A leading medical journal has 'expressed concern' over the study that said hydroxychloroquine raises the risk of death for Covid-19 patients, which halted global trials.
The Lancet, which published the controversial research on May 22, said there were 'serious questions' to be answered about the data used to come to its conclusions.
It comes after more than 120 leading scientists and doctors from around the world penned an open letter to the editor of The Lancet raising serious concerns about its methods.
They highlighted 10 major flaws, including patient data which they say does not match with public health records.
US and Swiss scientists ruled the Donald Trump-backed anti-malaria drug hydroxychloroquine doesn't work - and is even dangerous - for Covid-19 therapy.
The findings sparked panic among scientists across the world who halted major trials of hydroxychloroquine.
Despite the Lancet's concern, scientists said the 'harm was already done' and that it would be difficult to recruit patients into any clinical trials of hydroxychloroquine following the study.
The data was supplied by US-based healthcare data analytics company Surgisphere Corporation, which drew on hospital records for 100,000 people globally.
Editors of the New England Medical Journal have previously questioned the company over the validity of its data supplied for another Covid-19 trial.
A leading medical journal has 'expressed concern' over the study that said hydroxychloroquine (an antimalarial, pictured) raises the risk of death for Covid-19 patients, which halted global trials
The Lancet, which published the controversial research on May 22, said there were 'serious questions' to be answered about the data used to come to its conclusions in a letter last night
The paper published on May 22 found hydroxychloroquine and a weaker version chloroquine had been linked to increased risk of death and heart arrhythmias among people admitted to hospital with Covid-19.
The authors of the study said neither drug should be used to treat Covid-19 outside of clinical trials and said randomised clinical trials were needed.
Their finding prompted the UK’s drugs watchdog to temporarily suspend two major Oxford University clinical trials of the antimalarial.
The World Health Organization also pulled the plug on its SOLIDARITY study, on the back of the worrying results.
But on Tuesday evening, the Lancet's editors published an 'expression of concern' and said 'important scientific questions' had been raised about the data used in the study.
The data set was supplied by US-based healthcare data analytics company Surgisphere Corporation and its founder, Dr Sapan Desai, was one of the paper's four co-authors.
Among the criticisms were the seemingly high mortality rates linked to drugs that have been routinely prescribed since the 1950s.
The Lancet's editors said: 'Although an independent audit of the provenance and validity of the data has been commissioned by the authors not affiliated with Surgisphere and is ongoing, with results expected very shortly, we are issuing an expression of concern to alert readers to the fact that serious scientific questions have been brought to our attention.
'We will update this notice as soon as we have further information.'
Surgisphere said: 'In our hydroxychloroquine analysis, we studied a very specific group of hospitalised patients with Covid-19 and have clearly stated that the results of our analyses should not be over-interpreted to those that have yet to develop such disease or those that have not been hospitalised.
'We also clearly outlined the limitations of an observational study that cannot fully control for unobservable confounding measures, and we concluded that off-label use of the drug regimens outside of the context of a clinical trial should not be recommended.
'Our Covid-19 research was not funded by any drug company, private or public donor, or political organisation.
WHAT DID THE LANCET SAY?
The Lancet: Expression of concern: Hydroxychloroquine or chloroquine with or without a macrolide for treatment of COVID-19: a multinational registry analysis
Important scientific questions have been raised about data reported in the paper by Mandeep Mehra et al— Hydroxychloroquine or chloroquine with or without a macrolide for treatment of COVID-19: a multinational registry analysis1 —published in The Lancet on May 22, 2020.
Although an independent audit of the provenance and validity of the data has been commissioned by the authors not affiliated with Surgisphere and is ongoing, with results expected very shortly, we are issuing an Expression of Concern to alert readers to the fact that serious scientific questions have been brought to our attention. We will update this notice as soon as we have further information.
'Our research collaborators on the piece for The Lancet devoted their time through personal funds and resources because they saw the urgent humanitarian need and opportunity to inform rapidly evolving pandemic responses.'
The study analysed data from almost 15,000 patients with Covid-19 receiving the drugs alone or in combination with antibiotics.
It then compared this data with the hospital records of 81,000 controls who did not receive the drug - and claimed the data came from six continents.
Treatment with the medications among patients with Covid-19, either alone or in combination with antibiotics, was linked to an increased risk of serious heart rhythm complications and death.
But the authors stressed that anyone taking these drugs for other conditions should not stop taking them as the trial looked specifically at Covid-19.
The researchers estimated that the excess risk attributable to the use of the drugs rather than other factors such as underlying health issues ranged from 34 per cent to 45 per cent.
Professor Babak Javid, principal investigator at the Tsinghua University School of Medicine in Beijing, said 'mounting concerns' had questioned the validity of the data.
He said: 'For example, the number of Covid cases that were supposed to be from a subset of Australian hospitals was actually greater than the sum total of cases in Australia reported at the time.
'In many ways, the harm has already been done. No high-quality trials of hydroxychloroquine for Covid have yet reported, and some may now be unable to recruit sufficient patients to arrive at an answer.'
More than 120 scientists, including prominent Imperial College London epidemiologist Professor Neil Ferguson, whose stark warning that 250,000 Brits could die without action played a role in the UK triggering lockdown in March, wrote to The Lancet to address their concerns with the study.
The signatories of the letter said the study did not mention the countries or hospitals that contributed to the data source, meaning they cannot be fact checked.
They wrote: 'The authors have not adhered to standard practices in the machine learning and statistics community. They have not released their code or data.
'There was no ethics review... There was no mention of the countries or hospitals that contributed to the data source and no acknowledgments to their contributions.'
The scientists nodded towards the fact the Lancet paper included data from more Covid-19 deaths in Australia that existed at the time.
Dr Desai told the Guardian that this was due to an error that caused one hospital in Asia to be included in the Australian dataset, and a correction was made in the paper.
'This indicates the need for further error checking throughout the database,' scientists told The Lancet.
The letter, first seen by the Guardian, also states that data from Africa claims nearly a quarter of Covid-19 patients and 40 per cent of all deaths on the continent happened in hospitals where Surgisphere operates.
The experts say this is 'unlikely' to be true.
It follows The New England Medical Journal (NEMJ) writing an expression of concern over a study it published using data from Surgisphere.
The study found common blood pressure medicines do not put people at a higher risk of severe or fatal coronavirus symptoms, three major studies have found.
There had been concern arising from animal studies that these medicines ACE inhibitors and angiotensin receptor blockers (ARBs) might increase the body's levels of a protein called ACE2, which the coronavirus latches on to when it invades human cells, thus increasing people's vulnerability to the disease.
It used observational data from 169 hospitals in Asia, Europe, and North America to also find Covid-19 patients with cardiovascular disease were at an increased risk of death,
NEJM editor-in-chief Eric Rubin wrote in the expression of concern: 'Recently, substantive concerns have been raised about the quality of the information in that database.
'We have asked the authors to provide evidence that the data are reliable. In the interim and for the benefit of our readers, we are publishing this Expression of Concern about the reliability of their conclusions.'
US President Donald Trump has been criticised for promoting the drugs – which are used to treat malaria, arthritis and lupus – as a cure for the new virus.