Home tongue-swab testing helps find household tuberculosis cases
Andrew Medina-Marino led a study showing in-home tongue swab and sputum testing can identify TB among household contacts, though pooled swabs reduce sensitivity.
Tuberculosis programs often rely on household contacts going to clinics for testing, and many contacts never get tested. To address that gap, a team led by Andrew Medina-Marino conducted a prospective cohort study in the Eastern Cape, South Africa, to evaluate whether in-home molecular testing of tongue swabs could support household contact investigation. The study enrolled 909 household contacts (HHC) of people with TB between June 2021 and October 2024. Investigators collected sputum when possible and asked nearly everyone for a tongue swab (TS). To make testing more efficient, some TSs were pooled from up to three household contacts and then analyzed using Xpert Ultra on portable GeneExpert devices brought into the home. The goal was to see how well TS testing worked compared with sputum testing, and whether in-home testing could speed diagnosis and linkage to care at the household level without relying on clinic referrals.
The study found that TS collection was feasible: 99.1% of enrolled HHC provided a TS, while 31.6% provided sputum. Using Xpert Ultra on GeneExpert devices, the overall sensitivity of TS testing versus sputum was 61.9% (95% CI: 38.4%-81.9%), with specificity 100% (98.9%-100%). When swabs were tested individually rather than pooled, TS sensitivity was reported as 100% (47.8%-100%). Pooling reduced detection: among two-swab pools where at least one individual had a positive sputum result, 55.6% (21.2%-86.3%) of pools tested positive; among three-swab pools the figure was 42.9% (9.9%-81.6%). TS sensitivity also declined as the sputum Xpert Ultra semi-quantitative category decreased. At the household level, 27 of 439 households (6.2%) had an indication of secondary TB: 13 households (3.0%) were identified by both sputum and TS, 11 (2.5%) by sputum only, and 3 (0.7%) by TS only. Sputum testing identified 29 HHC with TB (yield = 3.2%), and 25 of those 29 (86.2%) were linked to care with a median time of 1 day (IQR 1-2).
These results show both promise and trade-offs for in-home molecular testing during household contact investigation. Bringing Xpert Ultra on portable GeneExpert devices into homes allowed rapid testing and, for those with sputum-positive results, fast linkage to care—most people began follow-up within a day. Tongue swabs made it possible to test almost every contact regardless of symptoms or ability to produce sputum, which could improve overall reach. However, pooling tongue swabs to save resources lowered the chance of detecting TB compared with testing individual swabs, and TS sensitivity fell when sputum bacterial load was lower. The study suggests programs can use in-home sputum and TS testing to expand screening and speed treatment, but must balance efficiency gains from pooling against the risk of missed cases and consider when to escalate diagnostics from pooled TS to individual TS or sputum testing.
In-home molecular testing can extend TB screening to nearly all household contacts and enable rapid linkage to care for people with sputum-positive results. Programs should weigh the resource savings of pooled tongue swabs against reduced sensitivity and potential missed diagnoses.
Author: Andrew Medina-Marino