Mpox rapid tests: a positive is usable, a negative is not

Design

retrospective diagnostic accuracy study, operators masked to other assay results, reference standard PCR

Population

lesion swabs from 190 patients with suspected mpox in nine health zones of Kinshasa province, DR Congo

Primary outcome

sensitivity and specificity of five antigen-based rapid diagnostic tests against PCR

Effect

best assay sensitivity 77.3% (95% CI 68.0–84.5), specificity 93.5% (86.6–97.0); lowest sensitivity 39.2% (30.1–49.1)

PCR is the reference standard for mpox and is unavailable across much of the region where mpox is now endemic. This study evaluated five antigen-based rapid tests on lesion swabs from 190 patients with suspected mpox in nine health zones of Kinshasa, against a reference PCR assay, with six swabs taken per patient so every test ran on the same clinical material.

Sensitivity ranged widely. The best, from Guangdong Wesail Biotech, reached 77.3% (95% CI 68.0–84.5) with specificity 93.5% (86.6–97.0). Hangzhou Testsea gave 72.2% sensitivity at the same specificity. Beijing Hotgen managed 59.8% sensitivity at 96.8% specificity, Contipharma 50.5%, and NG Biotech 39.2% — all with specificity above 93%.

So the tests are far better at confirming than at excluding. In a high-prevalence outbreak setting a positive antigen test has a high positive predictive value and can reasonably stand in for PCR to start isolation and contact tracing. A negative result in a patient with a compatible lesion means nothing — at best sensitivity, one in four true cases is missed, and with the worst assay, three in five.

• Use a positive rapid test to act; never use a negative one to stand down isolation.
• Assay choice matters more than usual here — a 39% and a 77% sensitivity are not interchangeable products.
• These were lesion swabs in patients with visible lesions; performance in early or atypical presentations is not addressed.
• India's mpox cases have been sporadic and imported, where prevalence is low and positive predictive value falls — PCR confirmation remains necessary here.
• Collect a PCR swab at the same visit where the pathway allows it; a second visit for confirmation often does not happen.

Why it matters

It gives outbreak responders a defensible way to use antigen tests without pretending they can exclude disease.

Don't overread it

This was retrospective testing on stored lesion swabs from patients already suspected of mpox, not prospective point-of-care use.

The statistics, in plain English

Sensitivity and specificity do not change with prevalence, but predictive value does — which is why this result is described as useful in high-prevalence settings specifically. At 93.5% specificity, roughly one negative sample in fifteen tests positive; where true cases are rare, most positives will be those false ones. The confidence intervals here are wide (68.0 to 84.5 for the best assay) because 97 PCR-positive samples is a modest denominator for a diagnostic accuracy study.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.