Back

Preferences for tongue swab- versus sputum-based testing in the context of TB care: a Best-Worst Scaling exercise in Vietnam and Zambia.

Castro, M. d. M.; Le, H.; Muzazu, S.; Pham, N.; Trinh, T.; Nyirenda, H.; Shabalu, P.; West, N.; Phan, H.; Cattamanchi, A.; Denkinger, C.; Muyoyeta, M.; Kerkhoff, A.

2024-12-29 infectious diseases
10.1101/2024.12.22.24319450 medRxiv
Show abstract

BackgroundThe development of non-sputum-based tests is an urgent priority to increase access to tuberculosis (TB) diagnostic testing. Understanding preferences of people undergoing testing is critical for designing tests and strategies aligned with their needs. MethodsWe conducted a survey and Best-Worst Scaling (BWS) exercise to determine relative preferences for tongue swab-vs. sputum-based testing among people ([&ge;]13 years) with presumptive TB at primary health centers in Vietnam and Zambia. The BWS assessed sixteen TB test features, including accuracy, sample type, turnaround time, cost, and service aspects. We analyzed preferences using Hierarchical Bayes modeling and identified distinct preference groups using Latent Class Multinomial Logit analyses (LCA). ResultsAmong 409 participants enrolled, 356 (87%) met quality criteria for analysis. When asked directly, most participants preferred providing tongue swabs over sputum (58% vs 29%, p<0.001; 13% no preference). In the BWS exercise, tongue swab was also preferred over sputum (mean rescaled preference weight [MPW] 6.4 [95%CI: 5.9-6.8] vs. 5.0 [95%CI: 4.6-5.4]). However, support and counseling (MPW=14.0), sensitivity (MPW=12.3), specificity (MPW=10.2), and provider attitude (MPW=7.4) were the most important features overall. Less important features included facility opening hours (MPW=3.4) and the influence of trusted sources on testing decisions (MPW=2.2). LCA identified five distinct preference groups, but support and counseling were universally valued. Participants in Groups 2 (27.3%; n=97) and 3 (17.1%; n=61) valued tongue swabs over many other features. Group 5 participants (11%; n=39) strongly valued sputum-based testing. ConclusionsParticipants in Vietnam and Zambia preferred tongue swab-based TB testing over sputum. However, sample type was less important than test accuracy and other TB care features affecting the testing experience.

Matching journals

The top 1 journal accounts 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.