Project EpiGraph

Strengthening outbreak response capacity in low-resource communities with open-source tools and collaborative support for epidemiological data management, analysis and reporting. 

Olivia Keiser

Director of Project EpiGraph

Head of Division of Infectious Diseases and Mathematical Modelling

Institute of Global Health, University of Geneva

Olivia.keiser@unige.ch

Emerging diseases disproportionately threaten LMICs, but human and technical capacity to quickly respond to outbreaks remains limited. Prior to the COVID-19 pandemic, the WHO African Region has grappled with over 100 public health emergencies annually. Many of these emergencies result in high morbidity, mortality and socioeconomic disruptions, which threaten national, and regional health security.

Half the world lacks access to essential health services, and there is a global shortage of health workers. Africa has 12.8 skilled health professionals per 10,000 population while the Americas and Europe have 82.6 and 105.3 per 10,000 population, respectively. There is also a lack of local researchers, data analysts, and outbreak response experts. Combined with limited access to digital tools, training and funding, these disparities constrain outbreak response capacity, sustainable national public health programs, and creates dependence of LMICs on external expertise for outbreak data analysis - slowing the availability of information for making evidence-based decisions that can save lives and livelihoods.

This project provides an open-source digital tool for rapid epidemic response and analysis, extensive technical training support for both use and further development, and a collaborative network to connect epidemic response teams across the global south and beyond.  

The COVID-19 pandemic has highlighted important shortcomings in capacities to perform timely collection and analysis of outbreak data. These shortfalls have been particularly pronounced in low-income African countries, where weak information management infrastructures and a dearth of trained data analysts have often hindered information flow.

In June of 2020, Project EpiGraph was conceived by researchers at the University of Geneva's Institute of Global Health after being approached by the World Health Organization's Africa Office to assist Sub-Saharan African countries in managing and analyzing their streams of COVID-19-related data. In response, we formed a diverse team of global health researchers, epidemiologists, and data analysts from 33 countries, including experts and students from 17 of the African countries being assisted, to respond to this call. 

Thus, Project EpiGraph is not just responding to the needs of a target audience - the audience itself is fully engaged in building and leading aspects of the solution. In-country members are currently lobbying their health ministries for further engagement and support, such that public health authorities at all levels are involved in providing input for development as it is scaled. 

Project EpiGraph provides several public goods. 

1. The EpiGraphHub is an open-source application, freely available to any user. 

2. The EpiGraph Training platform is a free tool available online. 

3. The GRAPH Network is a resource for public health professionals and students that serves as a platform for technical support, collaboration and learning. The network also produces several types of publications, including epidemic situation reports commissioned by country health ministries, and peer-reviewed publications. 

That said, it can be argued that the main public good that Project EpiGraph provides is enhanced capacity of public health professionals and decision makers in LMICs to monitor, understand, and control diseases outbreaks based on strengthened data analysis capacity, both technical and human.

The logic of how Project EpiGraph creates impact is outlined graphically in the Annex Figure 4. The main beneficiaries of the project’s outcome are the public health authorities of partner LMICs and, eventually, the global community due to enhanced collaboration with and between LMICs on outbreak control strategies. Rapid availability of scientifically-vetted information provided to decision makers early in an epidemic will save lives and livelihoods, both locally, and potentially globally for emerging pandemic diseases. Overall improvement in the management of infectious diseases will help LMICs reach sustainable development goals.

Transversal future use of standardized but adjustable, open-source, easy to use and inexpensive tools, such as EpiGraphHub, will enhance South-South collaboration in the medium to long term. Here, we provide a testimonial from one of our team members from Cameroon, Benedict Esseme, who has risen from student to mentor and training coordinator over the course of the project. Additional feedback and impacts are detailed in this draft manuscript in preparation for publication. 

We will scale our impact by piloting the full project in collaboration with 1-3 specific country health ministries over the first year. We will then expand roll-out to include at least 10 sub-Saharan African countries in the next 3 years, depending on the speed of up-take (see Annex Figure 5). The more countries that get involved, the more experts will join the network, and the more human-power there will be for carrying out additional country roll-outs. 

An important note here is that several non-African LMIC members are part of the network, and interest in Project EpiGraph for other WHO regions has been communicated. WHO central offices have so far committed to scaling the EpiGraph training platform, including the mentoring structures and likely also including fellowships, to allow it to function worldwide. Thus, this project will surely come to benefit many more than the 1.3 billion people on the African continent. 

