digital health needs to be strengthened, it needs to be invested in, at the same time, it’s not a panacea. You need to have technology but you need to also consider the human element, the incentive structures, the supporting environment that goes around it. And like I said earlier, just reporting that, hey, there’s a problem in my village is not enough, there needs to be a response. Health officials need to get active and engaged, they need to see the dashboard as an opportunity for them to learn, to test new public health interventions and then to follow evidence-based practices. So, it really will take a partnership from the health worker, the health official, throw in the development partners who are supporting the government, throw in public health researchers who can study these questions more deeply and also a citizen-centric approach.
Eventually, we also want to have the people themselves well-connected within this network. No matter if they’re in the village, they’re going to the district government hospital, they’re going to a private lab to get their health, that data should flow seamlessly and it should come back in a cohesive manner. Doing this for India is going to be challenging, doing this anywhere in the world is going to be challenging but I think we need to respect the fact that this is a complex challenge and work together and not try to step on each other’s toes but really work together to get it accomplished.

Peter Yeung:
Obviously there’s, well, alone in Rajasthan, massive diversity in terms of the populations that are being served, the geographies, of course, as well with the health workers. But what do the health workers, who are these people and what are they … Can you just explain what exactly they’re doing on the ground? What kind of gear are they using? Are they going around with tablets to take this data? What exactly does that look like?

Ruchit Nagar:
Yeah, and that’s a great question. To spend a day in the life of an ASHA worker or a nurse is truly something that is eye-opening. Our team has spent, collectively, over 200,000 hours alongside community health workers in co-creating these solutions. We have a team of 20 field monitors that work in the field every day and what you’ll notice is that they do a lot. They do a lot more than you would expect from a community health worker who might … You would think that they might focus on one or two diseases, no, they have six chapters of public health information that they have to cover. They’re doing everything from family planning, maternal health, checking on newborns, taking pregnant women to the hospital when it’s time for delivery, they’re looking for mosquito larva, they’re surveilling for TB, they were on the front lines during COVID. They’re really doing everything and they’re also experiencing the burdens of population displacement, climate change, other socioeconomic stressors on their populations all at the same time.
Now, on top of all of that, they have to still work on paper to a certain extent, juggle different applications, they have to travel to the primary health center and get some of that data entered. At times, there are situations in which they even have to convince the data entry operator to help submit their data so that they can get paid. And oftentimes their payment comes two to three months delayed, they’re getting paid maybe two to $3 a day and, you can imagine, for how vital their work is. Studies have shown that their work reduces things like neonatal mortality. Why are we not investing more? Why are states only able to even utilize 70% of their national health mission budget and why are we not able to pay community health workers on time on a performance basis when the technology exists to measure how they’re doing?
I think the other challenge is that, in this environment, you can imagine data quality is not perfect, it varies quite a bit. And some of it has to do with incentives and how those are aligned, you have to meet some target that was established based off a census 10 years ago. Sometimes support is not always present, infrastructure, other colleagues in HR that you need to work with. So, it’s a really tough environment and, especially in the hard-to-reach areas, it gets even tougher as you can probably imagine. So, that’s the kind of environment that ASHAs have to deal with and they may go to a home and try to convince a couple to undergo a vasectomy or hysterectomy for family planning purposes and they may receive threats from the family. They may have to go three or four times to the house of a family with a malnourished child in order to accompany and persuade the family to bring that child to the treatment center where a doctor may or may not be present or in which the outcome may or may not be assured and other kids are at home that the family has to take care of or the spouse is out of town.
It’s a complex situation with access to care barriers, sociocultural barriers and they’re trying to navigate all of it. What’s great about the ASHA is that she has trust with the community. She’s really seen as a friend, a peer and it’s interesting to see how that interface is going to change now that she has a phone, now that she’s entering in some data as she goes to speak with people and how that will affect the level of trust, how that will affect her ability to work efficiently. But ultimately, again, how can we use this technology, if we’re going to use it, how are we effectively using it to amplify her voice on the ground? Are we actually able to then allow her to report the ground truth in an easier manner, in a manner that is then supported with follow-up resources or is this just going to become another burden for her? And I think that is the line that we have to walk very carefully because we don’t want to tip onto one side.

Peter Yeung:
Right. And I imagine, yeah, it’s something of a messy transition. There’s no easy progress there but it is very pleasing to hear, obviously with your focus on the correct payment for this work. I think even places like the UK, very different context. The pandemic, there was a big shift of recognition towards the value of these frontline workers and putting their bodies on the line. Obviously, in this case with community health workers, as you say, doing really crucial work too. And I just wanted to touch on the point of, more broadly, the digitalization aspect of things. Now, as you say, before, these roles were still offered but paper-based and that kind of thing. But could you speak to the potential that you see in the digitalization? Obviously, with the data, I suppose the speed with which this information is transmitted but then also a lot more, presumably, complex analyses that can be done. Can you just explain what that potential is?

