Interview with Natalie Roberts : COVID-19 Alert and SMS Afia Yetu projects

How were the applications Alert COVID-19 in Niger and SMS Afia Yetu in the Democratic Republic of Congo born?

Digital health is not a new topic for MSF, but the COVID-19 pandemic has pushed us to rethink these issues.

The Foundation, through its director, had the opportunity to work with volunteer engineers from Google.org for a few weeks to develop digital projects related to the pandemic. After a comparison of several app publishers, our choice fell on Medic Mobile, a non-profit organization that develops open source software for community health projects in low-resource settings. The issue for us was whether it would help our teams or our patients, because the idea was not to implement a tool just for the tool. It had to be simple, adaptable to the context and meet operational needs. The Foundation provides that opportunity: manage the risk and initiate a pilot to see if it's useful. We initiated discussions with Epicentre and with the cells and the medical department. Often, the teams were interested in the topic but did not see the need. Nevertheless, two opportunities arose.

In Goma, where we support the Ministry of Health's HIV program, the unit wanted to use a tool to track chronically ill patients. At the time of the COVID-19 pandemic, not knowing whether HIV patients were likely to develop a severe form of the disease, the team had reduced face-to-face contact with patients to avoid the risk of infecting them. No longer able to provide regular physical follow-up, they wanted to develop remote monitoring for their chronic illness, any signs of COVID-19 infection, or any health problems related to the pandemic response that might reduce their access to care. The idea took shape after almost a month of brainstorming. We evaluated various projects that track cohorts of patients with chronic diseases, including HIV, but for a variety of reasons, Goma seemed like the most promising location for a pilot project. One factor was the strong interest in the idea from some of the patients in the cohort, as well as from our counterparts in the Ministry of Health.

In Niger, health authorities were afraid of missing cases of COVID-19, a fast-moving disease that requires a rapid response. In order to develop the response, it was necessary to find the patients early on. The Ministry of Health therefore asked Epicentre for support in monitoring the epidemic. The SAMU had a call center in Niamey, but no link with the Ministry of Health's decentralized and mobile COVID-19 response teams, which travel to conduct investigations when there is an alert (tests, triage, home-based care, etc.). The COVID-19 Alert platform helps to link all of this together and to better escalate alerts.

Have these projects put the needs of health authorities and patients back at the center of the MSF approach?

In both projects, the needs came from the MSF and Epicentre field teams, who continue to be the driving force in proposing new ideas and solutions, with the support of the Foundation, the Epicentre team in Paris, the medical and legal departments, etc.

For Niger, Epicentre approached the authorities asking them how we could be useful to them rather than proposing a ready-made tool and piloting it: there was a real collaboration with the authorities who wanted to improve their surveillance system and reduce the delays between the moment a case is identified and the response. They would like to adapt the same tool for other epidemic diseases (measles, cholera, meningitis).

In Goma, the problem was different. We did not always have a good understanding of the problems of the patients in our HIV cohort, whom we regularly lost touch with. Communication between patients and MSF or the health system in general was often limited to appointments. With this device, patients should receive regular text messages with questions to help with diagnosis and be called back, based on their answers, by the MSF team who receive an alert in a smartphone app. This gives the possibility of a two-way communication between patient and team according to the needs. In addition, the software is programmed to sort alerts to help staff organize their work. It's simple, easy and we think people will use it. So far, patients are enthusiastic because they finally have the opportunity to participate in care. The idea is to integrate this device into patients' lives and be able to follow them more regularly, even outside of the pandemic. They have also been very willing to participate in the development of the application, testing the system and giving feedback. At an early stage, it already seems that the development of the tool can strengthen the relationship between patients and their caregivers. But there are many technical constraints that we do not yet master, such as the delivery of SMS and the stability of telecom operators. As with any new system, many difficulties appear every day that we solve with the team, one after the other... It's a real long-distance race!

In both situations, we focus on the idea of deploying simple, context-specific tools that allow patients and caregivers to self-initiate an alert at the appropriate level to enable a rapid and appropriate response, while continuing with their usual activities. We believe this could be useful in the approach to epidemics in general, where it is vital to avoid delays.

What obstacles may have blocked the MSF teams?

Initially, it was a bit difficult to get the discussion going about different ways to approach a disease we didn't know much about, and to propose devoting resources in a time of crisis to an unproven concept. In the early stages of the pandemic, many of the fields were focused on establishing treatment centers.

When we started talking to the medical department and operations, some people jumped at the opportunity, others shared misgivings about involving a third-party company, which led to many discussions between the Foundation's director and the legal department. To me, Big Tech is like Big Pharma. We can't treat patients without drugs and we don't have the capacity to make our own drugs. So we have to understand how to work with them! Working with companies whose core business is this is essential to our learning.

Another reservation was based on the discriminatory idea that in our intervention areas, the population would not understand, that people don't have smartphones, no access to the internet, don't speak French or English... Yet, there are very few projects today where our patients don't have access to at least a simple cell phone, and in many places, the use of social media platforms is huge.