Furthermore, meeting the global need for epidemic response capacity in under-served and vulnerable populations often at the edge of the human-animal interface where zoonoses arise, stopping epidemics early, will help to stem future pandemic crises.

Project EpiGraph, and in particular the EpiGraphHub application development, deployment, implementation, and transfer processes innately integrate evaluation in the form of the constant collection of feedback from users, learners and partners, resulting in the updating of scripts and processes accordingly.

Tracking progress towards the outcomes indicated in the logical framework of the project consists of routine monitoring by tracking the quantifiable measurements of the project’s outputs and of periodic external evaluations by professional independent verification agents, such as data analysis and epidemiology professionals, with the use of relevant WHO performance frameworks (IHR Monitoring and Evaluation Framework, attributable WHO Emergency Response Framework indicators for countries, country indicators from the Monitoring and evaluation framework for the COVID-19 response in the WHO African Region) and their specifically developed impact evaluation methods. An increase in each country's WHO's health data and information systems SCORE assessment (next evaluated in 2025) is of particular interest. These independent evaluations will be conducted every 12 months of the project implementation with the terms of reference progressively transforming from data point verification to impact measurement with the growth of the project’s results.

Performance evaluation framework for the project roll out is outlined in the Annex Figure 1.

The main barrier is a lack of funding - for EpiGraphHub in particular. We are mitigating this risk by pursuing large and small donors from across public health, development and technology innovation sectors (BMGF, Fondation Botnar, The Global Fund, AFD, SDC, UNAIDS, European Commission, European Space Agency, as well as WHO central office. We have commitment of personnel time from partner countries, though these are - by definition - partners with limited resources, and from WHO for sustaining the network and scaling the training platform. That said, EpiGraphHub development has so far been a largely voluntary activity among GRAPH Network members that has continued even after funding expired. While this is not a sustainable model, and would make for a slow roll-out, it demonstrates our level of commitment and capacity to realize the maximum impact potential of this project. 

We have already begun discussions with health ministries in Burkina Faso, DRC, Senegal, Cote d’Ivoire, Malawi, Uganda, and Zimbabwe. There is a chance that the three other target countries (Congo, Mauritius, Guinea) do not accept our proposal. However, it is likely other countries will want to be included, and that uptake by neighboring states and WHO bannering will provide motivation.

University of Geneva

Swiss Research Association "Actions en Santé Publique"

World Health Organization

UNAIDS

Renku

École polytechnique fédérale de Lausanne (EPFL)

African Health Ministries (via WHO and directly for Mauritius, Burkina Faso, Senegal, Côte d'Ivoire, DRC)

African research and development projects benefit from less than 0.5% of world R&D expenditures. Very often, access to the world's research funded by the other 95% is also limited. We are applying to The Trinity Challenge to: 

(a) address the specific challenge topic, "How can data and analytics be used to better identify, respond to, and recover from global health emergencies?”, because we have a solution that is already being deployed to handle two of these objectives (identify and respond) from within the WHO; 

(b) help fund the project, which has a number of contributions and commitments from partners but which is not yet entirely funded;

(c) make connections with strategic partners, who can help us to achieve our goals of making technology and data accessible to African health researchers and health ministries (see potential partner descriptions below).

As an organization, Project EpiGraph has partnered with several important stakeholders, including the World Health Organization, African health ministries, UNAIDS, and those supplying technical infrastructure (Renku, EPFL). We have also made important steps towards partnering with synergistic projects such as  Global.health, RECON, and OpenIMIS, as well as the French geospatial tech SME Geomatys. We are in the process of adding other synergistic partners to this list, such as DHIS2 and SORMAS.

However, several Trinity Challenge Member organizations are also of interest to us as strategic partners. First and foremost, the Bill & Melinda Gates Foundation is a top priority, given their commitment to helping improve health worldwide, including protecting against future pandemics. 

Google and Facebook could both provide critical data on behaviors and movement of people, which can be made available on secure servers to countries for real-time epidemic data and response analysis. Blue Dot is a company that specializes in predictive analytics, who could add expertise in this area to the team and algorithms to the tool. We are eager to hear what contributions may be possible from other Trinity Challenge members as well.