Contact tracing is used by government health agencies to identify people who may have been exposed to health threats, such as the new coronavirus, either through direct contact with a person with COVID-19 or with someone who has been in close proximity to that person.
Investigators attempt to quickly identify and notify all potential contacts, break the chain of transmission if possible, provide medical evaluations when needed and quarantine contacts if they are (or become) symptomatic. Another outcome of contact tracing is to influence public health policies through the study of contamination chains.
The goal of contact tracing is diminishing the spread and duration of an epidemic or pandemic. With the COVID-19 pandemic, contact tracing will hopefully have the follow-on effect of helping governments know when it's safe to begin reopening society and to quickly identify and alert people to the inevitable reemergence of outbreaks.
Contact tracing webinar
In this webinar, you will learn about using contact tracing to identify the points of transmission and slow community spread.
Contact tracing sounds straightforward, but it can be overwhelmingly complex and resource consuming. And how can technology help? It's a highly manual process, and while it works well, technology can help improve it and complement it through entity resolution, case management, alert automation and network analytics.
Though contact tracing might seem like a relatively new concept to the general public, this practice has been used by health officials for decades as an attempt to stave off a variety of communicable diseases by painstakingly identifying the points of transmission and slowing community spread.
In fact, the state of New York recently unveiled its contact tracing plan that calls for the hiring of as many as 17,000 contact tracers to help in the fight against the coronavirus pandemic. And France is hiring large brigades of contact tracers. Belgium is hiring 2,000 “corona detectives” for a population of about 11.5 million people. Not all of these tracers are in the field. Some will be at call centers, others will work with incoming data and some will be public health experts.
These modern contact tracing efforts can be enhanced using data visualization and analytics to assist public health officials in uncovering insights from contact tracing efforts and publicly available health data to understand:
- Missing or unexpected linkages in contact data.
- Who should be tested.
- Where the virus is spreading.
- Which communities are at greatest risk.
This increased awareness is a valuable tool that enables epidemiologists and health care officials to respond faster in instituting containment measures and issuing public alerts to COVID-19 “hot zones.”
Thanks to advanced analytics and data visualization, public health officials and investigators are finding new channels to quickly identify people who have been exposed to COVID-19 so they can self-isolate, seek treatment if needed and impede the spread of infection.
Contact tracing: A history of disease tracking
One of the most famous cases of contact tracing is that of “Typhoid Mary” in New York City at the beginning of the 20th century. The source of contagion turned out to be an asymptomatic carrier who was a roving household cook for a number of the city's wealthy families. Only through the scientific and diligent work of George Soper (considered by many to be the first epidemiologist) were authorities able to not only identify the source but also realize that some people could carry and pass on the disease while appearing to be healthy.
Soper's detective work, interviewing typhoid patients and their families, became the foundation for modern contact tracing. Much of the work has remained unchanged and contact tracing has become a preventive medicine cornerstone. From AIDS to Ebola, contact tracing is instrumental in fighting disease outbreaks all over the world.
Contact tracing in today’s world
A journalist’s standard set of questions – who, what, when, where and why – are the same that a contact tracer uses when interviewing someone who has been diagnosed with a contagious disease, like COVID-19. They seek information directly from the source to help determine who the individual came in contact with, when these interactions happened and where they took place. Ultimately, the goal is to establish linkages to help slow the spread of the virus. But that’s easier said than done. What happens if patients can’t recall all of their movements over the last two weeks? What if a patient is incapacitated and can’t respond to questions? People who were unknowingly exposed to the infected individual might carry on with their lives, unaware that they, too, could be infected and further spread the disease. This is precisely where modern technology can play a pivotal role.
A once cumbersome process that relied on an individual’s often incomplete or inaccurate memory, contact tracing has entered the digital era. Thanks to advanced analytics and data visualization, public health officials and investigators are finding new channels to quickly identify (with their consent) people who have been exposed to COVID-19 so they can self-isolate, seek treatment if needed and impede the spread of infection.
Contact tracing: The new role of data management and analytics
The latest data management capabilities and cutting-edge analytics give contact tracers a new set of tools to work with. There are four areas where analytics can help.
Contact transaction databases with entity resolution
These databases are the home for contact tracing data. They enable entity resolution that, for example, links multiple records in perhaps multiple databases to the same person. This is a huge timesaver when speed is important (as it is with contact tracing). Using data management solutions means you can establish (and display) linkages between patients, their contacts and the places they might frequent to narrow and focus the efforts of contact tracing resources. By combining disparate data sources into a single repository that can be updated in real time, data scientists can gain insights into how linkages form and change over time. They can also suggest which contacts in a patient’s network have more current relevance and how wide their geographical territory might be.
