Can bidirectional contact tracing catch more COVID-19 cases?

By: Madeline Ward, University of Calgary

Throughout the COVID-19 pandemic, public health agencies have used contact tracing has been used to identify people who have been in close proximity to those who are known to be infectious. In Alberta, the standard for contact tracing has been to trace forward – meaning the goal is to identify any infections that could have been caused by a person who received a positive test result. When a person receives a positive COVID-19 test, they are asked to identify anyone who they came into close contact with between 48 hours before and 10 days after their symptoms appeared (or from when they received the positive test if they didn’t have symptoms). The contacts they identify are then asked to get tested and isolate themselves. However, a recent study published in Nature Communications this January suggests that performing “bidirectional” contact tracing, which aims to identify how the person who tested positive was infected, could be more than doubly as effective as only tracing forward.

In bidirectional contact tracing, reverse (or backward) tracing is done along with the standard forward tracing. If Patient A receives a positive test, reverse tracing means finding the source of Patient A’s infection, and, say they got it from Patient B, finding all the other contacts of Patient B who could have potentially been infected as well.

The authors of the study compared the effects of forward tracing versus bidirectional tracing using a metric called the effective reproduction number, or Re, which is the average number of new infections that one infected person will cause at some given point in time, taking into account factors like the state of vaccination and the levels of different variants . A lower Re means that there is less COVID-19 spread occurring; when Re falls below 1, this is a good indication that case counts are, or will start, falling.

The researchers found that when the current standard time frame (48 hours before symptoms or a positive test, and 10 days after) for contact tracing is used, bidirectional tracing can reduce Re by as much as 0.24 compared to forward tracing. In other words, there would be approximately one less case resulting from every four people with infections. If the tracing time frame window is extended so that contact tracers look at contacts within six days before first symptoms or positive test, the Re could potentially be reduced by another 0.42 – which is 2.75 times more effective compared to only tracing forward.

It’s important to note that the results of this study were obtained using simulations with a variety of assumptions (for example, how many cases are from contaminated surfaces or other environmental sources as opposed to from close contact with a contagious person?). This means the improvements seen in the study might not translate perfectly into real world contact tracing. However, in this study the authors tested their simulations under several different assumptions for how COVID-19 is spreading and found that bidirectional contact tracing was consistently better at slowing disease spread than forward tracing. This is a good indication that bidirectional contact tracing could be a good thing to test out in real-world scenarios.

So why is bidirectional contact tracing not already the standard? Unfortunately, reasonable to public health units may not be able to perform this degree of contact tracing all the time, especially during spikes in cases. During the second wave of COVID in Alberta from mid-October 2020 to mid-January 2021 for example, Alberta Health Services experienced a huge backlog of contact tracing cases.

Perhaps now as increasing rates of vaccination bring cases down to a steady and manageable number, more resources can be directed to bidirectional contact tracing. And as people who have been working or learning from home start to return to offices and schools, workplaces should keep in mind that tracing forward might not be sufficient for identifying people affected during any outbreaks.

Edited by B.G. Borowiec and A.E. McDonald. Header photo from Unsplash.