This interview originally appeared in the “Women of Public Health” interview series published by the John Hopkins Global mHealth Initiative. Yorghos Carabas, Global mHealth Initiative Program Coordinator at Johns Hopkins Bloomberg School of Public Health, conducted the interview.

Screen Shot 2015-12-02 at 1.18.45 PM


Yorghos: Let’s get to know both of you, how did you, with your different backgrounds end up converging in the world of mhealth, and ultimately, at Dimagi? We’ll stick to alphabetical order throughout as best we can, so Rowena, you jumped right into digital health while you were at Intel and then founded your own telemedicine non-profit named AMITA, what was that path like and how did it prepare you for what you’re doing now?

Rowena: I got my start in engineering. After working on a few projects, I realized I needed to understand the ‘why’ of what I was making, not just the ‘what’ that others told me to do. So, during my Masters, I joined the Technology and Infrastructure for Emerging Regions (TIER) research group at the University of California, Berkeley which is what gave me my real first taste of the mHealth space. What I really appreciated about this opportunity was that I was able to get a more research-oriented, non-partisan perspective on ICT (information communication technologies); explore the lay of the land and understand the state of research at the time.

At my core, though, I’m an implementer, not a researcher. I ended up leaving to try to continue the work from a different angle – so I set up an NGO in Canada focused on telemedicine systems in West Africa. This second phase helped me learn about how to operate a nonprofit, fundraising, administration, partnership building, etc. It gave me a better understanding of the sector overall. I also learned and worked through the challenges of developing technology within an NGO, which has its unique problems compared to the private or social enterprise sector. While I was working on a grant, a colleague of mine from Berkeley contacted me and told me to check out Dimagi, as they were doing exactly what I was dreaming up and had already formed a team passionate about the same cause. I was psyched to sign up.

Yorghos: Stella, you studied both law and computer science, and started at IBM before moving on to the Canadian Department of Criminal Justice and then as a Legal Advisor for African and Middle Eastern Refugees before arriving at Dimagi. An interesting route to be sure, what inspired you to work in digital health?

Stella: I’ve always been interested in technology for development. So I studied computer science and spoke to a CS professor at some point about how to combine my interest in international development with software development and he told me that I couldn’t combine those things and that I had to choose. So I chose international development, and I ended up studying international relations and got a law degree en route. I spent a lot of time working in various aspects of good governance and human rights, including supporting refugees fleeing their countries of origin. Rowena kept telling me about this great organization Dimagi and all the amazing things they were doing, but the real turning point was when I was offered a chance to be a Field Manager in West Africa and it sounded too good to be true – but it wasn’t. So I joined Dimagi. Of course the transition from law into mHealth is a big jump. I can’t remember a time when I learned more at such a fast pace than when I started to dive into the mHealth world.

Yorghos: Digital health implementations face various challenges, scaling is a common roadblock, as is sustainability – what are some methods you’ve used to overcome and achieve success in these areas?

Rowena: That’s a big question and we could write a book on the answers. From my perspective on the technology side, a key part of that equation is ensuring the long-term cost effectiveness, sustainability, and maintenance of the tools we deploy. At Dimagi, we now focus on building a technology product that can leverage economies of scale. We strive to have every user of CommCare using the same tool, managed on the same server, so that when a feature is built for one deployment it’s by default also able to benefit hundreds of others.

Coming from the era where Dimagi built a lot of our own custom software, I’m still getting requests for server maintenance from random countries even though I stopped coding years ago. I think the move to a community-sustained product is vitally important to ensure that you’re not always having to reinvest in the technology and infrastructure and reinvent the wheel. Long-term financial sustainability of these software systems is a hugely important considerations, and most projects could look more seriously at the ratio of cost to implement their initiatives versus the impact it will have for the governments that are who need to maintain it.

Stella: Aid projects have been around for decades, but there are still a lot of pretty basic things that haven’t been figured out (e.g. who is going to take up this project when the NGO leaves? Is there a government champion to help incorporate this change into the government system?), but this bad track record is no excuse. We need to think through the long-term adoption and sustainability plan – what happens when the pilot is over – and plan to make it happen, from early on. Beyond that, in terms of scaling, I think a lot of people are realizing that multiple government mechanisms need to be in place for a project to grow and be sustained. Even if imperfect, a locallyowned, but still running system is better than one that dissolves at the end of the project.

