The M-Pesa of Healthcare
As I write this, less than 7% of the Kenyan population is vaccinated.
In my 9 years living in Nairobi I have been blown away by the energy of the city. Healthcare associations and communities exchange hundreds of messages each day over whatsapp. Start-ups are “disrupting” everything from emergency care infrastructure to the supply chain for oral rehydration salts.
Why hasn’t the booming health innovation scene translated into higher vaccination rates?
I moved to Kenya in 2012 to launch a healthcare business called Access Afya. We currently serve over 10,000 people every month through community micro-clinics and a suite of digital health apps. Access Afya is a digital health business at its core that reaches people through lightweight clinics, a tireless focus on distributing the right health products at the right time and extensive community outreach.
As continued waves of coronavirus and it’s variants shake up how we work, shop and talk to doctors, everyone is making predictions on what this mysterious “new normal” will look like. I’ve got a clear picture in my mind of a distributed, connected healthcare system – one that ensures access for everyone.
I think this future of health could look a bit like our adjacent fintech scene.
M-Pesa, and what it takes to Leapfrog
M-Pesa is just one story of how an innovation can change the status quo – but it is an important story because it illustrates essential ingredients necessary to “leapfrog” or replace incumbent infrastructure beyond simply having good technology.
M-Pesa is a powerful financial access innovation. I still remember being amazed I could text my friends airtime during my first trip to Kenya in 2007. As people became familiar with texting airtime, that turned into sending shillings (Kenya’s currency) to a friend’s phone number. Today, M-Pesa is arguably one of the most successful financial inclusion programs in the world, with around 42 million users in Kenya (there are 52 million Kenyans) compared to around ⅓ of the country having a formal bank account.
Long before covid, my start-up tested fully cashless clinics using M-Pesa (1)
How can we then achieve the same viral success in the healthcare sector?
Investors, commentators, project teams at foundations say: “Digital health can leapfrog clinics like how M-Pesa leapfrogged banks.”
The obvious comparison here is telemedicine. In 1879, just three years after the first U.S. patent was awarded to Alexander Graham Bell for the invention of the telephone, The Lancet journal described how this new technology could be used to reduce unnecessary doctor visits (2). It’s widespread use is only just taking off.
Do we need telemedicine? Absolutely.
Does telemedicine “leapfrog” all of the “space, staff, stuff and systems” required in building functional healthcare systems (3)?
Not quite.
Did we leapfrog banks with phones? Only kind of. And focusing on that convenient narrative only distracts us from finding the M-Pesa of ______ (insert next industry here).
There are three features that have made M-Pesa wildly successful in Kenya, which provide more nuanced guidance for those looking to provide disruptive access to basic services.
We leapfrogged banks not with phones, but with a community of 45,000 strategically placed agents inside of existing shops where people were already going, paired with a phone-based money store and transfer mechanism. This gave people a place to go to learn about the service, and physically interact with their money whenever they wanted to.
Large investments in infrastructure, people and incentive design went into making M-Pesa work, which are the real reasons why we can interact with the “app alone” experience today. We would have never had mass adoption otherwise. Mass use trickled up, not down.
Banks are now on the M-Pesa platform offering value-add services that M-Pesa can’t or won’t do. (e.g. interest bearing savings accounts)
Using these principles, what does the M-Pesa of healthcare then look like?
(Disclaimer: a lot has been written on M-Pesa (4)-- I will only focus on the three areas listed above, which are also broadly applicable to any story of making technology take off in emerging markets. I’ve provided a few follow-up links in the notes for those looking at a broader literature review of M-Pesa and how it works.)
Distribution
The collection and distribution of cash was the real key to the success of M-Pesa. M-Pesa doesn’t just let you interact with it’s USSD/ app, it lets you give and receive real money to community agents. This was essential because as the technology was introduced, it was easy for people to approach agents and interact with their cash whenever they wanted or needed to.
