Back

Centralized molecular testing of community-collected tongue swabs for tuberculosis screening among household contacts: a cross-sectional diagnostic performance, yield, and implementation study

Olifant, S.; Medina-Marino, A.; Pieruccini, M.; Fiphaza, K.; Bezuidenhout, C.; Ruhwald, M.; Penn-Nicholson, A.; Peters, R. P.; Fourie, P. B.

2026-08-02 infectious diseases
10.64898/2026.07.30.26359194 medRxiv
Show abstract

Background Community-based tuberculosis screening can identify asymptomatic tuberculosis, but microbiological confirmation has been sputum-dependent. We evaluated the diagnostic performance, yield, and operational characteristics of centralized qPCR testing of community-collected tongue swabs (TS). Methods We conducted a cross-sectional study among household contacts (HHCs; [≥]18 years) of individuals receiving treatment for pulmonary TB in South Africa. During household visits, health workers collected TS specimens and, where possible, sputum, which was tested in participants' homes using Xpert MTB/RIF Ultra. TSs were transported non-refrigerated in molecular transport medium to a distant central laboratory for manual qPCR assay. The primary outcome was diagnostic performance of TS qPCR compared with sputum Xpert Ultra using paired results. Secondary outcomes were diagnostic yield by TS and operational indicators of the centralized testing workflow. Findings 909 HHCs were enrolled; median age 39 years (IQR 28-55), 532 (58.5%) were asymptomatic, 617 (67.9%) were sputum scarce. TS were collected from 901/909 (99.1%) and sputum from 292/909 (32.1%) participants. Among 271 paired TS-sputum results, TS qPCR sensitivity was 60.9% (95% CI 38.5-80.3), specificity 95.2% (91.7-97.5), and overall agreement 92.3% (88.5-95.0). Yield increased from 1.8% (95% CI 1.0-2.8) with symptom-restricted sputum testing to 3.2% (2.1-4.5) with symptom-agnostic sputum testing and 5.9% (4.5-7.7) when TS testing was implemented among sputum-scarce individuals, identifying 25 additional HHCs with positive molecular test results (86.2% increase). Operational indicators demonstrated successful implementation, with 2.3% specimen loss and valid molecular results for 93.8% who provided a TS. Interpretation Despite lower per-test sensitivity than sputum testing, TS may provide a scalable strategy for expanding microbiological screening beyond sputum-dependent pathways, thereby substantially increasing tuberculosis detection among HHCs. Funding U.S. NIH; Australian Department of Foreign Affairs and Trade; UK Foreign, Commonwealth and Development Office

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.