Ruchit Nagar:
Yeah, no, there’s tremendous potential. So, our solution is called Community Health Integrated Platform or CHIP for short. And when we started during the initial wave of the COVID Pandemic, we understood that we needed to make this technology widely accessible. We went away from deploying the software on tablets and we put it onto the Playstore and we didn’t even know if ASHA workers were going to be able to download this application, not all of them had phones. What happened as a result, they would borrow phones from their neighbor, maybe their son or daughter had a phone, they took the phone from their spouse, they convinced their spouse that I need to have my own phone because I have to do my work for the first time and we saw a tremendous increase in smartphone ownership, smartphone use.
And one thing led to another but it was that initial step of digital empowerment that really got us excited. That, now, ASHAs are digitally empowered, they have access to their own smartphone and now they can start to take decisions for their own communities. We think that this is just a starting point because the smartphone is not just a tool for data reporting, it’s a tool to help the ASHAs plan who they need to visit first, second, third, to prioritize based off different risk categories. It is a way for them to measure their progress on a weekly or monthly basis, to estimate how much they might earn for financial planning based off the services that they’ve delivered.
The phone is also a future diagnostic tool in of itself. You can use the camera to take photos of somebody’s eyelids to estimate things like their hemoglobin level, you can listen to the sound of their cough to stratify if this sounds like a TB-related cough or a COVID-related cough, you can take photos of rapid diagnostic tests to automatically interpret the results and now the smartphone can, therefore, be used as a way for you to verify that the health worker actually met the person and that there is some level of data quality assurance there.
The smartphone is a tool for learning. YouTube, WhatsApp, ASHAs are in these groups. We ourselves are in around 500 groups across the state of different ASHA coalitions. We’re able to see how they learn, share YouTube videos, do peer learning. Now, with large language models, that’s only going to increase. Even when there’s not an in-person mentor or somebody to talk to, there can be chatbots that allow them to learn from their own curriculum. Of course, you need to put in the right guardrails to prevent things from going off the wrong direction with hallucinations but there’s tremendous potential just in that phone. I think there is tremendous potential in terms of what we can do for the flip side of the coin which is the health officials, the partners, the researchers because, now, for the first time, these groups are getting access to highly granular health data. Yes, there’s a lot of signal and noise that is both coming in and you have to figure out what is the signal and what is the noise but, for the first time, you’re getting some sense which communities are sicker, which communities are healthier, which communities are at more risk. And not only can you see that on a map, but then you can start to think about what interventions can be targeted for those specific health problems and how do we use this whole platform as a tool for measurement.
It’s not just to see how well we are doing, how well we’re progressing but can we measure the public health interventions that we are trying to deploy on a local level? Now we have the tools to do that, now we have the interface to do that. So, that’s what really completes the loop. It’s getting people excited about, not just a streamlined way of collecting public health data, but getting them excited about, hey, this data can now be actionable for the first time and in of itself valuable. And despite all of the other challenges, the different apps that they have to use, payment delays, et cetera, if they can find value in the tool itself, they will come and use it and there will be a group of champion users that will lead this effort forward. And we’re looking to learn from them, learn from their experience, build off that experience in the coming years.

Peter Yeung:
Yeah, I think it’s easy to overlook the immense power that there are in smartphones because, I suppose, we’re used to it now in many ways but, really, the world at your fingertips. And I just wonder, so I had seen, obviously, you’ve done some RCT, some trials previously in the work you’d done. But I suppose can you take us through, I suppose, what exactly is the impact that you’ve managed to make and what are those successes?

Ruchit Nagar:
I love this question and it’s also such a difficult question to answer which is how does a digital health intervention create impact on the ground. It’s not like we are distributing food, building a school, it’s something that is working at a systems level. It’s not a direct intervention. There are stages to how we think about impact. First, we think about processes that we’re improving, then we think about attitudes, perceptions, behaviors that we’re improving and then, finally, practices and outcomes that we’re improving and it is a journey along that process. So, we first do things like measure is our system stable, operational, do users like it, are they engaged with it, does it change their daily activity. Also, when you’re starting off on a small scale, you can try to measure how their use of the system actually leads to health impacts more directly.
We did do a two-year, 3,200 mother randomized control trial back between 2016 and 2018. In one arm, the villages were getting the digital health intervention. The nurses had the mobile app, they would get a reminder call when it was time to go in for their health camp and their vaccination and, on the other hand, it was the standard of care. And what we found is that, when you put the digital health intervention there, you’re able to improve the infant’s vaccination rates, the full infant vaccination schedule through 12 months by 12 percentage points. And as a secondary effect, we were able to reduce malnutrition rates by four percentage points. These are meaningful impacts from a public health context and this was done across around 400 villages this study.
We had this health outcome data and then we went to the state saying that, hey, we want to replicate this, we want to scale this up. And they looked at the evidence and they considered it and they said, “Well, okay, you can take this but we want to cross apply your learnings and we have other challenges as well. We need to integrate all these systems and COVID came and we have to do something for that.” So, the whole nature of the intervention grew and evolved and now, as we think about the intervention which is operating at a systems level at the state, running a randomized control trial on whether or not the system at the state is working is not necessarily the same approach. Instead, what we have to think about is mixed methods research that looks at, okay, how are people using the platform, to what extent are they delivering data-driven interventions, to what extent are we actually able to identify new diseases.
So, out of the 45 million people that we’ve screened, around 5 million or so, we’ve identified as being high risk across different disease categories, whether maternal and child health, non-communicable disease or infectious disease. That data has now been shared with the departments for those respective programs to then take the required follow-up actions. But then we also have to measure the next step. Did people get referred, did people get put on treatment, were those treatments actually cured. So, the more deep you go, you fall into this cascade in which people drop out at each step and we’re still at the early parts of strengthening the cascade of it, at least improving disease detection. That’s currently where we are on our impact journey but we need to go further.
And then I’ll also say that it also requires us to do qualitative research. We need to engage with focus groups of different stakeholders to understand how they’re using the system, how they want to improve the system, why they feel the system is impactful along the whole value chain. Because that information about the impact is going to inform how we iterate upon our process and how we’re going to make it relevant for an ecosystem of multiple stakeholders. The government official wants something very different than the health worker. The government official may need some reports on a timely basis, the health worker may want a simplified user platform. How do you bring those two together, how well are we addressing both of their needs, we need to be able to measure that as well. And there are ways now with technology that we can measure this easily. We can, through the platform itself, we can do things like rapid polls and surveys where we get 500 answers from ASHAs within half an hour across the entire state which gives you a snapshot into how they’re thinking, what they have access to, where they need more support.
And I think the last thing putting this all together is that we are thinking of this digital health tool, not as an intervention in of itself, but rather as a tool to measure other interventions. This is a platform in which you can actually study did my public health intervention work or not. When I put a cooling center here in this area that experiences heat waves, did that make a difference or not on heat strokes, on kidney disease? When I sent out a message for family planning service promotion, did that increase the uptake of long-acting reversible contraceptions in this particular area? When I deployed a team to a specific geography where there were under-immunized children, were we able to have a successful vaccination camp and identify all those children? We are able to now measure the impact because community health workers have come together in this network and they’re sharing data from their communities. That I think is the actual strength and, when we know what works, then we can replicate those public health interventions in like-minded geographies.

