News, Culture & Society

Prestigious medical journal admits it published FIVE flawed studies

A prestigious medical journal yesterday admitted flawed research in six scientific papers – and retracted one.  

The New England Journal of Medicine took down a landmark study from 2013 that uncovered the benefits of a Mediterranean diet on the heart. 

But instead of removing the flawed data from their archives completely, they have reanalysed the data and dampened down its claims.

Little has changed in terms of the overall results of the Spanish research of nearly 7,500 people – but critics remain wary of the ‘sloppy’ experiment.

Corrections were issued for five other trials that contained errors, after they were flagged by an obscure report last year that scrutinzed thousands of studies to assess their validity. 

The New England Journal of Medicine took down a landmark study from 2013 that uncovered the benefits of a Mediterranean diet on the heart

Eating a Mediterranean diet supplemented with olive oil or nuts slashes the risk of heart attacks or strokes by 30 per cent, the retracted trial originally found.

A second glance at the University of Navarra data, by the same researchers who led the original study, found the percentage remained similar. 

However, instead of claiming the Mediterranean diet was responsible for the heart benefits, the revised paper noted just a correlation. 

‘Naive’ and ‘sloppy’ 

Dr Barnett Kramer, director of the division of cancer prevention at the National Cancer Institute, was one sceptic of the new analysis.

Speaking to the New York Times, he added: ‘Nothing they have done in this re-analyzed paper makes me more confident.’

Donald Berry, a statistician at MD Anderson Cancer Center in Houston, told the NY Times that the researchers were ‘naïve’ and ‘sloppy’.

‘Reassuring’ results 

Professor Naveed Sattar, from Glasgow University, said: ‘This is a highly unusual step and I am sure that the NEJM took this decision very seriously.’

He added that it was ‘reassuring’ to see results remain broadly similar – but argued some medical guidelines may already include the original findings.

WHY WAS THE PAPER RETRACTED?

Many scientific experiments randomly assign people to different groups to compare one treatment to another. 

The groups should be similar on height, weight, age and other factors, and statistical tests can suggest whether the distribution of these traits is implausible.

Without this, results of any trial could be compromised. 

The flag was first raised last June, when the editor of the journal Anaesthesia took a deeper look into more than 5,000 randomised experiments.  

Dr John Carlisle, branded ‘instrumental in exposing statistical anomalies’, used one such test to scrutinize the studies from 2000 through to 2015, including 934 in the NEJM and claimed 11 were suspicious.

The journal contacted each author and ‘within a week we resolved 10 of the 11 cases’, said the NEJM’s chief editor Dr Jeffrey Drazen. 

In five, Dr Carlisle was wrong. Five others were terminology errors by the authors and led to corrections by the journal yesterday. 

The last was the University of Navarra study, which made headlines across the world.

Dr Miguel Ángel Martínez González, part of the original study, dug through records and found not all procedures had been followed.

If one person in a household joined the study, others such as a spouse also were allowed in. A total of 14 per cent of participants weren’t randomised. 

When results were re-analyzed without those participants, the bottom line remained the same, and the journal is now publishing both versions.

Dr Ian Johnson, of the Quadram Institute Bioscience, said: ‘My reading of this is that some initial statistical discrepancies have now been properly accounted for.’ 

Many scientific experiments randomly assign people to different groups to compare one treatment to another. 

The groups should be similar on height, weight, age and other factors, and statistical tests can suggest whether the distribution of these traits is implausible.

Without this, results of any trial could be compromised. 

Red flags 

The flag was first raised last June, when the editor of the journal Anaesthesia took a deeper look into more than 5,000 randomised experiments.  

Dr John Carlisle used one such test to scrutinize the studies from 2000 through to 2015, including 934 in the NEJM and claimed 11 were suspicious.

The journal contacted each author and ‘within a week we resolved 10 of the 11 cases’, said the NEJM’s chief editor Dr Jeffrey Drazen. 

In five, Dr Carlisle was wrong. Five others were terminology errors by the authors and led to corrections by the journal yesterday. 

The last was the University of Navarra study, which made headlines across the world.

The reanalysis 

Dr Miguel Ángel Martínez González, part of the original study, dug through records and found not all procedures had been followed.

If one person in a household joined the study, others such as a spouse also were allowed in. A total of 14 per cent of participants weren’t randomised. 

When results were re-analyzed without those participants, the bottom line remained the same, and the journal is now publishing both versions.

Dr Drazen said: ‘When we discover a problem we work very hard to get to the bottom of it. There’s no fraud here as far as we can tell. But we needed to correct the record.’

Rising retractions 

The most known scientific paper that has later been retracted is Andrew Wakefield’s study in 1995 that suggested the MMR vaccine could cause autism.

Vaccination rates plummeted in the following years. The Lancet formally retracted the gastroenterologist’s research paper in 2010 – 15 years later.

Around  1,350 papers were retracted in 2016 out of two million published – less than a tenth of a per cent, but up from 36 out of 1 million in 2000. 

Studies are often the main source of evidence that guides doctors’ decision-making and patient care. 



Read more at DailyMail.co.uk


Do you like it? Share with your friends!


Comments are closed.