Ed Note. This item originally appeared in TechPresident and is reprinted with permission.
BY REBECCA CHAO
Every year, millions of Muslims from around the world embark on the Hajj, a spiritual rite through the dusty granite hills of Mecca in Saudi Arabia that forms one of Islam’s five pillars of faith. This year, the fatal Middle East Respiratory Syndrome (MERS) looms over the pilgrimage, which will take place in mid-October, leaving public health officials worried they may see another pandemic, as foreign pilgrims now account for more than half of Hajj participants and could bring the virus back home.
A few cases of this fatal disease — which has so far killed about half of those infected — have already emerged in Europe but it has not yet made its way to North America. Health officials in the U.S. are trying to keep that number to zero with the help of a new digital disease tracking tool called BioMosaic.
A Collaborative Effort
BioMosaic is an idea of Dr. Martin Cetron, the director of the Division of Global Migration and Quarantine at the Centers for Disease Control and Prevention (CDC) and a member of the World Health Organization’s (WHO) emergency committee concerning MERS. BioMosaic was an idea that began in 2010, says Dr. Cetron, but it has taken the last few years of collaboration to conceive and design. At this time, BioMosaic is an internal program used only by the CDC.
While there are now a number of health mapping tools what differentiates BioMosaic from others, explains Dr. Cetron, is its comprehensiveness. BioMosaic creates a visualization of the complicated interplay between the “holy trinity of infectious disease” – as Cetron calls it – the bug itself, the host and the environment. The platform not only tracks the disease’s movement – through people, animals, or insects – but also aids in understanding the unique formula that triggers pandemics.
To monitor disease, BioMosiac utilizes demographic information, as well as data extracted from two existing health-mapping tools. One, HealthMap, was was created by a team at Boston Children’s Hospital in 2006 and is overseen by Dr. John Brownstein, a professor at Harvard Medical School. It tracks reports of diseases from social media, newspapers, reports, publicly available information, as well as self-reported outbreaks through tools like Outbreaks Near Me, also created by the HealthMap team.
The other, BioDiaspora, follows human movement around the world and was created in 2008 by Dr. Kamran Khan, an infectious disease specialist and researcher at St. Michael’s Hospital in Toronto. It uses a trove of private data from around 4,000 airports – roughly 30 million flights a year – which took years of collaboration with governments, airlines and air traffic controllers to collect.
The BioMosaic tool. (image: Centers for Disease Control and Prevention)
Monitoring MERS
With MERS, BioMosaic enables U.S. health officials to analyze and visualize data — anonymized for privacy — about passengers flying between the U.S. and Saudi Arabia and other countries along the Arabian Peninsula. They can detect travel patterns to show which airports and cities received the largest number of travelers from Saudi Arabia and take necessary precautions to prevent the spread of MERS into the U.S.
Since the virus first emerged in June 2012 in Saudi Arabia, it has already appeared in Germany, France and the United Kingdom but has not yet reached the U.S. Of the 91 cases worldwide, only eight occurred outside of the Middle East and North Africa region, but half of those were fatal.
Dr. Khan recently completed a study of historic flight patterns in and out of Saudi Arabia during previous pilgrimages, examining peak travel periods and countries of origin.
According to the study, eight countries will most likely account for more than 50 percent of flights out of affected Middle Eastern countries and should be on their guard this year: India, Egypt, Pakistan, the United Kingdom, Kuwait, Bangladesh, Iran and Bahrain. These countries have “significant potential for MERS-CoV introduction via commercial flights,” writes Dr. Khan.
The report also discovered an important characteristic of these travelers. It notes that “two-thirds of visitors will be returning to low or lower-middle income countries where medical and public health capacity will be limited, and presumably where the risk of domestic transmission of imported MERS-CoV will be elevated.”
Many Middle Eastern countries also hold large populations of foreign workers – 28.7 percent are foreign-born in Saudi Arabia – increasing the risk of spreading MERS to other countries. Many of the foreigners in this region are low-income migrant workers with inadequate access to healthcare both in their host and home countries. These migrants, at high-risk for catching and spreading the disease, are also often overlooked by government disease prevention initiatives.
In an effort to prevent the spread of MERS, Saudi Arabia has already reduced foreign Hajj participants by 20 percent and those traveling from within Saudi Arabia by 50 percent. The country is also scrambling to discover the origins of the disease.
MERS, a relative of SARS, is believed to have also originated in an animal. So far, BioMosaic has incorporated information to help pinpoint the disease’s animal origins, such as bird migration, location of mosquitoes and even poultry distribution. It can analyze demographics and geographical information of the “host” country to see whether human population, rainfall, temperatures, types of animals and other environmental conditions have deviated significantly from historic averages. The information is particularly pertinent for the Hajj since millions of animals are sacrificed at the end of the Hajj and are eventually consumed by pilgrims or handed out to the poor, increasing the likeliness of animal to human passage of the disease.
