Mobile phones are everywhere these days – whether you live in the city or the countryside, have a basic nokia or a deluxe smartphone, have your own or share one – mobile phones are now an integral part of life on our planet.
In fact, more people now have mobile phones than have toilets.
It’s no wonder then that then that the health sectors have seized on mobile phones as a way to reach the communities that they serve – the mHealth working group, which brings together practitioners working in this field called ‘mHealth’ has over 1000 discussions on its discussion forum from its 1500+ members, while a search on google for mHealth brings up 1.7 million results.
However with so much out there it can be hard to know where to start when developing a programme that you think could (or should) involve mobile phones in some way.
Below I’ve included my top three ‘do’s’ and ‘don’ts’ to help guide your thought processes around using mobile phones, from my eight years of experience in designing and evaluating interventions using mobiles phones, whether text messages, voice messages or even using social networking sites.
“Do’s”
1. Do think about why someone would want to receive (and respond) to your message
You don’t want to be another annoyance in someone’s life – or worse, another advertising message that is completely ignored. Make your message appealing in some way; the best way we found for our sexual health messages was to make them funny, rhyming or link them into a relevant event for our audience (young people) such as Valentine’s Day which then resulted in forwarding to others (“Roses are red, daisies are white, use a condom if you get lucky tonight. Happy Valentines Day!”). MAMA ensures their messages are relevant to their audience of expectant mothers by sending messages tailored to the stage of pregnancy.
Another good approach for engaging your audience is to have quiz questions such as those run by Text to Change in Uganda , with small incentives or prize draws for individuals who respond correctly - those receiving the messages have a clear drive to respond (the incentive or prize draw) while you have the opportunity to educate those who respond incorrectly by sending them the correct answer.
Just because you work in the health sector, or in another sector for social change, doesn’t mean you should be boring! Take the 140 character limit on text messages as a challenge, and be creative as you can be with the small amount of space you have to maximise the amount of time someone will spend engaging with your message.
2. Do examine how your intended audience uses their mobile phone
What phone people have, and how they use it, varies greatly between and within countries. The Philippines is the text messaging capital of the world, with the average user sending 600 messages a month in 2010, while anyone who visits Israel is astounded by just how many calls an Israeli makes every day on their phone. Many people living in Africa and Asia use two, three or four prepaid SIMs concurrently, swapping at a frequency a European or North American with their monthly contract plan cannot conceive of. Kenyan’s can pay their bus fare, school fees and rent via through mobile money, a service that is only just being introduced in the United Kingdom seven years later. And despite what you might see on the streets of London, New York or Beijing, ‘feature’ phones, not smartphones, still dominate the global market.These short examples demonstrate how diverse the use of mobile phones is on our planet; you must factor this into the design of your mHealth intervention if you are to be successful. A 12 month program to follow up mothers and their newborns will not succeed if mothers change their phone number every three months and there is no mechanism to update contact details. Similarly, a text message based intervention to deliver epidemic alerts is unlikely to be successful in a country where the local script (or a romanisied version) is not commonly used by the local population to send text messages.A useful place to start learning more about mobile phones in your country is the Mobile for Development Impact site that pulls together mobile phone and health statistics, revealing for example that in Sierra Leone there were 69 mobile phone connections for every 100 people in 2013 compared to just 48 in 2011. Look around you on the streets where you live and work, and talk to as many people inside and outside your networks to find out exactly how your target population uses their mobile phone. 3. Do spend time developing and refining your approachThis isn’t rocket science, it’s simply good intervention development. As with every single community intervention, you need to spend time testing out your intervention with your intended audience. If you don’t, you risk having an irrelevant, or even harmful, intervention. An intervention I provided advice on in Cambodia found out through initial interviews and focus groups that the intended plan of using text messages was unlikely to work as most people didn’t have the ability to text in Khmer script on their phone and changed their intervention to instead successfully use voice messages (results to be published shortly). Had they not found this out early, they would have wasted months if not years of effort for something that could be addressed relatively easily at an earlier stage.Get out there and talk to people. Show them what you have been developing and ask them to honestly comment on it. Don’t wait until your intervention is almost finished – as with any innovation, the best thing you can do is develop a ‘prototype’ (early model of your intervention), test it, learn, improve the prototype, test it, learn and so on. Don’t let time pressures (real or perceived) to launch your intervention take precedent over this critical stage – no matter how much of an expert you are or how many hours of work you have put in, YOU are not the best judge of the quality of your intervention! “Don’ts” 1. Don’t assume using a mobile phone is what your programme needsBe realistic about when a mobile health intervention is appropriate, and when it isn’t. Coordinating emergency transport for women in labour via mobile phone calls may be useful – but not if network reception in the area is unreliable. Information about complex and unfamiliar issues such as stigma surrounding HIV or women’s rights are unlikely to be able to be usefully condensed into a 140 character text message.Mobile phone based interventions, particularly those using text messages, work best as reminders or prompts to individuals, to bring individual’s existing knowledge to the front of their mind, or remind them to do something relatively, like take their medication. Your starting point should NEVER be ‘how can I use mobile phones in my program’ but rather what challenge you want to solve, with mobile phones one potential to address this (regardless of how ‘sexy’ you think your donor will find an mHealth intervention). 2. Don’t underestimate the time you will need to find, understand and manage the software you will needDon’t panic – its not as hard as you think to find software to suit your needs. There are a plethora of open source solutions out there, such as FrontlineSMS, CommCare, ODK, Verboice, Vumi, and many others that cater for all manner of different mHealth interventions. My advice is to avoid developing your own, ‘bespoke’ solution – they always take longer, and cost more, than you initially planned and chances are, if there is something that you want to do, someone else has already thought about it. Instead, spend your time and energy choosing from the existing options, making sure you understand what they can do and how they work.Remember that even if the software itself is free, the time to customise it to your needs is not, nor is the cost of sending and receiving voice calls, messages or data. Having someone in your team – or someone you trust who you can call on – who can understand both software developers and health programming is invaluable when you are negotiating with potential suppliers, software developers and telecommunications providers. Many projects have taken far longer than expected to develop, as the programme implementers were unable to translate their needs into language that software developers can understand (and vice-versa). 3. Don’t forget about monitoring and evaluating your programmeBefore you launch any intervention, including one involving mobile phones, it’s important to consider how you judge success of your intervention, and how this will be measured. Will 100 people signing up for your Ebola alerts be considered a success? 1000? 10,000? Do you want people to just read your messages or also respond to them? What about forwarding them to others? All of these can be measured, but are best established at the outset so that tracking mechanisms can be put in place from day one.The best programmes have a strong theory of change (some statement of how what they are doing will lead to the effects they are intending), a clear way of monitoring this flow of events and the flexibility to improve the intervention over time. Unless you are running a scientific trial, there is no reason why you can’t change your approach midway, for instance if you find quiz messages generate 50% more responses than open ended questions.I hope you found this blog post helpful in inspiring your own mHealth programmes – a good place to start when designing your mHealth intervention is this online toolkit that talks you through initial concept development, solution design and testing, and planning for implementation. Good luck!This blog was written by Dr Judy Gold, a programme innovator, implementer and evaluator who has been involved in developing and evaluating mHealth interventions for the past eight years.Click on this link to access a downloadable version of this blog.