Study of lead levels in rice under scrutiny
Tests indicating that rice imported to the US contained high levels of lead have been cast into doubt.
At a conference in April, researchers reported that commercially available rice contained many times more lead than US food authorities deemed safe.
The findings sparked international concern over imported rice.
But preliminary independent checks on the findings have failed to replicate the results, and tests suggest the equipment used may have been to blame.
The initial findings, revealed at a meeting of the American Chemical Society, reported on tests of rice imported to the US from eight nations.
The team, led by Tsanangurayi Tongesayi of Monmouth University in New Jersey, US, analysed rice using a technique called X-ray fluorescence.
Researchers claimed to have found levels of lead exceeding by a factor between 20 and 40 the "provisional total tolerable intake" for adults, set by the US Food and Drug Administration. Their report suggested that untreated wastewater used in irrigation was a likely cause.
Media reports, including that by BBC News, have caused concern internationally, prompting two members of the European Parliament to raise the issue formally and a follow-up study by the Dutch food safety authority NVW.
However, attempts to replicate the results have found levels far below those initially reported - between 6 and 12 parts per million (6,000 to 12,000 parts per billion).
Dr Tongesayi's team sent samples to another laboratory for analysis using a different technique - that study recorded levels below one part per million.
The team then put on hold planned publication of the findings in the Journal of Environmental Science and Health, for what Dr Tongesayi told BBC News was a "data verification exercise". The American Chemical Society was asked to remove the press release on the work from its website.
"The most important issue for me at this point is to make sure the data is accurate," Dr Tongesayi told BBC News in late April. "If it is not accurate, we will obviously not publish the paper."
The team subsequently sent the instrument used in the study back to its manufacturer, which has since reported that the machine has calibration problems.
The Dutch authorities' independent study of 26 samples of rice imported from Asian nations found average levels of seven parts per billion - a thousandth of those found in the original study - and with no samples above the EU limit of 200 parts per billion.
Dr Tongesayi's findings also stand in stark contrast to prior published research on lead in rice, most of which have been established using a technique called mass spectrometry, which allows for more precision.
The International Rice Research Institute (IRRI) said in a blog post that "even where the soil is contaminated with a lead spill, a number of studies have shown that rice plants do not take up a significant amount of lead and move it to the grains".
Alex Waugh of the UK's Rice Association said that "in terms of work undertaken throughout Europe and the USA on rice of multiple origins, Dr Tongesayi's reported results stand out as being orders of magnitude higher than normal".
"This in itself ought to be enough to raise questions about whether his data are correct," he told BBC News.
In 2010, the European Food Safety Authority published a report outlining analysis of 612 rice samples from the EU, finding average levels of just a few tens of parts per billion.
An extensive study in 2012 by US magazine Consumer Reports, including rice and rice-based products such as rice cakes and drinks available in the US, measured lead levels even lower, with a majority of samples measuring less than five parts per billion.
The Federation of European Rice Millers told BBC News that "there is no published evidence of rice containing the levels of lead of [even] the same order of magnitude reported by Dr Tongesayi, and consequently no evidence of 'harmful levels' of lead in rice on the European market".
Dr Tongesayi told BBC News he was determined to reconcile his initial findings with the outcomes of subsequent analyses of his samples by other means, including the mass spectrometry method.
Sarah Beebout, a soil scientist with the IRRI, said: "I will be surprised if the independent analysis confirms these apparently anomalous results, but that will be a good starting point for scientific discussion and investigation if it happens."