It’s clear that the virus that causes Covid-19 travels freely. It needs no visa, having breached many points of entry. Experts at Imperial College London estimate that “in the absence of interventions, Covid-19 would have resulted in 7 billion infections and 40 million deaths globally this year.” Now is the time for governments to get ahead of the curve and respond to the emergency. It’s time to take a whole-of-government approach to strengthen testing at points of entry and institute mass testing at various points as South Korea did.

One such area of strengthening involves using digital tools to improve systems to manage real-time data and decision-support systems. In most low-and-middle-income countries, health information systems are weak and under-funded. A robust system to gather and analyze data and use the data to take action is required. These actions can involve isolating people, treating them, and managing consumer goods during lockdowns. Another aspect of managing the spread of the virus involves identifying elderly and at-risk populations with pre-conditions and then using the data to send targeted health communication messages and prepare for rapid response.

We propose five initiatives that ministries of health in low-and-middle-income countries need to focus on: improved data governance, rapid enhancement and implementation of digital solutions, real-time analysis and rapid response, data analytics and modelling support, and using technology to update health workers and the community.

Improving Data Governance Through a Whole-of-Government Approach

USAID’s report, Fighting Ebola with Information, refers to the “fog of information” that obscured the real picture of the Ebola crisis. Data was collected in silos with no standardization and no coordination at higher levels. To prevent a similar situation, more effort needs to be focused on building a lasting system to identify, share, and utilize real-time data through a coordinated, unified model of public-private partnership, said Dr. Raj Shah, President of the Rockefeller Foundation. Ministries of health should establish a Covid-19 response command center to oversee and coordinate all data collection and analysis across line ministries and partners. The command center should convene stakeholders such as implementing partners, the private sector, donors, and UN agencies to define governance protocols and develop a shared implementation plan for the country.

Rapid Enhancement, Implementation of Digital Solutions and Use of Standards

Quick response requires systems that can promptly detect and investigate cases. We recommend that ministries of health prioritize use of existing open source digital solutions for surveillance, laboratory reporting, case management, logistics management, and human resource management to provide real-time data. The best solution to manage data for the coronavirus outbreak should be based on existing technology in the country, local capacity, and plans for sustainability.

The Digital Square initiative has shared several global solutions, including DHIS2 Tracker; CommCare; SORMAS; ODK-X; OpenMRS; OpenSRP; Community Health Toolkit; and Go.Data; iHRIS (human resource information system); OpenELIS (laboratory information system); and OpenLMIS (logistic management information system) developed using open-source platforms. It has already released custom applications for tracking Covid-19 based on the World Health Organization’s contact tracing protocols. Countries should register their country implementation projects on WHO’s Digital Health Atlas, which is designed to help coordinate digital health activities at the country level.

Standard-based data exchange is key to enabling fast and smooth data exchange (for example, between clinical and laboratory information systems) to provide an immediate response and guide care. The OpenHIE (Health Information Exchange) community of practice has established a task force to support this initiative.

Real-time Analysis, Feedback, and Rapid Response

We have often observed that different systems are used to collect data in different formats, and it can be challenging to compare or match such data. To address this type of problem, Kenya’s Division of National AIDS and STI Control Program developed a data warehouse that allows weekly automated data transfer from different client level HIV data systems—including electronic medical records, laboratory information, or mhealth apps—to a central data repository using a standardized data format. When machine learning is applied to anonymized individual level data in a centralized database like this, it’s possible to analyze outbreaks such as Covid-19.

We recommend setting up decision support tools on top of the data warehouse and processes (taking a situation room approach) for early alerts, feedback, and managers’ actions. Specifically, one could conduct routine analysis of localized outbreaks, availability of medical supplies, hospital beds, and human resources available to respond to local needs. Similar analysis helped healthcare planners in China project fast-growing demand and pivot to address changing needs. As a result, hospitals with capacity for hundreds of beds were set up within days to be used as quarantine and treatment facilities.

It’s also important to support geospatial analyses (using open-street maps, QGIS, DHIS2, etc.) of cases, contacts, people/areas in quarantine, as well as assets such as health posts, schools, medical stores, and transportation infrastructure. The data can be used to plan and direct resources during the response to where they are needed most. We can engage telecom networks to assess knowledge and practices in the general community at scale by sending a short questionnaire via SMS. This can inform behavior change messages for the community.

Data Analytics and Modelling

A final area to focus on is advanced data analytics and modelling. This includes artificial intelligence and machine learning using data from multiple sources and sectors. Data from mobile phone operators, on smoking prevalence, and on the prevalence of other diseases can be used for vulnerability mapping and to develop early response health alert systems. Using innovative tools and analysis for mobility mapping, healthcare agencies could send alerts if new cases are identified in a region.

Using Technology for Updates

Another area to focus on involves using technology to update health workers and other personnel on clinical protocols, to screen individuals, and to help the community share health communication messages. Using mobile-enabled communication channels or messaging apps, providing feedback on data, and sending messages to update healthcare worker skills are all essential for the response. It’s also important to enhance the capacities of border-security officials to screen passengers at air and land points of entry by providing training via mobile applications. In the Republic of Korea, smart quarantine information system and free smartphone apps flagged infection hotspots by sending people text alerts on new local cases in the community.

Designing Data Systems for Future Scenarios

Could Covid-19 disappear completely? Such a scenario, said Dr. Bruce Aylward, an epidemiologist and Senior Adviser to the Director-General at the World Health Organization, would be “very, very unlikely,” because the virus spreads too easily among humans. We are more likely to see waves or low-level disease, he said. Countries need to adopt a broader and long-term vision to improving data systems to help manage mass testing, implementing the track-trace-treat continuum including contact tracing, isolation of sick patients, managing geographical outbreaks, and keeping in mind the future where a new rapid-test, self-testing, vaccine, and/or treatment will be introduced into the system. Data systems should be flexibly designed to adapt to future scenarios.

A well-coordinated response from the government, with support from the private sector, can help manage the spread of the virus and take corrective action. A whole-government, multi-sectoral approach is required to use outbreak data to manage consumer supplies, hospital beds, sharing of resources between hospitals, and deployment of healthcare workers. Using the appropriate technology solution to manage data, using a central data warehouse for standardized data, using data daily and weekly to take quick action, and investing in modelling to predict potential coronavirus outbreaks, are approaches that may make it possible to limit the spread of Covid-19 and prevent needless illness and deaths.