An M-Pesa agent amongst small shops
Banks were not reaching the majority of Kenyans because they didn’t want to invest in distribution. Safaricom, the primary telco in Kenya, already had. They were distributing airtime all across the country. (In Kenya, it is quite common to buy a physical scratch card to top up your phone airtime). At the time, a telco wasn’t the logical disrupter of the banking sector. But Safaricom could invest on top of its airtime distribution channels to quickly bring new products to market. This was essential to Safaricom’s strategy when launching M-Pesa, which was initially focused on rural-to-urban remittances. Their initial use case required them to get into last-mile distribution. Later, they’ve added on countless uses including salary payments, quick loans and community fundraising. At one point, they had over 400 trucks criss-crossing the country, running their distribution network.
Through distribution, M-Pesa got closer to the customer, and built a brand that people learned to trust.
Hospitals are the banks of healthcare: giant infrastructure that everyone has to trek to. Hospitals are too expensive to bring to everyone. We will never get to a scenario of universal healthcare by relying on the expansion of hospital networks.
M-Pesa for health organizes supply and demand by decentralizing the hospital.
To disrupt the distribution of healthcare we need a lightweight network of at least 10,000 healthcare agents similar to those dealing with our money on the M-Pesa network. A lightweight network is a flexible network: an M-Pesa agent is usually also selling phones or bread. A healthcare agent could provide services, rapid and digital diagnostics, medications, hygiene supplies, counseling and any other products and services that help communities live healthier lives. Their services could be tailored to evolving health needs and those specific to their communities.
There’s still a lot of “stuff” required to do healthcare. Even with teleconsultations, we still need people to have access to rapid diagnostic tests and labs, medications, first aid supplies, etc. etc. etc. And a lot of the “stuff” has some stipulation around how it is currently stored, shipped and even used. This adds complexity for the M-Pesa of healthcare.
To solve for “staff” and “stuff”, lightweight healthcare agents and patients themselves ultimately need to be equipped with comprehensive and easy to use digital health systems that monitor stock levels, temperature, symptoms, protocols and much more.
M-Pesa trickled up
Some things do trickle down (Tesla cars, for example) but this isn’t necessarily what we need to do with quality healthcare. M-Pesa went to the mass market first instead of waiting for the costs to go down with scale. Early adopters weren’t particularly tech-savvy or wealthy, they were just people with a really common problem, who were currently faced with terrible options to solve that problem. (This was the movement of money from family working in cities to their rural families.) Safaricom did not try and make too much money upfront. They invested a lot in building a great product, and kept pricing at a point that practically everyone could afford.
Neighbors and friends were the agents. Digital money could be instantly turned into physical money. Trust was built through community integration, but also by having a service that almost always worked.
M-Pesa as a platform integrating with the ecosystem
M-Pesa hasn’t replaced banks in Kenya. In fact, it’s opened up totally new markets for banks by letting them integrate and serve M-Pesa users. What M-Pesa has replaced is the default of starting and/or centering your financial life around a bank.
We shouldn’t want to replace hospitals. We need them.
What we need to do if we want to live in a world where everyone has access to decent healthcare is disrupt the paradigm of centering health around hospitals and large clinics. The M-Pesa of healthcare will enable hospitals to be more efficient. They will serve the right people – those whose conditions really need hospital level equipment and specialists.
An Access Afya Clinic amongst small shops
The M-Pesa of Healthcare
When I think about the future of healthcare, I do not see myself interacting solely with a phone, but rather a network of nimble healthcare agents with the tools they need to help people. This lightweight and flexible healthcare infrastructure will constantly adapt to the shifting needs of a society: today, a place for vaccine distribution, tonight physiotherapy sessions, Sundays, eye examinations and glasses fittings. M-Pesa is simply a platform that facilitates a distributed network, which is exactly what is needed for healthcare.
This was what I kept in mind when building Access Afya. Using the building blocks of distribution, trust and tech we integrate:
A consumer facing tech application: mDaktari. This gives people the option of talking to a trained healthcare professional using USSD to request a call back, an app for voice/ video chat, questions and support groups and mDaktari+ for scheduled video consultations with specialists. This lets us reach a variety of users from simple to smart phones, monitoring their symptoms, medical conditions and outcomes.