Rowena: It’s less about building brand new infrastructure in a country and more about reinforcing existing capacity. It’s about identifying who will be part of the long-term story of an intervention and how do we engage those people early on so it’s not just some western organization jumping in. One of the big focus of our tools is giving local users and administrators the ability to modify the tools to fit their particular needs. Whether you’re building a project to last 5 years or 500 years, you’re going to have changes. Ideally governments can own these tools and change them themselves.

Stella: In the Indian context, what’s really interesting is that there’s so much capacity, ambition and intelligence already here, as well as many local initiatives to change and improve things, so it becomes a matter of figuring out how to contribute our specific area of expertise and technology and identifying the best way to contribute in a meaningful and respectful way.

Yorghos: Health systems are comprised of myriad moving pieces, including human components like front-line health workers and clinicians; how do you build for accountability, and consequently enforce or incentivize participation?

Rowena: To incentivize participation, make it useful. One of the issues with the industry is that you have the people you’re serving and the people who are paying for it and so many systems are built for the funders and not for the communities they’re meant to serve. They might end up using it because they have no choice, but the better model is empowering the people who it’s designed for. We prefer to create tools that help people do their jobs better and more easily with decision support. Just as important is connecting those frontline workers with their immediate supervisory level so that health workers stop feeling and acting like they’re alone at the last mile, but rather part of a cohesive and functional health system.

With regards to your question on accountability, let’s take a look at where we were 20 years ago, when health workers would be trained and left to do the work in the field alone and asked to report on their own performance once a month, by memory. Of course the numbers were off and the accountability was lacking. Now we can collect metadata automatically from digital tools they want to use every day to know where a person is located when a form is filled out and if the collection times align. One of the most interesting and powerful potential wins about mHealth is that even if the application itself doesn’t do anything, just measuring community health worker participation and activity accurately is a huge step forward for the sector. Just having better data on which CHW programs are delivering their services more consistently unleashes the power of every health worker NGO to identify failures and iterate/improve faster than ever before.

Stella: One thing that I warn partners who start in mHealth is that it does provide higher visibility and greater access to more data, which can identify deficient areas. Real data with some deficiencies may end up looking bad compared to all the beautiful reports out there full of fake data. So while real-time digital data collection has tonnes of advantages, transparency comes with its own set of risks.

Yorghos: A tool is only as good as it is useful, how do you handle usability and UX testing? What do you look for when engaging users, be they patients or health care workers?

Rowena: Usability is very important to me, and was the primary focus of my graduate education. Yet, the biggest question I’ve needed to ask, which is a bit bigger than ‘usability’ itself, is: are we solving a problem that the user cares enough about?

The people that we work with have such a hard life and so many transportation and communication issues. But if they care about something, they’ll make it work (as demonstrated by how hard they’ll work to share music and movies with each other). Usability in the West is more concerned with identifying or even creating a problem and coming up with a user model and creating demand to address it; usability here is sifting through the many well-known problems being faced and selecting the one that needs to be prioritized first, that matches with the threadbare resources available to deal with it.

Yorghos: Engaging on a scale like Dimagi does, you both likely encounter and work with various portions of various supply chains, how do you handle commodity stockouts and breaks in the chain? Have modern mobile data tools alleviated any issues previously faced?

Rowena: What really surprises me about supply chain, which is my primary focus area, is how much less you hear about mHealth supply chain work than you do behavior change or monitoring and evaluation. Supply chain and mobile health are natural partners: it’s a problem of data, and calculations, and coordination. It’s so common for someone in a facility to need medication, and send out a request, then it just disappears and there’s no tracking method and no way to know what’s happening and if that medication is coming. Yet mobile phones and digital technology are so well suited to this challenge.

