The study, led by the University of East Anglia (UEA) in England, found these technologies are not effective in real world settings.
The devices claim to work by removing unwanted particles to improve indoor air quality. They were put forward as a way of stopping the spread of Covid-19 and making indoor mixing safer when the pandemic started in 2020.
Researchers looked at observational and experimental studies published around the world from 1970 to 2022, searching for data on infection and symptom outcomes for people who spent a minimum of 20 hours a week in shared indoor spaces using the technologies.
They said pooled statistics from 32 studies “suggested no net benefits of air treatment technologies for symptom severity or symptom presence, in absence of confirmed infection”.
Dr Julii Brainard, of UEA’s Norwich Medical School, added: “The kinds of technologies that we considered included filtration, germicidal lights, ionisers and any other way of safely removing viruses or deactivating them in breathable air.
“In short, we found no strong evidence that air treatment technologies are likely to protect people in real world settings.”
Professor Paul Hunter, of UEA’s Norwich Medical School, said: “When the Covid pandemic hit, many large companies and governments – including the NHS, the British military, and New York City and regional German governments – investigated installing this type of technology in a bid to reduce airborne virus particles in buildings and small spaces.
“But air treatment technologies can be expensive. So it’s reasonable to weigh up the benefits against costs, and to understand the current capabilities of such technologies.”
Dr Brainard added: “There is a lot of existing evidence that environmental and surface contamination can be reduced by several air treatment strategies, especially germicidal lights and high efficiency particulate air filtration (HEPA).
“But the combined evidence was that these technologies don’t stop or reduce illness.
“There was some weak evidence that the air treatment methods reduced likelihood of infection, but this evidence seems biased and imbalanced.
“We strongly suspect that there were some relevant studies with very minor or no effect but these were never published.”
She said that while her team’s findings are “disappointing” it is “vital that public health decision makers have a full picture”.