Peter Yeung:
Yeah, some really, really interesting work and thanks for taking the time to explain that. And although you do already seem to have a lot of work and a focus on a lot of projects already, I just wonder, in the future, are you looking to … What are the plans of the future? Obviously, we spoke about with your regional focus at the moment with Rajasthan but is it in the near future that you’re looking spread to other states as well or, I suppose, developing other aspects of the work?

Ruchit Nagar:
Absolutely. So, we are looking to move to multiple states as well. So, Karnataka and Maharashtra which are neighboring states which will allow us to span from North India to South India, some 250 million plus people compared to the 50 million that we’ve currently been working with right now. And each new area, in fact, each new district has its own contextualization and localization that you need to do but being able to work across multiple states will be a new challenge that we’re excited to tackle. With multi-state evidence of a system working, we have a better chance of advocating at the union or central level to make this a digital public good and that’s our end goal.
Our end goal is to make this software solution, all of our standard operating practices for how to implement this solution on the ground available for all these departments of health, no matter what state you are in India or if you’re the central government, to adopt for free and be a resource to them as they look to contextualize the solution for their specific geographies. That’s how we think we can ultimately make the biggest impact on a systems level.
And at the same time, as I talk about going broad and expanding more broadly, we are also looking to make impact on a more deep vertical basis. So, whether that means helping identify and immunize over a thousand zero-dose children in the state of Rajasthan who have never received a vaccine, it’s incredibly difficult to find where they are. But once we find them, we want to ensure that the health camps are there so that they can get their catch-up vaccines, doing this across every single domain. We have some 15,000 individuals who are suspected for TB, ensuring that they get their confirmatory testing, they’re put on treatment and are followed through their treatment journeys until they are converting to negative. Each program requires you to go really deep and we also want to go deep in Rajasthan where we already have a base of working there for four years to make even greater impact.

Peter Yeung:
I suppose, could you describe to us, this India that you’re working towards, what it looks like and what the vision is in the future that you hope the country to be in terms of the health situation?

Ruchit Nagar:
I think we have to be aspirational and I love this idea of a program called aspirational blocks. These are the blocks or geographies across India where you have the hardest socioeconomic conditions, the poorest health outcomes. Why are they called aspirational blocks? They’re called aspirational blocks because we want to make them, not only come up to the standard of the neighboring geographies, but we want them to become centers of excellence. We want them to become places where you want to aspire to raise your level towards. I think that’s the philosophy that we need to take. It’s not to elevate the public health system to the bare minimum but really to make it excellent. And I think, in many ways, India is leapfrogging, has a history of leapfrogging. They leapfrogged the landlines and went almost directly to cell phones when it comes to mobile banking or the unified payment interface.
In many ways, India has leveraged technology to leapfrog and get ahead in various other domains. Health remains one domain in which there still needs to be more investment. About 2% of the GDP is being invested and not all of it is even being utilized. We need to make efforts to leapfrog in health as well. And when you look at how health is connected and we’re here, we’re talking about climate and health, India’s spending around 6% of its GDP on climate preparedness, there needs to be some crosstalk between the two.
In general and on the whole, there needs to be more collaboration. We need to move away from this vertical siloed approach of this is my domain of maternal health or child health or TB, et cetera, and really think about what are the linkages. Not just within the health department but between the health department and the Department of Women and Child, education, public planning. There’s this concept of one health that is looking at animal health, human health, environmental health altogether and there’s these committees that they’re trying to assemble from the forestry department, the health department, the animal husbandry department to get together and share data sets and work together for the first time.
We’ve never had to do this on this level except in disasters, in situations like COVID in which you had to coordinate very quickly. But sooner rather than later, we are going to feel the pressure of this exponential curve and we’re going to need to sit together at the table. So, I think, really, my vision for the future is to make technologies accessible, to make collaboration easy, and to make it truly multi-sectoral. Government, development partners, researchers, citizens, community health workers all on the same … Get them on the same page and get them aligned towards the common goal. Even if that means, initially, that the additional transparency may make things appear to be worse than they previously were, over time, we will see the benefits of how a coordinated system can reduce adverse health outcomes.

Peter Yeung:
Well, thank you so much for your time.

Ruchit Nagar:
Thank you so much.

Introduction (00:02):
Welcome to Role Models for Change, a series of conversations with social entrepreneurs and other innovators working on the front lines of some of the world’s most pressing problems.

Dr. Aparna Hegde (00:13):
Hi, I’m Dr. Aparna Hegde. I’m the founder of the NGO ARMMAN that leverages technology to create cost-effective scalable solutions to impact maternal and child health in 21 states of India.

(00:26):
In collaboration with the government of India, our mobile health programs have provided over 47 million women with preventive care information through pregnancy and infancy, and trained over 371,000 health workers in the early identification, treatment, and referral of high-risk factors to reduce maternal and child mortality and morbidity in India.

Matthew Beighley (00:49):
How does it work exactly? What’s the technology and what kind of messages are going out?

Dr. Aparna Hegde (00:54):
In the next one hour, three women will die in childbirth somewhere in India. Two children under age five die every minute in India. India’s malnutrition rates are worsening. India has one of the highest under five mortality rates in the world.

(01:12):
Two primary reasons are, one, lack of access to information. Women do not have access to preventive care information through pregnancy and infancy, and hence they do not demand care for themselves and their children, do not recognize the risk factors, the complications.

(01:28):
And, on the other hand, the health workers at every level of the health system, the front-line health worker, that is the ASHA, the nurse, the medical officer, and the specialist are not really aware of their part of the treatment process. How to identify the risk factors early, managing time before it’s too late, and that leads to very high numbers of delayed or complicated referrals to tertiary hospitals or irrational referrals leading to increased maternal and child mortality and morbidity.