One great example is our work with a community based SMS supply chain tool called cStock which we developed in Malawi in partnership with Ministry of Health and JSI. Based on simple reports of stock on hand and receipts, the system would auto-calculate reorder timing and quantity, trigger resupply, and let the health worker know exactly when the shipment was ready for pickup. 99% of its users preferred the SMS system to a paper system because it saved them from wasting hours, sometimes days, going to the nearest health facility for stock only to find that the supplies they needed weren’t available. Such travel is a complete waste of their precious time – and completely avoidable. cStock today is used to handle resupply for all 3,800 community health workers in the country.

Yorghos: Culture, both country-wide and regional can impact the efficacy of a digital health implementation; Rowena, I saw a piece you wrote about women in Northern Ghana and the difficulty revolving around phone-sharing and household phone ownership, can you tell us a bit about gender challenges?

Stella: It’s an interesting question, gender issues are everywhere and how they play out in the real world varies from one context to another. mHealth is uniquely positioned because it’s such a young field that there aren’t many experts which means there isn’t that traditional gender barrier which I remember quite well from my days in the legal world. Being in this emerging space allowed me to take on leadership roles from country to country. I can walk into a board meeting and be evaluated based on my experience and contribution in this field at the same level as male peers with decades more of other experience.

Rowena: I’ll talk about two points, first in terms of the communities that we serve, second to those women interested in doing work in this space. An experience that sticks out in my mind is one of our deployments in western Afghanistan with World Vision a few years back where we were standing up a community health worker program.

A lot of things are different there of course, but the specific things I remember was sitting down with an 80 year old grandmother who had never touched a phone in her life and had a very basic level of literacy – giving her a phone, her very first one, and thinking “there’s no way this is going to work.” But seeing her energy and her excitement and how determined she was… she wouldn’t go home until she figured out how to use the program. It was particularly meaningful because her husband, son, and grandson, who was 7 years old all had phones, but she had never been allowed to touch or handle one. So it was one of those things where you have this glimpse into this vast space of untapped human potential. It was a very tricky situation at first, but that group of women ended up being some of our very best users, and still to this day they’re using the application even though our visits wrapped up long ago.

On a completely different side of the gender equation: in the development sector in general – not related specifically to mHealth – a lot of us find ourselves working in very tough environments and working with challenging partners. Women can sometimes find themselves harassed, diminished, or held back because of their gender, and sometimes it feels like that’s just the way this world runs and you have to deal with it if you want to work in such environments. To those reading this article: if this ever happens to you, speak up. The right for women to be safe from harassment and discrimination is a core part of making the world a better place – we have to start by ensuring this truth within the small teams where we work before we can realistically go after the big task of changing countries and governments.

Yorghos: Stella, you’re stationed in the field (India currently) I believe, what experiences can you share regarding cultural or gender obstacles?

Stella: With my team here in India, I didn’t quite realize this when I came here originally, but I do now: we hire these passionate, intelligent and qualified people, many female, and it is very unusual for a single Indian female in her 20s to travel all over rural India on her own the way our staff do, in order to train and implement programs and help NGO teams design and deploy mHealth systems. Though the battle for equality is complex and waged on many fronts, I’d like to believe we are a part of it, every time one of these women firebrands shows a community healthworker (also female) how to use a phone for the first time and also every time I walk into a room full of businessmen, whether it’s in India or anywhere in the world.

Yorghos: What projects are you both working on right now? What stage are you at and what direction are you heading in?

Rowena: Over the next two years, we’ll be deploying a tablet-based version of CommCare to 400 clinics in Burkina Faso with our partner Terre Des Hommes. It’s a decision-support tool to help nurses diagnose children under five for the five greatest killers of children under five: pneumonia, malaria, malnutrition, etc. It’s replacing a 40 page paper protocol, which of course, nobody in practice actually uses because it’s 40 pages… and unwieldy. We spent a tonne of time to ensure the usability of our digital version of this protocol, to make it fast and efficient, to automatically calculate medicine dosage and capture drug interactions, and to provide a tonne of visual feedback that makes sense to the nurses. They go through so many patients in a day and they need to be able to diagnose correctly and quickly, provide the right medication, take measurements and record data. It’s easy to get something wrong when you’re seeing hundreds of children a day; this tool will help them to get it right each time.