Enriched contact tracing data
Using analytics and machine learning gives health officials new and important insights into each patient’s network. Analytics can help determine who might be directly linked to a patient – such as family, friends and neighbors – but also employer rosters, passenger manifests and school rosters. Machine learning can help automate that effort by building analytical models that reflect real-world data and conditions.
Once the contact tracing team has sufficiently identified a patient’s links and network, it can begin the task of automatic notification of people in that network via text/SMS messages or emails. Based on the likelihood of close, extended contact, such as a work colleague, they may be notified to get a health assessment. Or in the case of a restaurant, the staff may be notified that someone with the virus was a patron so that the restaurant can do a thorough cleaning. Sometimes for large groups, it may be more feasible to send an alert and ask people to contact their physician or the health department for screening.
Public health insights
When assessing a patient’s network, public officials can rely on analytical insights from the data to fill in some of the gaps to answer questions such as:
- Who should be tested?
- Who is most likely to spread the virus?
- How do I find missing or unknown linkages?
- Which communities are at greatest risk?
- Is social distancing working?
The current pandemic has exposed weaknesses in the traditional approach to contact tracing. The swiftness of the COVID-19 outbreak has overstressed the public health infrastructure and has underlined the need for innovative approaches to increase contact tracing capabilities.
How we can help
With the COVID-19 pandemic affecting societies around the world, public health officials from different regions have varying needs and approaches when it comes to contact tracing.
For example, analytics combines robust link analysis and visualization with text and geospatial search and analysis, interactive network building, entity generation and contact analysis. These capabilities enable health officials and investigators to proactively identify risk contacts and superspreaders.
Analytics software can integrate data from a variety of external sources to deploy the right data quickly to our cloud-ready investigation and incident management solution. From there, users can easily create, triage and manage their efforts to make contact tracing more complete.
Advanced analytical modeling tools help health officials and governments answer the critical questions needed to implement smart public health policies. With data visualization abilities, users can perform deeper investigations of contacts and data to uncover hidden patterns and share them across various health agencies.
Analytics can help governments establish contact tracing databases where public health workers can enter contact tracing data in real time. Additionally, analytics can enrich the data collected from contact tracing interviews by establishing more comprehensive linkages using direct link data, inferred link data and communication method data. These links can be visually displayed and analyzed over time.
Once links have been determined, you can generate alerts that public health officials can dispatch to communicate with contacts for a given patient. These alerts can convey health risk warnings and can be customized for each recipient, such as directing them to obtain a COVID-19 test at a specific facility or to self-quarantine for a specific number of days. They can be sent via automated channels, such as text/SMS messages and emails.
Contact tracing and privacy
Contact tracing gathers data that can often be considered protected data under existing privacy laws, such as name, address, phone number, email address and, of course, health condition.
A discussion of contact tracing is not complete without at least touching on the sensitive and real privacy concerns we all have. And the level of intrusion into a patient’s privacy (and that of the patient's network) varies from country to country. Public health officials can again turn to analytics – this time, to help protect privacy. Data scientists can use techniques such as masking, de-identification and role-based security to maintain a healthy balance between privacy and good public health.
It’s important to be aware, however, that some data privacy laws govern, but do not prohibit, contact tracing. And most data privacy laws allow for sharing of data during public health emergencies, such as pandemics.
Debates will, and should, continue about the legal and ethical challenges of the tradeoff between the public welfare and individual privacy, but without access to as much contact data as possible, contact tracing efforts will be as difficult now as in the 1900s.
- Unemployment fraud meets analytics: Battle lines are clearly drawnMany fraudsters seized opportunities presented by the COVID-19 pandemic. During the crisis, unemployment fraud became a battleground between international criminal networks and government agencies. Learn how analytics can save billions – and deliver benefits to those truly in need.
- Jump-start COVID-19 research with text analytics Applying text analytics and AI to discover trends and commonalities among COVID-19 research documents can lead to new insights about the disease and help answer many unanswered questions.
- AI in government: The path to adoption and deploymentThe government sector is lagging in AI adoption, but awareness of the importance of AI in the public sector is increasing. Our survey indicates that operational issues are requiring governments to turn their attention to AI projects as a way to address important public issues.