(01:59):
Given the sheer scale of India’s numbers, it is very important that any solution I designed had to be scalable, accessible to every woman in her home and get cost-effective.

(02:12):
Technology allows that. We leverage our unique tech plus touch model. That is, we use mobile health solutions to reach out to women and health workers alike weekly, but we also have the requisite touch points for feedback, two-way communication, and hence the solution is scalable.

(02:31):
Our programs have two arms. On the beneficiary side directly to the women, we have a program called Kilkari in collaboration government of India, which is the largest search mobile health messaging program in the world. Kilkari is a voice-based program that sends preventive care information directly to the phones of the enrolled women through pregnancy until the child is one year of age.

(02:55):
If women have access to information, they’ll demand care, but on other hand, we have the health workers at every level of the health system. That is the frontline health worker called the ASHA, then the nurse, then the medical officer, and the specialist, and we have programs for all of them.

(03:12):
Our program for the frontline health worker that is the ASHA is called Mobile Academy, and it is in collaboration with the government of India. It is currently a voice-based program. There are four [inaudible 00:03:23] of content. There’s a quiz at the end of each of the 11 sections, and there’s a bookmarking system. We’ve reached around 371,000 ASHAs in 17 states of the country, and we hope to reach 850,000 ASHAs PAN India in the next three to five years.

(03:42):
Moving forward, Mobile Academy will have an version two, which will be purely multimedia through WhatsApp or a [inaudible 00:03:51]. There will be live action videos, simulations, interactive quizzes, and more handholding.

(03:58):
Then we have the program called Integrated high Risk Pregnancy Management and Tracking program for the nurses, medical offices and specialists. High risk factor in pregnancy is where the mortality happens because each layer of the cadre does not understand what is their part of the role in management of high risk factors, how to treat the woman, when to refer, what counseling to give, what to do when the woman comes back to them.

(04:24):
So we have created protocols for high risk pregnancy management for every level of the cadre. These are clear, clearly detailed, algorithmic color coded protocols for 35 high risk conditions. They have been modified to suit the state situation in three states of the country, and now they become policy in two states.

(04:47):
Going forward we are actually, I’ll start this going forward again. Is that okay?

Matthew Beighley (04:51):
Sure, yeah.

Dr. Aparna Hegde (04:51):
Till now, it’s okay?

Matthew Beighley (04:58):
Perfect. Yeah, really great.

Dr. Aparna Hegde (04:58):
Really? Yeah.

Matthew Beighley (04:59):
No, trust me. I’d be starting you again if I wasn’t getting [inaudible 00:05:00].

(05:00):
Great.

Dr. Aparna Hegde (05:00):
Yeah. And now on two states of the country, we are doing training of the nurses medical office and specialists face-to-face in these protocols. And in addition to face-to-face training, we also have a learning management system on their smartphone or tablet where all the protocols have been converted into live action videos, simulations, interactive quizzes and notifications for self-paced learning. And when they implement the protocols in the field and they have any query, we have a WhatsApp help desk.

(05:30):
Going forward, we’ll be tracking the pregnant woman with high risk factors end to end. So if a woman becomes pregnant, she becomes a part of the database. If she’s got a high risk factor, it’s tracked. So it leads to dashboards at every level. So we know where we have to go into iterative program improvement. So if a particular part of the state is doing poorly in pregnancy-induced hypertension, we can train the nurses, medical offices, and specialists more in that particular part of the state.

(05:53):
Moving forward, we are hoping to integrate all the three programs, namely Kilkari, Mobile Academy, and integrated high risk pregnancy management and tracking program with AI added in.

(06:06):
So if part of the state is doing poorly in a particular condition, say pregnancy induced hypertension, Kilkari can inform the women more about it. Mobile academy can train the ASHAs more about it.

(06:17):
An integrated high risk pregnancy management program can train the nurses, medical officers, specialists more about it. And with AI, we can do a lot of analytics to improve program efficiency, and in fact even have prediction of high risk factors.

Matthew Beighley (06:31):
Why is it growing so fast? What was the need? What’s your feeling as to why this is just expanding?

Dr. Aparna Hegde (06:39):
Yes, that’s an interesting question. It’s expanding because it’s need of the hour. It’s right for the hour. Women in the country are using phones every day. The smartphone coverage in the country is astounding in rural, tribal, Haryana and Jharkhand, 60% of women own their own phones, own their own smartphones. And 100% of women have access to a family phone. So it’s very logical that the next step would be they to move on to multimedia content, two-way communication through WhatsApp, because WhatsApp is being used throughout the country. Or, even when possible a [inaudible 00:07:17] app.

(07:18):
And so information is everybody’s need and the fact that they have access to phone in their homes it makes it very easy for us to reach out to them with information weekly. The same is the case for the health workers.

(07:33):
Every health worker now has a smartphone. They need to be trained, they need to be hand-held. They need to ask questions. And what better tool than the smartphone, which you and I also use for the same purposes. And that’s why, because these programs so logical and the next step anyway for training world over, if you look at any training program world over, the next step is obviously this.

(07:56):
And in the pandemic, the use of digital health was very obvious. Everybody moved online, including laypeople and health workers. So we are just keeping pace with the changes that are naturally happening in the world, and that’s why the programs are expanding so exponentially. And I also want to highlight here the critical aspect of government support.

(08:21):
The government of India is committed to ensuring that every woman and every health worker is trained and they’re providing the finances for it. They’re supporting these programs. And so the financial issues also is taken care of because they’re solely funded by the government of India. And that’s why these programs are expanding exponentially.

(08:39):
Now, what does this mean for the individual mother and her journey through the health system during her pregnancy and her child’s infancy? Imagine Aruna in rural tribal India. She gets pregnant and the ASHA worker pays her a home visit. And when she’s getting registered into the database at the government of India through the ASHA, the ASHA informs her that there’s a program called Khilkari that will come to her weekly in her phone with important critical information of how to take care of herself and her child in that particular week.

