This is a three part series on Mobile Health. Look for two more posts from my colleagues, Karen Chen a mhealth researcher and Amy Lehman MBA/MD of the Floating Tanganyika Health Clinic

Photo credit Joel Selanikio, www.datadyne.org

The mobile technology infrastructure in Africa sets a powerful platform to begin to address some the problems around healthcare access and delivery. As I will mention in this 3 part series on mHealth, technology alone cannot address the issues, but likely a mix of services enabled by mobile services can create highly compelling, scalable and affordable solutions to real health problems.

Latest figures by the GSMA, show 65% mobile phone penetration in Africa. 96% of subscriptions are pre-paid. The figures might also be misleading because you may end up getting 1 person subscribing to multiple services- so its very important to check per country statistics and get a feel of the real adoption in that region. Either way, the fact that we are fast approaching universal coverage is good news and offers opportunities to invest or redesign services inc. healthcare. In this first post on mhealth, lets dive into the technology components and some business models that have the most potential for delivering mhealth services.

The cloud- the backbone infrastructure
We are in the cloud computing era- this means the need for expensive infrastructure to set up that once cost millions can be pushed onto service providers. Providers like Safaricom and Seven Seas are starting to offer public clouds that brings affordable solutions and turns what used to be capital Expense into an operating expense. As more data centers go up, the viability only improves. Big uses of using the cloud is to not only store data such as personal health care records at scale, enable scable web applications, but also to offload computational intensive tasks to the benefit of health- one example would be for genomics research I talked about.

SMS- ultimate in delivery and notification
Perhaps the single most important and often least appreciated technology is SMS. This simple yet pervasive technology is sometimes understated by technologists as to the impact and scale. Billions of SMS are sent everyday. In Africa, it’s probably more important communication medium than e-mail and phone calls themselves. Given that it works across any phone, feature of smartphone- it is a service that should be at the heart of many mhealth services. One of the most common uses of this could be to allow users to request information as well as to provide short notifications and basic information. Where it begins to fall short is when you try do too much interactive connectivity to the point where the user experience does not make sense. You might use SMS for appointment reminders but maybe not to allow a patient to view their entire personal health record stored in the cloud (but you might initiate it via an SMS request). The trickiness comes in working with a good third party SMS providers that can provide the scale, integration, delivery and ease of use that is needed- there is often a high barrier to entry to set up a service and one needs to often set up a deposit account of credits to get going. Another advantage of SMS is that you can use premium SMS services as a source of revenue to cover sms costs or even add a small profit margin all seamingly built in.
What is really needed is allow a developer like Twilio onto the continent to disrupt the whole third party distribution business and make it easier to develop cloud enabled SMS applications.

Simtoolkit/USSD- lets add some interaction
This service built into simcards allows more menu based interactive services but it is literally contrained to the simcard level which means working closely with carriers or 3rd party operators that provide this functionality. The big upside is that it works on almost every phone. After all M-PESA mobile money in Kenya is implemented using this service, but no surprise it works given its a carrier controlled service. I think USSD may make sense for some very rich interactive mhealth features delivered broadly. Also the cost once the development is done of communicating is much cheaper than working with SMS services. While it is not useful as a notification service like SMS, it is often paired with it to achieve good results.

Mobile Web- Rich content & services, backward and future proof
Where SMS and USSD ends, mobile web for rich content and interactivity begins.Mobile web is the next big thing in Africa and opens so many options. When pages of information, text, images and menus are required (e.g. say for describing symptoms or creating a health information portal), this wins.
A sleeping giant in Africa in the mobile web space is the Opera mini browser, reporting up to 70% share of mobile web requests. The magic of opera mini is that it is bundled with many handsets and supported to be bandwidth efficient making service available quickly and cost effectively. It would be wrong to not mention HTML5 when talking about mobile browsers, set to revolutionize the web and actually allows Africa developers to quickly tap into this technology for delivering rich services without having to go via cumbersome app stores. Some core components not to be ignored in the future are “offline access”, “location based services”, and “web hooks” – all of which allow unique services limited only by your imagination. Of course, the HTML5 standard is not fully supported across all browsers, so need to be careful- but Opera mini supports a no. of them. To my surprise, I was shocked to see that Opera has a appstore and even works with local carriers like Safaricom to integrate mobile money services.
One to watch in this area is windows phone operating system on Nokia devices in future- whilst symbian powered Nokia phones maybe around a little longer, to remain competitive with Android- Nokia and Microsoft has no choice but bring richer solutions to market to this region of the world to remain relevant.
The other thing not to overlook about mobile web is that it is backwards compatible with feature phones as well as with emerging tablet computers in future and of course normal computers. A well designed and integrated service with the right views allows multiple devices to work seamlessly and designing that is the challenge.
One final obvious point is to pay attention to connectivity rates and coverage, whilst browsers like Opera mini help in this regard- one should not expect to have fast 3G internet available in the most rural areas currently, but this will surely change with time.

