Natural pandemics have historically killed millions of people and caused widespread suffering.
Advances in biotechnology have opened up the possibility of novel diseases being accidentally or deliberately released, and these could cause more harm than natural diseases.
Biosurveillance as a possible solution
The field of biosurveillance aims to help with detecting and tracking outbreaks of disease and is one component of improving biosecurity. It has two aims:
- Early event detection – detect as early as possible events that suggest a public health emergency
- Situational awareness – to help response to outbreak and monitor it’s progress
What you could build
The parts of a typical biosurveillance system
To work out what you might build, it’s helpful to see the parts of a typical biosurveillance system:
We’ll explore what you could build for each part.
Syndromic surveillance systems look for leading indicators of a disease outbreak. Unlike traditional health surveillance, which relies on confirmed cases of a disease, such a system may, for example, look for an increase in individuals presenting with sore-throats in a particular area. In theory, this can help spot an outbreak before the traditional methods do.
Data from medical records
Data from hospital medical records could be pulled into a more centralised public health monitoring system at local, national, or international levels. This could be used to detect changes in patterns of diseases that are diagnosed in the records. Alternatively, the symptoms mentioned in the records could be analysed to detect misdiagnoses of rare diseases that could have a big impact on public health. As Alex Demarsh says “One issue biosurveillance people think about a lot is the fact that doctors are told ‘When you see hoofprints think horses not zebras’. But in biosurveillance we’re looking for those zebras, unusual cases that could be the beginning of something bigger.”
However, it could be very difficult to access this data both because of regulations for patient privacy and because the IT systems that hospitals use may be very difficult to work with.
It’s possible to gather data from individuals about health around them. Flu near you is an example of this. If you build a system that can consistently get good data in different regions, you could then use it to enhance surveillance in high-risk areas or during an emergency. For example, the WHO helped set up an SMS-based surveillance system for use during and after typhoons.
Scraping data from online news sites and social media can catch outbreaks before traditional methods. An example of this is the GPHIN system, which proved useful during the SARS outbreak. According to “GPHIN’s information was ≈2–3 days ahead of the official WHO report of confirmed and probable cases worldwide.”
Data on factors that can affect spread
For example, human movement data can help with predicting the spread of many diseases; weather and landscape data can help when studying vector-borne diseases. For example, article about “Environmental Biosurveillance for Epidemic Prediction Experience with Rift Valley Fever” describes a system designed to warn of climate conditions favourable for an outbreak of the mosquito-born Rift Valley fever. As another example, much of Bluedot’s work is on the spread of diseases (such as flu, ebola, and dengue) via global airline transport.
Connect together multiple data sources
Alex Demarsh says “We need interoperable systems that can talk to one another. We need to be able to link together e.g. survey data, administrative data, environmental data, road networks, and maybe vector populations and movements. One thing that makes this difficult is that data is often hard to move between administrative units (such as states or provinces). This data infrastructure would help us to predict the spread of disease and would support decision making on how to handle an outbreak.”
Detecting an initial outbreak of disease is not enough. You also need to be able to judge whether a report indicates an outbreak, and if there’s an outbreak whether it’s likely to be important. Historically, predicting the importance of an outbreak has been very difficult.
Better tools for communicating to decision-makers
Alex Demarsh mentions that “It would also be good to make it easier to communicate results of surveillance an analysis to non-technical and non-scientific people who have to make decisions. We need better tools for communicating data effectively to people in government who make decisions on how to respond. Often agencies have to search manually through data when there would be ways to make that process a lot quicker using better software”
Which diseases should you concentrate on?
To make progress, you’ll need to start by working on a disease that has outbreaks often enough that you can test your system. Ideally you’d also work with a disease that is globally important, which will often mean diseases that are most prevalent in the developing world. See Chan et al. 2010 for data on outbreak frequency in different regions.
One good option would be flu. According to Alex Demarsh “In general, influenza is a good disease to work on as a lot of people die of it each year, there’s a risk of an influenza pandemic, and there’s a lot of data on it.”