(09:12):
She might get voice calls if she got a feature phone, but if she got a smartphone, she’ll get multimedia content, live action videos. She can even ask questions and answer some interactive quizzes. It’ll handhold her through pregnancy and infancy, so she’ll know exactly what are the risk factors, what are the danger signs, what are the complications, why she needs to access regular care, what medication to take, when to go and ask questions and how to prepare for delivery.

(09:40):
And it not only will give you her big picture information, but also handhold her emotionally. Tell her how the baby’s growing in her womb, when her child develops fingers. I mean, what is psychosocial stimulation of the child? When the child understands color? So it’s like a companion for her through pregnancy and infancy, answering all the questions she needs.

(10:01):
And, if for any reason she’s unable to get the voice calls, the ASHA would be informed about that because system will tell the ASHA, this woman is not getting the information weekly. So the ASHA will come and do some troubleshooting, so the information will come back to her.

(10:15):
Similarly, when the ASHA is trained on how to take care of the mother, at the same time, if the woman has a problem, a risk factor, and when she approaches the ASHA, the ASHA will actually know what to do in that situation, she will take the woman to the nurse.

(10:33):
Now, the nurse medical specialist is also trained now because they know exactly what is their role in each high risk factor. So the nurse will be able to do the right diagnostics, pick up the risk factor in time, know what she has to do, what medication to give at that particular point in time, if there’s a sudden emergency, when to refer, where to refer, and exactly what counseling to give. And when the woman is referred to either the specialist in the district hospital or the medical office in a primary health center, each of them would know they’re part of the role.

(11:01):
So that would mean that the Arunas risk factor is picked up early because she’s asked and recognized the risk factors. Because she recognizes the risk factor, she’ll be able to know and go and ask the right questions because the ASHA nurse, medical officer specialist know what they need to do. The risk factor will not become complicated. There’ll be no delayed referral. Treatment will be given to her at the right time, and this would mean that she would have a healthy outcome both for herself and her child.

Matthew Beighley (11:33):
What’s your vision for, you’ve done so much in the last five years since [inaudible 00:11:37], but what’s your vision for the next 15 years? What’s the world you want to see around you, in India or even beyond?

Dr. Aparna Hegde (11:44):
The world, I want to see is a world where every woman is empowered and every child is healthy. I want every woman to have access to information. That allows us to take the right decision for herself and her child. I want every health worker who is often female to be empowered, to have the knowledge, to be able to give the right treatments and the right counseling to the patients and the women that she visits and the children that she visits, and to be able to stand up to the community dynamics.

(12:14):
Women and health workers alike face multiple obstacles because the pervasive patriarchy. They do not have agency. They do not have decision-making power. The gatekeepers are the husband or the in-laws. Even though they are aware inside that a particular action they need to take is needed for themselves or their child, they can’t take that action because the husband won’t allow it, or the in-laws won’t allow it.

(12:44):
And that’s why information is empowerment, because then information comes in and validates what they already know inside. They’re able to kind of stand-up because there’s validation. They can actually throw off the yoke of patriarchy. I’ll give you an example.

(12:57):
There was this Azaan preacher’s wife who was pregnant and in the first pregnancy she didn’t get our program, and then she got pregnant again in the next three months, and she knew that this is not right for her body because there’s only three months gap between the two pregnancies. But her in-laws or husband did not agree to do something about it. But in that pregnancy, she got our program and the program spoke about family planning or spacing, that there should be three years gap between two pregnancies. She’s so convinced that she convinced a husband to undergo family planning.

(13:36):
So information is empowerment and it allows them to stand up to what they know is right. Their own sense of justice within them. It ignites that spark within each woman.

Introduction (00:02):
Welcome to Role Models for Change, a series of conversations with social entrepreneurs and other innovators working on the front lines of some of the world’s most pressing problems.

Ramesh Padmanabhan (00:12):
I’m Ramesh Padmanabhan. I’m the CEO of ARMMAN. I’ve been with ARMMAN for five years now. I am a corporate crossover from the corporate world into the social sector space. I’m basically a technology guy. Started my career as a programmer many years back and my last stint in corporate was as the CEO of a tech firm called NSEiT for about eight years. And that is when I decided to shift to the social sector to leverage my expertise and skills in technology and scaling organizations. And that’s what I’ve been doing in ARMMAN for the last five years. ARMMAN is a India-based nonprofit which works in maternal and child health. We work to reduce mortality and morbidity in mothers and children and our programs are running in 21 states as I speak today.

Matthew Beighley (01:09):
Why is there a need for ARMMAN? What’s the problem?

Ramesh Padmanabhan (01:12):
So the lack of focus on primary healthcare leads to a situation where when women go for seeking help, they basically don’t get the right kind of care in the institutions that they go to. There are two reasons for this that happens. One is the lack of access to preventive care information for the pregnant woman and not understanding the risks and the complications during the pregnancy well enough. So the woman is not able to take the decision and even when she takes the decision and she goes to a health facility, the care provided is inferior or not up to the mark for multiple reasons. The health worker is overworked, not motivated enough, not skilled enough. So these two lead to mortality and morbidity in women and children. And this also leads to a big problem because when there is a high-risk, which is not addressed during pregnancy, the child obviously is affected and the mother is also affected and it doesn’t really lead to the potential of the child as she would kind of be as she grows up.

Matthew Beighley (02:31):
And so as far as the solution is concerned, can you talk to me about why tech? Why was tech a integral part of your solution?

