An inquiry into some tuberculosis tests on sale to developing countries has found that most were missing more cases than they found.
The evaluation of 19 commercially available rapid diagnostic tests, which detect antibodies to tuberculosis bacteria, found that on average they could identify only about a quarter of cases. Most detected less than half and one test identified less than one in a hundred cases.
"This means that ninety-nine out of 100 TB-infected people would continue to go about their daily lives coughing and infecting people," said Rosanna Peeling, director of diagnostics research and development at the WHO Special Programme for Research and Training in Tropical Diseases (WHO-TDR), which conducted the research.
"Imagine if this goes on for one year, how many more people would they have infected?"
Peeling was speaking at a workshop for diagnostics experts this month (3 December) which called for a global body to regulate diagnostic tests, particularly those used in the control of diseases that disproportionately affect developing countries.
Such a body could set standards for tests, provide guidelines for their evaluation, ensure participating manufacturers adhere to a code of conduct and provide a 'seal of approval' for quality products, they said.
Few developing countries have regulatory infrastructure for approving the sale and use of diagnostic tests, Peeling told the 'Global Health Diagnostics: Research, Development and Regulation' workshop, organised by the Academy of Medical Sciences in London, United Kingdom.
Governments, disease control programmes, donors and the general public could all be buying useless tests.
The problem is particularly acute in the case of rapid diagnostic tests (RDTs), which act in a way similar to pregnancy tests and have been hailed as the solution to providing affordable and timely point-of-care diagnosis in developing countries.
"Very few countries actually look at the quality of the products before they are allowed to go on sale," said Peeling.
The study, published last October, focussed on RDTs for TB. It concluded that "the tests are sold and used in disease-endemic countries, without evidence of effectiveness". It pointed out that the majority of the tests were technically simple and, if performance could be improved, would be appropriate for their target use.
Some contributors to the London meeting said that the onus should be put on manufacturers to ensure that their tests meet quality standards.
"It would be good to have a seal of approval from a diagnostics manufacturing association that we could trust," says Peeling.
WHO-TDR is working with networks of laboratories in over 40 countries to engage them in the evaluation of RDTs for their country or region.
"If national regulatory authorities can become more active using their reference laboratories to properly evaluate diagnostic tests, then hopefully national authorities will learn to distinguish between reliable and unreliable diagnostics," said Peeling.
"Since reliance upon bad diagnostics can be public health and life-threatening, unreliable products would thus be removed from the market either by voluntary or regulatory actions."