Bidirectional contact tracing could be pandemic game-changer

Bidirectional contact tracing, wherein both parent and subsequent cases of a known infected patient, could help drastically improve control of the coronavirus disease 2019 (COVID-19) pandemic, a recent study reports.

Using a stochastic branching-process model, the researchers extended a current forward-tracing COVID-19 protocol to assess the efficacy of bidirectional tracing. The model had the following initial assumptions: 10 percent of transmission was environmental and untraceable, while 48 percent occurred presymptomatically; 45 percent of cases were asymptomatic; and the virus had 50-percent infectiousness. The basic reproduction number was initially set at 2.5.

The baseline protocol involved manual forward tracing and isolation of people who had come into contact with the known case 48 hours before symptom onset or diagnosis. In this scenario, the effective reproduction number (R_eff) was predicted to drop by as much as 0.24 relative to the no-tracing scenario.

The adoption of manual bidirectional tracing, without changing the 48-hour window, further reduced the R_eff by another 0.24, effectively doubling the efficacy of the forward-tracing protocol.

In addition, extending the bidirectional tracing window to 6 days before symptom onset led to another 0.42-unit drop in R_eff, representing an approximately 85-percent improvement relative to the 2-day window standard. Compared to the forward-tracing-only protocol with a 2-day window, this new protocol was 275-percent more effective.

Notably, the researchers also found that implementing a digital tracing system, with instantaneous notifications for potential contacts, could also effectively stem the tide of the pandemic even without manual tracing.

However, such an approach was highly sensitive to network fragmentation: the digital-only protocol worked when all cases opted in, and even small decreases in the population of participating individuals substantially weakened its efficacy.