Novel near-point-of-care TB molecular tests: navigating the repercussions of sensitivity tradeoffs.

New near-point-of-care, swab-based molecular tests for TB offer substantial operational advantages including improved accessibility and cost. However, early data suggest some loss of sensitivity: in particular, these tests tend to miss the lowest-positive molecular signals, such as 'trace' Xpert MTB/RIF Ultra sputum test results. Recent cohort studies suggest most trace signals represent TB disease, even if not currently confirmable by culture. However, the underdiagnosis arising from reduced sensitivity will be greatest in populations with mild disease, where immediate clinical and transmission risks may be lowest and benefits of expanded access may be greatest. Altered test characteristics may also alter empiric TB treatment practices, with variable effects on over-diagnosis. The improved access and yield offered by near-point-of-care platforms and swab specimens are likely to justify these tradeoffs, more than offsetting detrimental effects on case detection in most contexts. Implementation strategies are proposed to mitigate potential harms.