Connected Diagnostics for Early Warning of Disease Outbreaks

We propose to build a global platform that leverages our existing network of over 900 point-of-care, multi-parameter diagnostic devices deployed in Asia and Africa, and our strong operational and research capabilities to build a scalable early warning system for future pandemics.

Ramanan Laxminarayan is a globally renowned infectious diseases researcher and thought leader in connected diagnostics. He is founder of the Center for Disease Dynamics, Economics & Policy (CDDEP) and HealthCubed.

Early detection of pandemics is a global challenge. Researchers estimate it may be possible to contain an outbreak of pandemic influenza only if human-to-human transmission is detected within two weeks after the first occurrence.    

Early warning systems that are based on data that are routinely collected for other purposes are likely to be more successful and sustainable than those that require new technologies and systems to be deployed.  Long before pandemics are reported through official channels, there are signals of their existence in unusual clusters of fevers of unexplained etiology, unusual population movements, changes in microbial composition of sewage, and chatter in social media. However, no single source is entirely reliable. Social media signals may be non-specific without the context of underlying disease factors or clinical information. The risk of new outbreaks is concentrated in a few hotspots around the world, and most outbreaks are related to drug-resistant infections that alter our ability to treat infections.

The problem that we are trying to solve is how to leverage routinely collected data from distributed, connected point-of-care diagnostics, in conjunction with other datasets including information on underlying disease risk to provide early warnings of pandemics.

The primary goal of HealthCubed technology is to make diagnostics for common infectious and non-communicable diseases easily accessible to people everywhere.  An estimated 4 billion people on the planet lack access to reliable diagnostics and this impedes their access to healthcare.  In this project, we will leverage our existing, on-the-ground connect to over a million patients to use anonymized data to identify abnormal clusters of disease.  Our solution helps bring malaria diagnosis to an individual in rural Kenya but that information then immediately informs public health officials of malaria prevalence in that area. Our target audiences are patients and healthcare providers (who are the primary users of our technology), local governments (who use our data on disease patterns), and national governments and global organizations such as WHO (who could respond to our early warnings).  We work with hundreds of healthcare providers and over a million patients to constantly improve our technology and usability.  CDDEP works globally on policy issues related to pandemics and zoonotic diseases including antimicrobial resistance.  Ramanan Laxminarayan is  formally appointed as advisor to two state governments (Andhra Pradesh and Tamil Nadu) and is well positioned to ensure that our solution is responsive to their needs.

There are multiple public goods that are direct result of our solution.

1. The principal public good is an open access platform that shows maps of symptoms, and diagnostic information at every location that HealthCube devices are installed.  

2.  HealthCube data can also be used to identify clusters of routine diseases including malaria, dengue, chikungunya and typhoid (among many others) that are useful for public health surveillance and response. 

3.  Research that will combine the large datasets generated by HealthCube along with other datasets available to CDDEP researchers on AMR (see ResistanceMap.org) and other infectious diseases can help shed greater light on risk factors for spillovers of zoonotic diseases.

4.  Although not a public good in a standard economic definition, the availability of diagnostics, and with it telehealth and remote patient monitoring in remote populations is an important public good that benefits many who currently have little or no access to formal healthcare. 

We propose to simultaneously expand access to diagnostics to the roughly 4 billion people on the planet that lack them, and help the world with understanding clusters of symptoms that can predict the next pandemic. 

The solution we propose will serve as an early warning for future pandemics and disease outbreaks. This has direct impact on (i) communities that are part of the network  (ii) communities that are not currently within the network but may be impacted and (iii) Government policies.

Communities that are being actively surveyed can be helped as soon as evidence of an outbreak becomes available. Our solution would be able to detect this sooner than existing systems, which rely on isolated data collection. Alerting the clinical community and agreeing on modes of diagnosis and treatment early can help prevent morbidity and mortality. The lack of such responses, as seen with a number of recent outbreaks, can result in serious losses within the region and globally.

Intelligent data systems like the one proposed here, can have an impact even on communities that are outside the network. Modeling based on existing data can help predict outcomes and alert communities to take appropriate precautions.

Finally, our solution will provide data to ensure that Government resources are allocated to addressing a nascent outbreak. Aligning the Government’s machinery to undertake appropriate interventions will greatly prevent further spread.