From SARS to MERS
When SARS broke out in China 10 years ago, it touched every country in the world, infecting around 8,000 and killing over 900. Technology played a major role in both its spread and its containment.
After the SARS epidemic in 2003, studies revealed just how air travel could lead to the spread of disease. Cities with direct flights from Hong Kong were 25 times more likely to report incidences of SARS than cities with connecting flights. In fact, cities that were two flight connections away from Hong Kong had no cases of SARS.
“If international connection through air travel helped spread infection, digital connections–local and international–helped spread the ideas required to fight it,” wrote Ethan Zuckerman in his book, Rewire: Digital Cosmopolitans in the Age of Connection.
He noted that in March 2003, as the disease began to pick up speed, the WHO created a website that allowed researchers and disease specialists from around the world to hold videoconferences. They could share lung x-rays of infected patients and create a procedure for diagnosing the disease. “The alerts proved remarkably effective–90 percent of all the SARS cases occurred before the WHO’s advisories were issued,” wrote Zuckerman.
The WHO also used the Global Public Health Intelligence Network (GPHIN), a program created by the Canadian government, that collects media alerts and mentions of SARS on the Internet. “More than one-third of the rumors identified by GPHIN led the WHO to identify and isolate cases of SARS,” wrote Zuckerman.
Since 2003, dozens of health mapping apps have been created, speeding up the way in which disease is reported.
“CDC is the gold standard,” said Sumiko Mekaru, an epidemiologist who works with Dr. Brownstein, managing several HealthMap projects. But since there is a significant delay in getting reports out – data has to go from the local to the state to the national level with lots of other steps in between – digital tools can provide an advance warning.
“The Internet enables us to find epidemics when they are smaller in size, when there are 20 cases instead of a thousand,” said Dr. Cetron. “This allows investigation and preventive tools and minimizes the impact of epidemics.”
In fact, worldwide, the lag between a reported outbreak and an official announcement has been shrinking for some time.
According to one study analyzing WHO data, the median time from outbreak start to outbreak discovery and to public communication about the outbreak has decreased. The lag dropped from 40 days in 1996 to 19 days in 2009. In terms of discovering the outbreak, lag time decreased from 29.5 days in 1996 to 13.5 in 2009. The differences changed depending on region, with shortest delays in Southeast Asia and the Western Pacific.
Digital Limits
While filling information gaps is vital to stem the spread of disease, disease and information spread in different ways.
According to one report, when a disease first emerges, there is a sudden surge of interest and social media activity. As the disease progresses, interest wanes, the amount of information on the disease decreases, and the effectiveness of digital tools also goes down. In particular, for newly emerging diseases, the report notes that there is an initial surge of media reporting rather than first-person accounts via social media. As a result, it makes it difficult for the tool to provide data on early detection of these emerging diseases. This can be mitigated by human analysis, which can provide context to the data.
Another challenge is that without proper context, digital tools can often produce misleading results. Google Flu was a case in point, grossly overestimating the percentage of Americans with flu this year, as Miranda Neubauer reported in February.
Mekaru notes, however, that not all wrong information is bad information.
Outbreaks Near Me is one tool that has shown how self-reporting, even when flawed, can help with preventing the spread of diseases.
Outbreaks Near Me allows self-reporting of symptoms with yellow pins denoting lower likeliness of disease and red pins a higher one. The platform revealed an outbreak of Syphilis in a town in New York. It turned out that many users were misdiagnosing themselves and reporting the wrong disease. “The County Health department quickly responded to our query and said they were actually seeing an increase in chlamydia cases,” explained Mekaru. “The community appeared to be mixed up about which STD was circulating.” This mix-up was actually useful for public health officials who can then correct misinformation and get the public what it needs.
Another challenge in using these digital tools, however, is that there can be “a lot of noise in the system,” said Dr. Cetron. “We need to understand all the factors of why epidemics don’t occur even when the conditions are ripe.” Such scenarios could raise false alarms and waste resources and reduce credibility in a method or a tool.
As health experts try to outpace disease with increasingly sophisticated tools, there are problems that even the best technology cannot solve. “The enemy of knowledge is not ignorance but the illusion of knowledge,” said Dr. Cetron, quoting Stephen Hawking. “You might think there is causality when all you are seeing are associations.”
The key to avoiding those mistakes, Dr. Cetron advised, is good old fashioned humble pie – looking honestly at what conclusions can be drawn from the glut of big data.