On the use of LFA tests in contact tracing: preliminary findings

Rationale We compare two policies for the efficacy of Test, Trace and Isolate (TTI): 1. Status quo: isolation of symptomatic individuals for 10 days post symptom onset and quarantine of 14 days for within-household contacts and tracing of contacts with 14 day post last exposure for outside of household contacts. 2. Daily Lateral Flow Assay (LFA) testing of all contacts with no quarantine for up to 14 days of negative tests; 10 days self-isolation for contacts testing positive. Contact tracing is initiated on a positive test result. If contacts of a confirmed case do not quarantine and instead test regularly, there is a risk that they will transmit infection if their infection is not picked up by testing. On the other hand, the current system of only testing contacts who become symptomatic will miss contacts who are asymptomatic or do not report symptoms; thus the risk might be offset in gains via identifying more index cases for tracing. We use a household structured branching process model of infection and contact tracing to investigate. Main points • Daily LFA testing of traced contacts may offer an improvement over the current contact tracing strategy. • These are preliminary results. A full sensitivity analysis has not been completed and many of the parameters have significant uncertainty. In particular, we were not able to have uncertainty in the test sensitivity curves. • To make the strategies comparable for this analysis, we model uptake and adherence to self-isolation and quarantine (status quo policy) and daily LFA testing of 100%. This is not a realistic assumption in practice. Assumptions about the proposed daily LFAT testing of contacts policy: • If an individual has symptoms, and is not being LFA tested, then they must isolate and request a PCR test. • When an individual tests positive (through either LFA or PCR), an individual initiates contact tracing for their recent contacts. Their household members begin LFA testing if they haven't already.