The Center for Data Innovation spoke with Ghassan Aziz, Health Surveillance Program Manager at Médecins Sans Frontières (Doctors Without Borders) Middle East Unit in Amman, Jordan. Aziz discussed the challenges involved in gathering data in conflict zones and how a new tool called the Dharma Platform could make these challenges easier.
This interview has been lightly edited.
Joshua New: Can you explain the concept of health surveillance and why it’s useful for organizations like Médecins Sans Frontières (MSF)?
Ghassan Aziz: MSF is an international, independent, medical humanitarian organization that delivers emergency aid to people affected by armed conflicts, epidemics, natural disasters, and exclusion from healthcare. In 2016 MSF had established a decentralized Middle East unit based in Amman, Jordan to support MSF operations in the Middle East. Within the unit we have the MSF Center for the Advancement of Humanitarian Medicine (CAHM) which runs some support activities, one of which is the Health Surveillance Program (HSP). In general, health surveillance is essential for a sustainable and successful public health intervention designed for prevention purposes. MSF HSP is an operational data program that provides open, reliable, and timely information on the health situation and essential service needs of the people affected by conflict in the Middle East, allowing the utilization of actionable time-series information from crisis-affected communities where MSF serves in order to detect and respond to the evolving population health services’ needs. This allows MSF to better design projects and better allocate resources to provide the maximum support for the people in need.
New: You have helped MSF deploy software called the Dharma Platform in several countries, most recently in Syria. What does this software let MSF do with data that they couldn’t do already?
Aziz: Dharma Platform is one of the new platforms that MSF started to pilot side by side with the other standard software and applications that are already being regularly used by different MSF projects. MSF mainly uses Dharma for the challenging large-scale community-based households surveys. Instead of the common paper-based assessments, we were using Dharma as a mobile data collection tool that can work in some of the most difficult settings like Syria. But using Dharma was far beyond simple data collection— it helped us to have a kind of “all-in-one” solution where we were able to collect responses in the people’s spoken language with excellent built-in skip logics and conditioned questionnaires, real-time results visualization in English, built-in analytics, optimal visualization of the results, multiple data export options, and real-time data collection staff tracking. This all in one solution of using Dharma helped reduce the effort it takes to conduct a survey, perform data entry, and analyze data until final reporting. This saved us two to three months of time, allowing MSF to have data-driven decisions much faster than usual. Just as an example, MSF teams in Iraq interviewed 6,600 randomly selected displaced Iraqi individuals. Each one of those individuals answered an average of 20 health and health-related questions over four days of data collection. Final results were available on the MSF Dharma dashboard by the end of those four days.
From that start point and with the continuous evolution of Dharma we in HSP decided to use Dharma in solving one additional challenge facing us working in difficult settings. The challenge was to get accurate and timely data from some of the health facilities that we are supporting across borders without having physical presence in them. Language barriers, patient overloads, and the lack of enough qualified human resources all played a role in making receiving health facilities data a long and complicated process. Building an individual-based data collection project on Dharma was the solution. With its forms building options and ease of use even by non-medical staff in the field we managed to use Dharma to collect detailed health facility activity data since March 2017. Now, we have detailed data of around 50,000 patients treated just in one health facility, all collected, visualized, and analysed using the Dharma Platform.
New: What kind of impact is this software having in Syria? Or is it too early to tell?
Aziz: In all the countries that we used Dharma inside the Middle East, data collected through Dharma is being used for operational and advocacy purposes. Just one example about our findings in Syria is that we believe that our recent detailed immunization coverage data for children younger than five is a very important, and we are sharing this data with several humanitarian actors. This will lead to significant change in policies and approaches currently used, aiming to increase the coverage of vaccination among Syrian children inside Syria to prevent disease outbreaks.
New: Working in a conflict zone typically means humanitarian workers have to go out in the field to collect data, rather than rely on phone or Internet surveys. Can you explain the challenges this poses and how MSF is working to overcome them?
Aziz: Indeed, there is a long list of challenges. Security concerns take the lead—you should avoid being in the wrong place at the wrong time, but you can never predict what could happen on a particular day on a particular road. Overcoming this always requires very careful and continuous monitoring of the situation, and this is even a harder job when you have to run everything from a different country.
Introducing a new technology is also a challenge. You will have to deal with a group of staff that have been selected because of their experience in certain medical fields and suddenly you ask them to use some new technology while their set of skills is usually not that technology-friendly, and this requires more time for training.
Some of the challenges were more logistical and environmental ones, like having the staff moving in the middle of Iraqi summer where the temperature is above 50 degrees Celsius (122 degrees Fahrenheit) or in snow in other locations. It is also always challenging to find a tablet that can keep running for more than eight or nine hours in the field since you cannot recharge it on the spot!
New: Whenever I hear about data-driven humanitarian efforts, it seems like aid workers frequently rely on community-developed, open-source software. Has this been your experience? Why do you think the humanitarian sector doesn’t have ready access to the same variety of tools as other sectors?
Aziz: This is a challenging question that needs proper clarifications. First of all, there is this general idea of humanitarian organizations being technology-phobic, which is incorrect. Humanitarian sector needs in term of technology are very complicated. For example, in MSF we work in more than 70 countries around the world, those countries have different laws and regulations and different level of experiences among their staff. What we need in terms of software should be very scalable, meaning it should be possible to use it in literally every single project in any given country. It also needs to be very easy to use to the extent that the staff turnover across any project should not be a problem for using that platform or software.
So to the contrary, we actually do not tend to rely on self-developed or community-developed software but we use the ones that have been in use enough so that finding experts users is never an issue. And for this reason it is always hard to introduce new software within our systems no matter how well developed it is! But we always keep our eyes open for software that we believe we can smoothly incorporate within our systems.