Additional file 9 of Quantifying the potential value of antigen-detection rapid diagnostic tests for COVID-19: a modelling analysis

Additional file 9: Figure S6. Sensitivity analysis to patient behaviour in relation to self-isolation, in the community setting. As a focal model output, all figures show the proportion of simulations in which an Ag-RDT was favourable, with different algorithms labelled by the different line colours. Panel A shows the impact of compliance amongst those required to self-isolate after a positive final test result. Panel B shows the impact of test-negative individuals voluntarily self-isolating. This sensitivity analysis was restricted to the community setting as it is likely that hospitals will enforce compliance to isolation guidelines. Results illustrate that increasing the proportion of compliance to isolation recommendations increased the favourability of both “Ag-RDT-only” and “confirm Ag-RDT negative” strategies, from 86% and 68% of simulations being favourable with 50% compliance to 98% and 80% with 100% compliance, respectively. The benefit of an Ag-RDT test in rapidly detecting COVID cases, and hence averting onward transmission, is reduced if these individuals did not isolate. However, the opposite was seen with the “confirm Ag-RDT positive” strategy, with the favourability of the algorithm decreasing from 8% to 0% if compliance doubled from 50% to 100%. Generally, this strategy detected fewer COVID cases than a NAT-based strategy, due to the reduction in overall sensitivity caused by inclusion of NAT confirmation; thus, increasing the proportion of individuals that did comply had a greater effect on a NAT-based strategy than the Ag-RDT strategy, hence increasing the latter’s favourability. Similar results were seen for voluntary self-isolation (where false negatives voluntarily self-isolate).