Ramesh Padmanabhan (02:37):
So when Dr. Aparna Hegde started ARMMAN 15 years back, she knew that whatever we built should scale and should go all over India and the mobile penetration was just happening at that point in time and the mobiles were being introduced across the world at that point in time. And she had this vision that at some point in time, every woman in the country will have a mobile phone in her hand and unless we are able to communicate what we want to communicate through a mobile phone, we will not reach every woman in India. So the whole idea is to kind of create programs which are digital so that they scale, they are non-linear, they don’t increase the cost. So it’s cost-effective as it scales the program and basically you’re able to kind of innovate on that. And for example, we have gone, we had voice calls earlier, now we have a WhatsApp channel that we have built for our programs. We are using AI to make our programs more effective. So artificial intelligence, machine learning. We have also built some large language models for training health workers.

Matthew Beighley (03:44):
Can you walk me through the experience from the mother’s perspective of what is she doing physically? Does she have a phone, is it an app?

Ramesh Padmanabhan (03:52):
Okay.

Matthew Beighley (03:52):
How frequently are the messages?

Ramesh Padmanabhan (03:54):
Okay.

Matthew Beighley (03:54):
Yeah.

Ramesh Padmanabhan (03:55):
Okay. So the programs, Kilkari is one of the largest ML programs in the world. Kilkari is a government of India program and we work with the government to implement it and it is implemented in, like I said, 20 states now. The program works as that, so like I said earlier, the problem was the lack of preventive care information that pregnant women have and understanding of high-risk and complications during their pregnancy. So what Kilkari does is it provides that information to the pregnant woman from the second trimester, so from the fourth month of her pregnancy till the ninth month or six months of her pregnancy, she gets information. Every week she gets one call and that call gives her information on health, nutrition, about the child, family planning, everything. So six months of the pregnancy gets everything around the pregnancy and once the child is born on the next 12 months, she gets information on how to take care of the child and herself.

(04:47):
So 72 voice calls go over 18 months, 18 into for 72. And these 72 messages actually are built by experts, validated by the government, validated by technical experts and shared. Initially we started with a voice service because that was the mode of communication then. Now with smartphones coming in, we have built a WhatsApp channel. So now there are two channels that we have. So there is a voice channel which sends the 72 voice calls and there is a WhatsApp channel which sends 72 WhatsApp messages. WhatsApp obviously gives you more flexibility to give richer information so we can give multimedia information. So there are videos that we will share. There is two-way communication possible. So we can send quizzes to the woman. If we send a message on iron tablets or breastfeeding, we can send a quiz saying that these are the three options, this is the message we send. How did you understand that? And she can choose one.

(05:47):
So we get a two-way communication channel. WhatsApp is very effective and WhatsApp is asynchronous. With the voice calling, the woman gets the voice call, but she has to be there to take the voice calls. If she’s not, then we try each call nine times. So there are nine iterations of the same call. If she doesn’t pick up in the nine calls, she can always call back and listen to the message also. But in WhatsApp what happens is once you deliver the message, it sits on her phone and she can listen to the message or would see the video at her convenience. It also gives the advantage of the family seeing the messages because once it’s on her phone, the husband, the mother, the mother-in-law, all of them can see the messages together and understand what’s happening during her pregnancy.

Matthew Beighley (06:30):
Mobile Academy, is that the other component? Do you want to,-

Ramesh Padmanabhan (06:32):
Yeah. Yeah.

Matthew Beighley (06:32):
Give me more information?

Ramesh Padmanabhan (06:35):
Yeah, so there are two sides of the challenge that we are trying to solve. Like I said, there is one side is informing the women, which is what I talked about just now. The other side is training the health workers. So Mobile Academy is a training program for ASHA workers. ASHA workers are the health workers in the villages, the last mile health workers. They have to be trained on what happens to the woman during the pregnancy and the child. So Mobile Academy is a program which trains the ASHA workers and gives them digital training. It’s a digital program where they call in, listen to the training content and it’s a four-hour content, but they don’t have to finish it in one shot. They can do it over time. So they listen to half an hour when they have free time. They can disconnect and the program will continue from where they stopped last time. So over time they finish four hours of content and there’s a quiz in the program. So if they pass the quiz, they get a certificate that they are certified for the program assets.

(07:34):
So basically between these two programs, any woman who is pregnant in the country in India should get the information through Kilkari and the ASHA workers get enabled through Mobile Academy. The whole idea of these two program is to inform the woman so that she’s empowered with the information and enable the health worker. So she’s enabled with the information to provide care.

Matthew Beighley (07:57):
Do you want to discuss, Aparna mentioned something that was interesting to me about the husbands being involved too.

Ramesh Padmanabhan (08:03):
Yeah.

Matthew Beighley (08:03):
Do you want to talk about that,-

Ramesh Padmanabhan (08:04):
Yeah. So we’ve realized that the information that we give has to be gender transformative and in the new content that we have developed both for the voice and the WhatsApp, we’ve created a drama kind of format in the whole thing. We have created characters, the wife, the husband, the mother-in-law, some community person who’s knowledgeable about all the information whom she can go to. The whole idea is that unless the husband also thinks that he’s going to become a father, the mother-in-law plays a role in the whole process. The pregnant woman doesn’t get the right attention. So the whole idea is to kind of transform the whole content in such a way that everybody gets involved as part of the process. So the content also plays along with those roles as it plays the thing. Yeah.

Matthew Beighley (08:54):
Are good with numbers right now. Do you have a sense of the reach and how many women and how many states you’ve expanded to?

Ramesh Padmanabhan (09:00):
Yeah, so currently ARMMAN operates in 21 states in India and in the next two years we should be PAN India with Kilkari and Mobile Academy. So far we have reached 47 million pregnant women and their children and we have reached about 367,000 health workers. In the next seven years by 2030 or next six years, by 2030 we will reach 70 million women, mothers and their children and 850,000 health workers. So I think all our programs, like I said earlier, are digital. So they scale, but it is non-linear. And when we do the programs, ARMMAN as a team actually builds the technology, builds the content, but the implementation of the program happens through the government systems. So the government health workers, the government administrative functions, all of them manage the program. So that’s how it scales PAN India.

Matthew Beighley (09:57):
So objectively it is just growing very fast. Anyone would say that and my question to you is why? Why is this working so well?

