Mpox study reveals high rate of false-positive PCR tests

In August 2024, the WHO declared the mpox upsurge in the Democratic Republic of the Congo (DRC), and its spread to neighboring countries, a Public Health Emergency of International Concern. The emergence of a new strain and its rapid spread across eastern DRC, and later to other African countries and beyond, made reliable diagnosis essential for understanding the scale of the epidemic and guide vaccination and other public health measures.

A study led by the University of Geneva (UNIGE), the Institute of Tropical Medicine Antwerp (Belgium) and the National Institute of Biomedical Research (DRC), analyzed more than 2,700 PCR results for mpox in four Congolese cities — Goma, Kamituga, Kinshasa, and Uvira — between April 2024 and April 2026. The findings show that 35% of samples reported as PCR-positive may not have represented true infections, but rather environmental contamination with mpox viral DNA. Published in The Lancet Infectious Diseases, the study has direct implications for outbreak surveillance and for the individuals falsely diagnosed.

Mpox is a viral disease characterised by fever and a distinctive rash that can cause painful and disfiguring skin lesions and, in some cases, serious or fatal complications. Fast, reliable case detection is essential for measuring the scale of an outbreak, targeting vaccination, adjusting public health measures, and slowing transmission — particularly in resource-limited settings.

PCR testing plays a central role in mpox diagnosis as clinical symptoms are not specific to this virus. But when the same unusual pattern emerged in results from several different treatment centres, the team led by Andrew Azman, associate professor at the Institute of Global Health and the Center for Emerging Viral Diseases at the UNIGE Faculty of Medicine, took notice. Instead of a continuous distribution of CT values (the indicator reflecting the amount of virus detected in a sample — the lower the value, the higher the viral load), the results clustered sharply into two distinct groups.

The first group corresponds to low values, consistent with a high viral load and therefore a genuine active infection. The second cluster sat just at the detection threshold, pointing to an extremely low viral load — atypical for a real infection, but typical of a residual viral trace. What's striking is that this clear split between the two groups appeared almost identically across all four sites, which is unlikely to be a coincidence."

Andrew Azman, Associate Professor, Institute of Global Health and the Center for Emerging Viral Diseases, UNIGE Faculty of Medicine

Contaminated surfaces

To understand why, the team swabbed surfaces in the treatment areas of two care centres. Many of the surfaces tested positive for traces of the virus, while surfaces sampled outside the treatment areas came back negative. Mpox DNA is highly stable and can persist for several weeks on surfaces, or even remain in the air of treatment centres. This environmental contamination can then be picked up by mistake when sampling other patients, with no connection to an actual infection.

"We also looked at serological data for these individuals where available," explains Megan O'Driscoll, the study's first author and a postdoctoral researcher at the Institute of Global Health. "89% of the people that our model predicts had false-positive diagnoses showed no trace of antibodies against the virus." In a small separate study, the team tested samples from patients with suspected mpox for multiple rash-causing viruses and found that most people had other viruses, like varicella (chickenpox) and measles, with few testing positive for mpox.

Adjusting positivity threshold

These false diagnoses carry real consequences — for public health resource allocation, and for the people affected: unnecessary overtreatment, hospitalisation, or quarantine at a mpox treatment centre, along with the social stigma associated with the disease, and a very real risk of contracting an actual infection there.

"The simplest and fastest fix is to lower the PCR positivity threshold," Azman adds. "A lower threshold significantly reduces the number of false positives without meaningfully increasing false negatives — a proposal currently under discussion with decision-makers. In the long term, this shows that we need to improve the training of lab staff in low-resource settings like this so they can spot these anomalies in real time, rather than after the acute emergency is over."

This work is part of the mandate of the Geneva Center for Emerging Viral Diseases, a WHO collaborating centre for diagnostics, which supports several countries in improving their diagnostic practices.