Submission ID 132046
| Issue/Objective | Effective pandemic response depends on rapid diagnosis, timely action, and equitable access to care. In resource-constrained settings, delays in diagnostic turnaround propagate downstream inefficiencies across health systems. Digital health strategies may reorganize diagnostic-care pathways by accelerating testing and shifting entry points closer to communities, yet robust randomized evidence from equity-seeking communities remains sparse. |
|---|---|
| Methodology/Approach | We conducted a superiority, open-label, single-blinded, parallel-arm randomized controlled trial (RCT; ClinicalTrials.gov identifier: 174921) in township populations recruited from community clinics in Cape Town, South Africa. Consenting eligible adults (N=605) were randomized 1:1 to a digitally supported rapid self-testing strategy (COVIDSmart CARE! program) or conventional facility-based testing. The primary outcome was turnaround time (TAT) from testing to actionable test result. Secondary outcomes included the proportion of participants linked to action plans and detection of new infections by RT-PCR. Tertiary outcomes included diagnostic accuracy of the self-test against RT-PCR. An accelerated failure time model with Weibull distribution was used for the primary analysis. |
| Results | Of 605 participants screened, 589 were included in the primary analysis (285 digital self-testing; 304 conventional). Median TAT was 1.98 hours in the digital arm versus 53.23 hours in the conventional arm (p < 2 × 10-16). In the adjusted accelerated failure time model, the digital arm achieved a TAT that was approximately 2.34% (95% CI: 2.21-2.49%) of the conventional arm. Over 99% of participants in both arms were linked to appropriate action plans (digital: 99.30%; conventional: 99.34%). New infections were detected in 10 participants in the digital arm and 4 in the conventional arm (OR 2.73, 95% CI: 0.85-8.80; p = 0.105). Sensitivity of the self-test was 10.0% (95% CI: 0.3-44.5%) and specificity was 99.3% (95% CI: 97.4-99.9%). |
| Discussion/Conclusion | A digitally supported self-testing strategy substantially compressed turnaround time to actionable results while maintaining near-universal linkage to care. Low sensitivity reflects high vaccination coverage and declining infection prevalence. These findings provide the first randomized evidence from South African township populations that digital self-testing can restructure the diagnostic-care cascade in pandemic settings, with relevance for future respiratory pathogen outbreaks. |
| Presenters and Affiliations | Hani Rukh-E-Qamar McGill University Nitika Pant Pai McGill University Aliasgar Esmail University of Cape Town Qihuang Zhang McGill University Jeremi Swanepoel Department of Medicine, University of Cape Town Apoorva Anand McGill University Health Centre Mincen Liu McGill University Keertan Dheda University of Cape Town |