Mobile Money- huge potential for fee payment services
This leads on the huge trend in mobile money- forecasted to with 350M accounts by 2015. M-PESA does more transactions than western does globally and responsible for $7B is transfer in 4 years since launch. Mobile money is another service that when well integrated into any pay per fee service can extremely compelling to mhealth service users. Suddenly the viability of micro health insurance to the masses becomes a real possibility. This is particularly important given the increasing role the private sector in the overal health care space in a world of budget and capacity constrained public services and the popularity of fee for service payments to access health services. Again, important to check per country how this might vary. Similar, for mobile money- adoptions rates vary widely, Kenya is the standout best market for mobile money at 70% adult penetration and dominant and widely used M-PESA. It becomes more difficult in going to other countries like Tanzania, where adoption is still low but growing fast, also fragmentation of mobile money services due to no single incumbent means you have to pay attention to being able to integrate multiple services- not a fun task, but solutions are emerging from a number of startups for both remote payments as well as Point of Sale (POS) integration.
Some surprising problem areas that you may not expect, when thinking mhealth services, its easy to just focus on fee for service- but also look for other problems that can be solved beyond the obvious. For instance, why not help coordinate and pay for expensive transportation to get patients to clinics and hospitals faster. How about allowing diaspora and migrants to directly pay for health service like Willstream is doing with remittance solution in Senegal? mHealth and especially the payment side is not what you might expect to just pay for medical services.

Business models

This is probably one of the hardest areas for entrepreneurs to think about for making a health service viable. The reality is that the people that need the service often have the least means to afford it. As I had mentioned micro-insurance models via mobile payments make a ton of sense as they help bring the masses into a formal healthcare system. Serious Bottom of the Pyramid thinking is required.

Prepaid subscriptions may also work where mobile money is not appropriate- for instance to access content. The ease of use in subscribing via sms cannot be underestimated, for instance in Tanzania it is very easy to send an SMS to enroll various packages for data and/or content and just as easy to turn them off- all the billing and deduction whether daily, weekly etc.. can be handled for you by the carrier or some mobile technology provider and as I mentioned, premium sms revenue when applied at scale may even cover operational costs of the services- but getting scale is key.
The same approach can be applied for enrolling on USSD enabled service. The downside with this is the need to work with carriers to implement the service and they may take a cut, but the big upside pay off is that if it works, you get instant distributions to millions of subcribers and opportunities to co-promoting efforts to gain adoption. As carriers continue to fight out for marketshare and saturation point is reached- they will increasingly turn to value added services to retain subscribers- one of which would be mhealth services. As such this channel offers a very lucrative payoff if the upfront investment and time to work with such partners is put in.

The other common way to enable mhealth service is to shift the burden of revenue to the institutions themselves. This includes at the clinic or hospital level, research institution that are willing to subsidize the service on behalf of their patients or target populations. The scope may not be a grand since many research institutions target vertical areas as opposed to broad health issues. However the solution or service offered may be very meaningful and your service may help set up interactions between service providers and patients- for instance an appointment or reminder for doctors schedule is highly relevant to local population, or an app or mobileweb service that allows find a qualified doctor or medical practitioner. What is required here is deeper collaboration with institutions but I feel the results can be worth it and may allow the venture to demonstrate a proof of concept and access more serious financing by impact and ordinary investors alike for scaling up. I think this is my preferred approach because it forces the entrepreneurs to build tight partnerships with health service providers rather than trying to provide a service that is mainly technology or information access that might risk being out of touch with the real needs of the local population.

Privacy & Regulatory issues
One of the biggest sticking point concerns of mhealth services in the western world is the privacy and regulatory rules that one might encounter in making a service viable or the bureaucracy that might burden you to even get the service of the ground. For example in USA, the HIPPA rules are critical to observe when handling any patient or health data.
In Africa, given the urgency of such services to solve huge problems and often less concerns about privacy means you have a better shot of lanching mhealth services. This does not mean you can ignore basic issues. For instance, all patient health data should be protected and encrypted when being help on the web. Regulations may require that data be held in data centers on shore rather that on a public cloud infrastructure such as Amazon web services. This area actually remains quite murky given many Governments in Africa’s lack of technology, so it makes sense to help educate and work with authorities then try and alienate yourselves from them. By working closely, you may get unexpected benefits such as a government mandating your technology implementation as a standard. Given mHealth technology and interporably are important when working with different stakeholders, working with Government is probably an absolute requirement.