End-user preferences and trade-offs for tuberculosis infection testing in a discrete choice experiment: What do people want?

IntroductionTuberculosis infection testing is important but testing uptake is variable. Test-types and attributes may affect engagement, but evidence on end-user values and preferences is lacking. This study aimed to examine end-users preferences and understand trade-offs in test accuracy, operability, cost and convenience. MethodsA cross-sectional discrete choice experiment (DCE) was designed - one for consumers and for health providers. The DCE covered 6 attributes relating test acceptability, accuracy, feasibility and equity. Multinomial logistic regression and willingness-to-accept analyses were used to compare results within and across designs. A minimum sample size of 45 was deemed sufficient to draw statistical inference. ResultsA total of 234 people from 59 countries undertook the DCE (48 consumers, 186 providers); 59% female, most from India (26%) and United States (16%). Providers and consumers preferred tests which minimize false positive ({beta}=0.790,p<0.01 consumers; {beta}=0.189,p<0.05 providers) and false negative results ({beta}=0.967,p<0.01; {beta} =0.169,p<0.1). Consumers strongly preferred testing in the community ({beta}=0.395,p=0.07) and in primary care ({beta}=0.427,p=0.02) compared to hospitals. Avoiding in-person follow-up ({beta}=0.151,p=0.02) and not requiring specialist staff or equipment to administer or interpret the test ({beta}=0.144,p=0.01) affected provider preferences. Providers preferred more expensive tests, though overall, test cost was the least important determinant ({beta}=0.133,p=0.05). Preferences between providers and consumers differed with avoidance of in-person follow-up valued eight times more by providers than consumers. Consumers viewed an increase in likelihood of a false negative result more negatively than an increase in likelihood of a false positive, but the opposite was true among providers. ConclusionsPreferences on TB infection testing was influenced not only by test accuracy but also convenience, portability, and ability to administer in a community / primary care setting. End-users would trade-off key test attributes including test accuracy for convenience. Choice evidence should inform target product profiles for development of novel tests and programmatic implementation. Key MessagesO_LIWhat is already known on this topic O_LITesting for infection plays a critical role in diagnosing disease and guiding preventive treatment to reduce progression and transmission. However, uptake remains inconsistent, potentially due to test characteristics that do not align with user preferences and expectations. C_LI C_LIO_LIWhat this study adds O_LIWhile both providers and consumers value test accuracy, their preferences differ. C_LIO_LIConsumers prioritize accessibility through community or primary care testing, whereas providers emphasize operational simplicity and minimizing in-person follow-up. C_LI C_LIO_LIHow this study might affect research, practice or policy O_LIDCE can be implemented across diverse populations to gain broader insights into TB infection testing preferences, emphasizing that test design and delivery should be guided by both provider and consumer perspectives to ensure high uptake and programmatic success. C_LI C_LI