A complimentary tech product monitoring quality. M-Pesa (for the most part) doesn’t care if you want 100 or 100,000 shillings. In healthcare it’s a red flag if you want 100,000 painkillers. Our AfyaSmart product is monitoring data coming in from digital diagnostics, sales transactions, text messages, security cameras and more to make sure evidence-based healthcare is being delivered.
A micro-clinic model, where pharmacy, diagnostics, consultations, chronic condition management, ultrasound, tele-connections to specialists and more all happen in a lean space the size of 1.5 shipping containers at a fraction of the set-up cost of larger clinics or hospitals.
A franchise platform, so entrepreneurial community members can run our healthcare model, tapping into mDaktari, our distribution and quality assurance software, our online training academy, partnerships with other healthcare players and more.
A strategy to build trust. Our outreach, which is primarily educational or fun events, drives people to talk to healthcare agents even when they don’t feel sick. Building this trust and education is essential: 80% of our chronic care patients didn’t know they had a condition before talking to us, 87% of mDaktari users are first time telemed patients.
Strong distribution with thousands of products moving to and between healthcare agents/ micro-clinics using trucks/ cars/ bikes, tracked and providing analytics on what is needed where.
Mass use is trickling up. We started by solving a problem that over half the country experienced: a safe place to receive healthcare advice, services and products for the majority of the country who is uninsured. Now, the company’s franchise model is facilitating mass scale with local entrepreneurs applying to run their own micro-clinic in their own community.
Access Afya is just one example of a platform for a distributed healthcare network, and others are starting to emerge around the world. M-Pesa worked because of its tech, trust, and distribution, and we probably would not have heard of it had it skimped on just one of those. Let’s not forget this as we work across sectors to disrupt the status quo.
—
I love talking about this topic! Please contact me if you agree, disagree or just want to swap ideas about healthcare disruption.
** Thanks to John Nicholas, Chris Wong, Oswald Jumira, Peter Park and Karena de Souza for the conversations and edits **
Notes:
“The Cashless Clinic: Acceptability and Benefits of Mobile Money in Kenya's Health Sector.” By: Marilyn Ommeh, Melissa Menke, Alexander Kohnstamm, Cees Hesp. [PMID: 30707810] Findings from our research running a fully cashless clinic, with support from PharmAccess Foundation. We found about a quarter of patients more than happy to pay with M-Pesa, but I give some tips (especially in the blog link) for how M-Pesa could be even more usable in retail settings.
28. Aronson S. The Lancet on the telephone, 1876–1975. Med Hist 1977;21(1):69–87. [PubMed: 320411]
The mantra of Dr. Paul Farmer, who has written about this prolifically, including a piece from 2014 about the Ebola outbreak in West Africa, which eerily rings true today (highlights and bolding my own):
I’ve been asked more than once what the formula for effective action against Ebola might be. It’s often those reluctant to invest in a comprehensive model of prevention and care for the poor who ask for ready-made solutions. What’s the ‘model’ or the ‘minimum basic package’? What are the ‘metrics’ to evaluate ‘cost-effectiveness’? The desire for simple solutions and for proof of a high ‘return on investment’ will be encountered by anyone aiming to deliver comprehensive services (which will necessarily include both prevention and care, all too often pitted against each other) to the poor. Anyone whose metrics or proof are judged wanting is likely to receive a cool reception, even though the Ebola crisis should serve as an object lesson and rebuke to those who tolerate anaemic state funding of, or even cutbacks in, public health and healthcare delivery. Without staff, stuff, space and systems, nothing can be done.
See, for example, multiple case studies available from the Harvard Business Review.
Amazon moving back to physical stores is not just about the experience, but about last-mile delivery: https://www.theatlantic.com/technology/archive/2021/08/amazon-department-stores-prime-shipping/619860/