Alternatively, instead of working on detecting outbreaks of currently important diseases, you could work on detecting new diseases. According to this paper “The majority of all human infectious diseases and pandemics have originated through the cross-species transmission of microorganisms from animals to humans, overwhelmingly in the Old World.” An example of this is HIV – according to the same paper “It is now generally accepted that the hunting and butchering of wild nonhuman primates in the early 20th century led to the introduction of simian immunodeficiency virus into the human population, giving rise to our modern day HIV pandemic.” If we had detected the early spread of HIV or the even earlier transmission of simian immunodeficiency virus into humans we would have been decades earlier than the discovery of AIDS in the 80s. In TED talk is a useful introduction to this issue.
We’re not sure how easy it is to build a viable business in this area. There are existing companies such as Epidemico, BlueDot, and Metabiota but it’s unclear from reading their websites how their business model works. A good next step would be to reach out to these companies and talk to them about their businesses and where they think there are gaps.
Evaluation of potential impact
Based on the Open Philanthropy Project’s research on the topic, we think that the biosecurity and pandemic preparedness is an important cause area. However, we’re unsure of how effective biosurveillance is at tackling this and have not looked into this question in depth. A discussion of the effectiveness of biosurveillance can be found here http://faculty.nps.edu/rdfricke/docs/Encyclopedia%20Article.pdf#page=7.
As well as looking at empirical work on the impact of biosurveillance, you could look at existing modelling of interventions to prevent disease spread to see if biosurveillance systems could affect some of the model parameters.
Others working on this
- Governmental and intergovernmental organisations such as the WHO, CDC, and the Public Health Agency of Canada
- Companies such as Epidemico, BlueDot, and Metabiota
- Academics such as in Harvard’s Department of Biomedical Informatics, McGill’s Surveillance lab, UC Davis’ Infectious Disease Ecology lab, and the University of Pittsburgh’s Real-Time Outbreak and Disease Surveillance system
- Nonprofits such as Global Viral and EcoHealth Alliance
- Open Philanthropy Project’s cause report on biosecurity, including the conversation notes linked to in the sources section http://www.openphilanthropy.org/research/cause-reports/biosecurity#sources
- 80,000 Hours problem profile on biosecurity https://80000hours.org/problem-profiles/biosecurity/
- This encyclopedia article on biosurveillance http://faculty.nps.edu/rdfricke/docs/Encyclopedia%20Article.pdf#page=7
- The Handbook of Biosurveillance http://www.amazon.co.uk/Handbook-Biosurveillance-Michael-M-Wagner/dp/0123693780/
Beckstead, Nick, Niel Bowerman, Owen Cotton-Barratt, William MacAskill, Seán Ó hÉigeartaigh, and Toby Ord. “Unprecedented Technological Risks” Global Priorities Project, 2014. Accessed August 17, 2016.
Chan, Emily H., Timothy F. Brewer, Lawrence C. Madoff, Marjorie P. Pollack, Amy L. Sonricker, Mikaela Keller, Clark C. Freifeld, Michael Blench, Abla Mawudeku, and John S. Brownstein. “Global Capacity for Emerging Infectious Disease Detection.” Proceedings of the National Academy of Sciences 107, no. 50 (December 14, 2010): 21701–6. doi:10.1073/pnas.1006219107.
Fricker, Ronald D. “Biosurveillance: detecting, tracking, and mitigating the effects of natural disease and bioterrorism.” Wiley Encyclopedia of Operations Research and Management Science (2011).
Keller, Mikaela, Michael Blench, Herman Tolentino, Clark C. Freifeld, Kenneth D. Mandl, Abla Mawudeku, Gunther Eysenbach, and John S. Brownstein. “Use of Unstructured Event-Based Reports for Global Infectious Disease Surveillance.” Emerging Infectious Diseases 15, no. 5 (May 2009): 689–95. doi:10.3201/eid1505.081114.
Nouri, Ali, and Christopher F. Chyba. “Biotechnology and biosecurity.” Global catastrophic risks 1 (2008): 444.
Pike, Brian L., Karen E. Saylors, Joseph N. Fair, Matthew LeBreton, Ubald Tamoufe, Cyrille F. Djoko, Anne W. Rimoin, and Nathan D. Wolfe. “The Origin and Prevention of Pandemics.” Clinical Infectious Diseases 50, no. 12 (June 15, 2010): 1636–40. doi:10.1086/652860.
Silver, Nate. The signal and the noise: Why so many predictions fail-but some don’t. Penguin.