Ramesh Padmanabhan (10:05):
So it is, I think it is an information. Both Kilkari and Mobile Academy are two things that should have been there in the system and they’re there in the system and they’ve not gone PAN India for various reasons so far. And it is important that every woman in India gets the information and every health worker gets the training that we’re providing through Mobile Academy. The reason for it growing fast is because we have a very good relationship with the government. We share plans with the government and obviously the government also prioritizes what they want to do. So our relationship with the government is very strong. We build plans for the next two, three, five years in terms of what we are going to implement, how we are going to implement. Obviously the state have to get the state buy-in, and then when we work with the government so closely, it’s very easy to implement the programs on the ground because you get complete state buy-in terms of implementing the programs and it’s easier to implement also because it’s digital.

(11:04):
So when you inform the state about what’s going and you communicate and train the state government officials in terms of how the program is going to implement it, the implementation happens fast because there is a maternal and child health system database in the country called the RCH database. And every state has its own RCH database. RCH is a reproductive and child health database. So every state has a RCH database which sends information to a central database. So there’s a central RCH. We get information from there and we get the information on the mobile number, the last menstrual period, the date of birth, and all of the information that’s required for the Kilkari program. So launching the program, once we get buy-in from the governments, we have done the training, it’s very, very quick because it’s very digital. So once we get all this information, we can start the calls immediately.

(12:00):
And so for example, if a state decides to go live in the next month, then maybe four weeks before we’ll prepare for that, we’ll go to the state, we’ll tell them what’s going to happen, send the information across the state in terms of Kilkari being launched. And then the database of the state syncs with the central database and we start getting information from the central database and we can start sending the calls in four weeks. So that’s how it gets done very fast because it’s very digital, easy to launch, but the touch part is important. All the programs have a tech plus touch approach. Like I said earlier, the technology is the piece, but the touch is the health worker on the ground. And so it’s very important to kind of inform the state and inform the administration and inform the ASHA worker that program is going to be launched. And so that, what does her role in the whole process with Mobile Academy and Kilkari?

Matthew Beighley (12:51):
Let’s touch on, tech plus touch, like what is that and why is that so important?

Ramesh Padmanabhan (12:55):
So like I said, when Aparna started ARMMAN 15 years ago, one of the key things was we should be digital, we should be health based so that it reaches every part of India. But we were also aware that technology alone will not work. Just sending calls to women is not enough because there are lots of information available on the net that people can see also, which is not valid content. Right. So it was very important to give her a touch point where she could go. So registration is a touch point, but even during her pregnancy and the 12 months of the child, she should have a place to go to in case she has a question, she has a query, she can call. So the touch is either a person, in most cases in India where Kilkari is launched, the ASHA worker becomes a touch point and she’s the touch, so we kind of ensure that the touch, which is the health worker on the ground, understands the program, is able to help the pregnant woman and the mother of the child on the ground.

(13:54):
The touch could also be virtual. Like for example, when we launch the WhatsApp program and when we do high-risk pregnancy programs in Kilkari, we will provide a call center or some handholding for the mother. And that could be a virtual touch where it’ll be a toll-free number that she can call and get help. It’s very important that while we share a lot of information, every woman will not understand the information that we share the way we want them to, and so that they have a point to go back to and understand it better if they want to. And that is the touch point. And the tech plus touch is a blended tech plus touch.

(14:26):
I’ll elaborate on that a little further. The whole idea is that you’re able to tweak the technology and tweak the touch based on the type of community you’re serving. So India has different populations living in different regions, and there are a lot of places where there are problems of reach, there are caste problems, there are religion problems. Being the program may not reach as effectively as we want to.

(14:54):
So in all those regions, so to go to the most vulnerable and marginalized communities, we have to kind of understand why we want to send a voice or a WhatsApp message. What is the most effective way it will reach the last woman on the ground? What is the best touch on the ground? Like I said, ASHA worker will work in most places, but there’ll be very remote areas where you may need self-help groups or you’ll have mother groups or you have other community groups, or you’ll have religious groups, which could be the touch point for the pregnant woman and the mother. So it’s very important to understand what is the technology, what is the touch which will work for that particular situation, blend the two together so that the program is the most effective on the ground.

Matthew Beighley (15:36):
Kind of two last questions. [inaudible 00:15:39].

Ramesh Padmanabhan (15:39):
I want to talk on technology a bit, artificial intelligence, if that’s,-

Matthew Beighley (15:42):
Yeah, let’s do that now.

Ramesh Padmanabhan (15:43):
Okay.

Matthew Beighley (15:44):
Yeah. Do you want to just highlight the importance of AI?

Ramesh Padmanabhan (15:46):
Yeah. So like I said, all our programs are mHealth interventions. We have a lot of data and over period we have realized that we can use this data to really make our programs more effective. And we’ve started working with Google Research and Harvard about three, four years back to embed AI in the programs that we are doing and to improve the effectiveness of the program. So I’ll give you one example of what we’ve done with Google. The effectiveness of the program is largely dependent on the woman listening and the woman’s engagement in the program. So what we did was we took the listening patterns of woman for a particular year in 2018 data we took and we gave it to Google, obviously anonymized, taking care of data, security, privacy and all that. But we ran a machine learning model on that to understand how the woman listening pattern is.

(16:41):
This is important because if the woman drops out of the program, many women, some women drop out of the program as the program progresses. So it’s very important to kind of intervene with them and say that, okay, don’t drop out of the program. It’s a very important program you should listen to. So how do you predict a woman being dropped out? So it’s a proactive thing because after she’s dropped out, it’s very difficult to bring her back into the program. So how do we predict that a woman is a potential dropout? That we are doing using a machine learning model using artificial intelligence. So the model has been trained over time. It kind of throws up alerts in terms of in a particular geography, which are the women who are potentially going to drop out. And the health worker, that’s the touch that I talked about earlier. The touch person on the ground can intervene with that pregnant woman and say that, okay, you look like you’re going to be a potential dropout.