One other project is the “Informed Push Model” in Senegal with Ministry of Health and IntraHealth International, which is a government partnership geared towards changing the way that family planning commodities are supplied to the country. So instead of each facility trying to figure out for itself how much of any product they need to order, they’ve trained a small but dedicated team of mobile logisticians to manage hundreds of facilities each, equipped with tablets running CommCare Supply. These tablets contain the history of each facility and can assess the stock, calculate consumption, forecast need, and inform reorder amounts on the spot, before the logistician moves on to the next facility to continue the process. It’s really exciting because we’re able to scale this program across not just family planning, but also malaria and other programs, which is critical for the sustainability of such implementations. Today we’re supporting 1,400 facilities across the country.

Stella: The project I’ve been working on is with the central government in India to improve nutrition by tracking and monitoring malnourished children, which is a huge problem right now. There’s more children in India than any other country in the world and about 40% of children under five are underweight. So the numbers are enormous. We’ll be supporting community nutrition workers across eight states across India to help make this happen. To my knowledge this is the largest mHealth project that has ever happened, so you can imagine, much like India itself, it’s orders of magnitude more complicated than anything else I’ve ever worked on. Really, the technology is the easiest part of this, managing all the people and processes is the real challenge, the logistics alone are a massive undertaking.

Yorghos: The tech in health industry is evolving rapidly, what would you like to see happen in the near future that would benefit the work you do and maybe health systems in general?

Stella: One thing I’d like to see is as we transition from this pilot world is that all the different players really need to start working together more, collaborating and combining resources more, and sharing lessons learned. There’s a lot of organizations out there who need to move away from “here’s my app” “here’s my app” “oh and here’s my app.” We need to promote best practices and learn from each other, even if a project doesn’t work out, it’s good to share that experience in good faith so we can move forward improving health systems.

Rowena: I’ll take a different perspective, yes influenced by my technology background. . . collaboration is essential, but what’s really interesting that’s happened in the last two years is the drastic drop in the price of android phones, coupled with the end of Nokia-before-Microsoft, and the beginning of a much more open market for pairing apps with appropriate hardware that can be fitted to the needs of any given deployment. Mobile innovators today are much less constrained by the platform than they were two years ago, and the effort to get these apps running on a huge variety of devices has dropped dramatically. Reducing such barriers to entry opens up a floodgate of possibilities. It’s exciting to see this change and to see more and more tiers of people able to communicate, connect, and collaborate with each other even in the toughest environments in the world.

Stella: Another trend: many governments have done a good job of mandating open access to data. Once systems are able to talk to each other, it enables an ecosystem of care records and provider information, wherein one organization can tackled the problem of medication adherence while another goes after defaulters and a third is analyzing big data, and all this energy can actually be complementary instead of duplicative. These policy level interventions make a real impact on the real-world possibility of how we can run health systems better.

Yorghos: As always, inquiring minds want to know – what advice would you give to someone interested in working in digital health?

Rowena: Mine is one of my favorite phrase from working here at Dimagi, “Just figure it out.” When Stella and I started in this space, there was no way to sign up for a course or to train in mHealth; now there’s a lot more resources out there, which is great to see and I’m happy it’s getting better day by day. But the landscape is changing constantly. This industry is only 10 years old and working out its awkward adolescence. We’re only now just scratching the surface of how to run national-scale systems wherein all health workers have the right information tools they need, and how those tools can amplify and interact with each other. It’s changing so rapidly, and it will continue to change… so don’t wait for someone else to figure it out and tell you what to do. You need to have the energy and the passion to talk to people, understand what’s going on, identify the gaps and be part of the answer. Don’t wait for someone to tell you what to do; just figure it out.

Stella: Keep your horizons open. I don’t actually think the area of mHealth is the be all and end all in the general sense, but I do think it is arguably the most exciting area at the current time for those interested in making people’s lives better and in radically transforming how that’s even done. The opportunity for step change impact, as this kind of innovation scales, is enormous, but hopefully in 10 or 20 years, mHealth will just become “how it’s done”. The more general principle is that it’s good to find out what you believe is the game changing driver of progress now and figure out how to help make that happen. (To misappropriate a Sheryl Sandberg quote) When you see a rocketship, get on it.