Contact tracing: Key to fall reopening for colleges and universities
Contact tracing has been a cornerstone of epidemiology for over a century and is one of a number of tools and solutions colleges and universities are exploring to help safely and confidently reopen campuses in the fall.
While public health officials remain hard at work in their contact tracing efforts, colleges and universities like ours, Oklahoma State University, have a much more closely-knit population. We have access to data sources that allow us to work faster and gain better understanding of where and how the virus has spread. Ultimately when it comes to contact tracing, speed saves lives, which is why universities like ours are investing in an analytics-driven approach.
A unique contact tracing opportunity
Due to their contained nature and access to many data sources, colleges and universities can more easily create direct links between and among students, faculty members and staff. Information on dormitory and class assignments, and other data from buildings and offices help us better understand who is in direct contact.
Those data sources can be pre-populated into contact tracing databases so that when a person tests positive for COVID-19, campus and public health officials can easily see where that person has been recently, and who has spent meaningful time in the same space.
Additional data sources can create inferred linkages to determine roughly where students or faculty members have been. In addition to helping find others that may have been exposed, the data can also identify hot spots within a dorm, building or even find “super spreaders”—individuals who are linked to a large number of cases or a more risky group of students.
Contact tracing at Oklahoma State
We implemented contact tracing using SAS data analytics tools that allow us to not only integrate, clean and maintain data, but also to create data visualizations and reports for health officials.
To be ready for the return of students, faculty and staff this semester, we integrated a variety of data sources that can create both tangible and inferred links between people and places, and show times these individuals may have interacted. We connected institutional time and location-based data from sources like card swipes, campus store purchases, class schedules and more than 5,100 campus Wi-Fi access points, including in Greek housing.
If someone tests positive, we can see on a map where they have been in the previous 48 hours and who they may have spent more than 10 minutes with. The students are represented by their student ID numbers and information is only shared with authorized health officials who will then follow up with the affected people.
If data shows a person spent more than 10 minutes with the infected individual, we take another step to reduce false positives. For instance, let’s say the infected student spent time in the library. If another student used the library WiFi during the same time period, they might need to be contacted. However, what if they logged on then left to get coffee? If we see they swiped their ID in another building during that time we know they were not in the library consistently and we may be able to eliminate them from the contact tracing process.
The data also helps us identify high traffic areas, places in need of cleaning, and where social distancing should be monitored to ensure students and faculty are being safe.
Contact tracing is and will continue to be critical to Oklahoma State University’s re-opening strategy. As colleges and universities across the nation finalize their plans for the Fall 2020 semester and beyond, we hope they consider these data sources and strategies as tools to help inform policies to ensure a safe and effective learning environment.
Christie Hawkins is associate vice president for administration and finance, and the director of institutional research and analytics, at Oklahoma State University. Larry Burns is assistant director of institutional research and analytics at Oklahoma State.