Mead Over recently wrote compellingly about the importance of collective action to strengthen surveillance of the swine flu and other contagious diseases. A major issue, of course, is the cost of such surveillance measures, the timely receipt of data on potential infections, and the accuracy (and completeness) of such information. Mobile phones may be part of the solution. Mexico, for example, with some 110 million people, has only 1.5 doctors per 1,000 people but by 2005 already had some 44 million cell phone subscribers. It seems a safe bet that by now half the population has a cell phone. Can we tap this huge information network for public health?
If public health workers are well-trained to diagnose and report potential swine flu cases, they can easily (and cheaply) send these data over long distances – therefore minimizing the needs for more health infrastructure or costly field visits by health professionals to rural areas. The technology exists – the health workers just need a cell phone and a number where they can send their data (such as via Frontline SMS, which can be useful for sending and receiving market information).
This is already happening at the grassroots level. The “crowdsourcing” platform Ushahidi allows individuals to report instances of swine flu in their region. The tool maps reported cases in real-time, and allows other users to confirm or reject the reports. (For technological novices, “crowdsourcing” refers to outsourcing a task to a large community or group of people. So, rather than have the Ministry of Health workers report swine flu cases, this technology lets the “crowd” – more specifically regular citizens – report cases via SMS or the Internet, and allows others to confirm or deny these reports. I learned about this during an interesting discussion at the recent Mobile for Social Change bar camp co-organized by MobileActive.org).
The Ushahidi tool is a variation of other mobile health (m-health) projects in developing countries, such as the use of PDAs by health workers to track measles outbreaks in the Zambia, monitoring avian flu outbreaks in Africa, and disease surveillance in India and Uganda, among others.
Great idea, right? Of course. But you could also imagine potential problems. Citizens could under-report if they’re afraid that the Ministry of Health would quarantine their village, or if they think that someone else will report. Or they could over-report, if they perceive some financial or social benefit to their community. Verifying reported cases is a potential problem in both situations.
This doesn’t mean that cell phones can’t or shouldn’t be used to monitor and prepare for future epidemics. It just means that governments and donors need to carefully think about what they want to monitor, how they want to monitor it and how cell phones can fit into the process. This means answering the following questions:
- What are we trying to measure? In other words, do we want to simply know the geographic location of potential outbreaks, the actual number of cases in a specific location, or both? While the number of confirmed cases of a specific disease will be important to manage an epidemic, timely information on the outbreak of potential cases can help to better concentrate scarce resources.
- How and from whom do we want to obtain this information? Cell phones in mobile health can be useful in two ways – to obtain information from the general public (crowdsourcing) or from trained experts (such as public health workers). Both groups can be useful sources of information – but it depends upon the type of information that we want (and need), the measurement of that information and the strength of the public health infrastructure in the country.
- How can we design incentives for the truthful data collection? This is not only a problem for crowdsourcing but also for trained professionals. Part of this is related to the free-rider problem – why should I report if someone else is? -- and part of it is related to the actual cost of reporting – in other words, can I afford to make the call or send the SMS? This not only requires a information campaign on the disease and its potential effects, but also ensuring that any costs of reporting are covered.
- How can we link with the private sector for this venture? Surveillance measures are often criticized for being expensive. And calling or texting can be prohibitively expensive for the rural poor. This points to the need for a public-private partnership with cell phone service providers in developing countries. In Ghana, for example, election monitors were able to call or text a free “election hotline.” Yes, the cell phone company lost out on potential profits from those calls – but probably won in terms of the number of potential clients in the future and business from other organizations.