(17:32):
Explain the importance of the program. Explain why she has to listen to the program, explain the benefits of the program, nudge her so that she continues in the program. So that is one area where we use machine learning, artificial intelligence to do predictive analytics and predict the dropouts from the program. There’s a second area where we have used artificial intelligence is building large language models. So we also train health workers in a high-risk program that we are running in India. And the high-risk program has a content that Aparna and a set of team have built protocols for high-risk management for nurses, medical offices and specialists. So the nurses get trained on that and then they use it on the ground, but they may have queries and we had built a WhatsApp support hand-holding for them. So they could have a WhatsApp support where any query that they have, they can put it on WhatsApp. It’ll go to their supervisor and the supervisor will respond.

(18:33):
Now with large language models, what we are doing is we are building a large language model where we have taken our content and build a large language model. So when the nurse puts a query on WhatsApp, it goes to our Copilot that we built on large language model first, and that large language model kind of will build the answer. And if the answer is good, obviously we can directly send the response to the nurse and it’ll not go outside the domain of what we’re, because it doesn’t kind of hallucinate, is a technical word that’s used a lot nowadays. So the hallucination is not there because we restrict it to the domain that we have taught her. And if the response is not good, then a human comes into the loop and is able to intervene and give the answer. So it kind of obviously increases the effectiveness of the program multifold because many answers are answered by the Copilot and there are very few which will get intervened by the human and the human will give the answer.

(19:37):
So there are two areas where and artificial intelligence, something that we are going to use across all our programs, be it predictive analytics, be it large language models to make our programs more effective as we kind of progress. So the vision of ARMMAN is to reach every pregnant woman and child in the country with information that provides preventive care information to the pregnant woman and the mother of the child. The way we are structuring the programs or the interventions is in such a way that it’s a fit-for-purpose approach where every woman gets the information that she needs and the intervention that she needs. I’ll explain this in three layers that we are building out all our programs. The bottom-most layer is what we call the low-risk layer, the LOW. So every pregnant woman, according to us, is a low-risk. There’s nothing like no risk. And so the low-risk woman, we have to provide information to her on preventive care and high-risk and we have to train the health workers there.

(20:39):
So Kilkari and Mobile Academy are the two programs which work on the two sides that I talked about. So Kilkari provides the information, Mobile Academy trains the health workers. When you go one level above, we have the high-risk women and the high-risk children. So we have a program for training health workers on how to manage high-risk, which is called IHRPTM, Integrated High-Risk Pregnancy Tracking and Management. And on Kilkari we are building content which is for high-risk. Suppose you have a high-risk like anemia or hypertension or gestational diabetes. Once that is identified, you will get the basic Kilkari information and you’ll get a Kilkari plus, which will be the high-risk information for that woman. At the same time, the health worker on the other side is trained using IHRPTM, Integrated High-Risk Pregnancy Tracking and Management, which trains the health worker on how to manage the risk, what is the protocol, when should she refer it so that that is more effective.

(21:38):
Then also children could have a risk. So malnutrition is one of the biggest risks in India. So we have a program for children called Swasth Kadam, which means healthy steps in English, which basically, when a child is underweight, now the general identification of a child happens in malnourishment, happens after the child has become a severely acutely malnourished child. What we do is we identify the child earlier if she’s underweight and if the child is identified as underweight, we start intervening. And we have done a research on that to kind of see that, how this program works. So once the child is underweight, we do two things. We start counseling for the mother of the child, and we also start sending information on WhatsApp to the mother of the child, both these helping her to take care of the child. And in the studies that we have done so far, we’ve realized that the number of children who reduced [inaudible 00:22:35] has reduced and the number of children who have improved their health has increased over time in the control group and an intervention group where we did a study.

(22:43):
So I’ll repeat what I’m saying. So there is a low-risk, there is a high-risk for pregnant women and there’s a high-risk for children. And there is another risk which is very relevant in India, which is called, we call the equity risk. So there are lots of women who get left out of program because of caste, because of religion, because of gender, because of geography, because they migrate. So there are lots of such populations which kind of do not get the full benefit of public health programs because of one of these encumbrances that they would have. We treat that also a risk and we say that how do we make our programs more effective in those areas? And we say, what is the kind of technology that will help us reach them? What is the content, kind of content? Because the content could change, the dialects may change. You may have to maybe break down the content for this particular set of women.

(23:34):
So what is the kind of content that will work? What is the kind of public health system? What is the kind of support you can give on the ground? So I think equity is also a risk that we identify and that we can design our program for that. So to put everything together, there is a fit for purpose approach, which is, like I said, it’s digital, it is mHealth based, and it basically has interventions for all pregnant women and all ASHA workers, high-risk pregnant women, high-risk children for health workers who handle high-risk and also for equity in encumbered populations. That’s our model.

Matthew Beighley (24:11):
Where would you like to see ARMMAN in 25 years?

Ramesh Padmanabhan (24:15):
So in 25 years, I think we obviously would have reached every pregnant woman, every child in the country. We are very painfully aware that we will not reach 100%. It’s quite possible that will happen. We may be at 99%, hopefully at that point in time and providing the information and the health worker is trained to do that. And all these programs are government programs. It goes PAN India in 25 years. And what our role at that point in time I would see is helping the government or working with the government in innovating the programs on a continual basis. While we implement everything through the government systems we think in 25 years, all pregnant women, all children, all high-risk women, all high-risk children, all equity encumbered women and children get the programs that we are trying to kind of build now. And it goes across India in the next 25 years. We basically are able to provide this information to other countries.

(25:18):
Basically the high-risk management protocols or any program that design the concept, the ideas, we’re able to kind of take it to other countries and share it with them. We are able to advocate on some of these areas that we speak about. And we continue to kind of innovate and build programs for pregnant women and children to make it better and better using technology, using artificial intelligence. And obviously there’ll be more tools that will come in the next 25 years.