Our solution is already making a tangible impact in the lives of many who have access to ready diagnostics.  We currently serve over a million patients in eight countries using 900 HealthCube installations.  Our trajectory is as follows.

2021:  Cover another 3.2 million beneficiaries this year with 4000 HealthCubes.

By 2024: Cover 20 million beneficiaries with 20,000 HealthCubes.

For example, our work with the Tibetan population in India at the request of His Holiness, the Dalai Lama, helped identify high levels of hypertension and diabetes in that population while also provide access to testing for infectious diseases in the remote locations where Tibetans live within India. Our systems have helped identify school children at risk for anemia in Mehboobnagar district in Telengana state through mass screenings. 

Going forward, the deployment of additional systems by itself will result in greater access to diagnostics.  The data generated will then provide reliable information to populate big data models of pandemic risk and early warnings of outbreaks. 

Our progress will be indicated by our success in deploying HealthCube devices in the most widely distribute manner across geographies.  We were able to scale from 200 devices in 2018 to 600 in 2019 to 900 in 2020.  The Covid pandemic slowed down our efforts but we are already regaining momentum and expect to reach 4,000 installations this year.  Part of our success during the Covid period was to lower the cost per system from GBP 900 down to GBP 400 through re-engineering. 

In our current installations, we are able to generate data on roughly 600 patients per HealthCube.  We propose to increase this performance to 800 over the coming year and eventually to 1000.  

Converting these data into usable information to provide early warnings of disease outbreaks is the next step and will be completed under the scope of the Trinity Challenge.  We are not funded to make this transition.

We have developed a technical and usage indicators to assess progress including number of users registered and tests performed.  These provide some indication of utilization and will be collected and assessed on an ongoing basis and in real time to identify and improve any systematic issues in our technology. 

Our barriers to expansion are primarily financial and regulatory.  The commercial feasibility of entering some markets of great need may be low and the gap will have to be filled with philanthropy.  We propose to raise Series B venture capital funds to pay some of the expansion but this will likely be easier to justify for high-income countries like the US where the ability to pay is higher than in places where the need for diagnostics is great and where their value as early warning systems for future pandemics is greatest.  This is the main reason, we hope to win the Trinity Challenge - it could help support our expansion into communities where local governments and economies are unable to support diagnostics, or improved access to healthcare. 

Bill & Melinda Gates Foundation, University of Gothenburg, US National Science Foundation, Imperial College, London, ETH Zurich, University of California, Berkeley, University of Virginia, Princeton University, University of Kwazulu-Natal, South Africa, University of Strathclyde, TechMahindra, Reliance Jio, Apollo Telehealth.

Although CDDEP and HealthCubed are adequately funded to carry out their routine operations, building a platform for risk identification and early warnings of future pandemics will require some dedicated investments on three fronts that we are unable to obtain from our existing funders.

1. Expand HealthCube installations to more areas of projected pandemic risk, particularly in South-east Asia.

2. Tie the data from HealthCube systems to underlying risk data and other data sources including laboratory and sewage surveillance data from partner organizations. 

3.  Support an online platform that can provide the information on risk and early warnings in real-time back to policymakers and other data users.  

Although we work with many of the Trinity Challenge partners already on various other projects, connecting with them in the specific context of the objective of identifying, delaying and mitigating future pandemics would be helpful to our collective goal. 

We would like to partner with the following organizations (and have already been in contact with them but not specifically in the context of pandemic surveillance)

Bill & Melinda Gates Foundation - Both CDDEP and HealthCubed are grantees for various project but not specifically for the pandemic preparedness challenge)

Facebook and Google - We would like to work with Facebook and Google on supplementing our risk maps with search and social media data that could help with early identification of outbreaks. CDDEP is already part of an NSF-funded research consortium that is targeted to infectious disease computational modeling in which Google is a partner. 

CHAI - We have been discussing how CHAI in India could help us scale up the deployment of HealthCube diagnostic systems in remote districts.

Imperial College - We already work with infectious disease modelers at Imperial in Neil Ferguson's team but not specifically in the context of pandemics.  We would like to find ways to expand that collaboration.