Carlisle Method highlights new spate of data inaccuracies in medical trials

A sophisticated computer programme that has previously spotted statistical inaccuracies within professional published work has once again highlighted fabricated data across multiple academic papers, according to a new study.

The Myles, Carlisle and Scarr paper, published today (18 February 2019) in the journal Anaesthesia, indicates that there are significant statistical errors, image duplication and unnatural distributions that are sufficient enough to support an investigation into work undertaken by Mario Schietroma’s group, specifically  that which looks at the benefits of liberal peri-operative inspired oxygen.

The screening tool, developed in 2012 by consultant anaesthetist and member of the Association of Anaesthetists, Dr John Carlisle, was first used to check the published work of Yoshitaka Fujii, a Japanese researcher in anaesthesia. It resulted in the retraction of 183 scientific papers and Fujii’s dismissal from his position as an associate professor at the Toho University Faculty of Medicine.

This most recent application of the Carlisle Method to Schietroma’s work revealed that the group’s trials had unnatural distributions of baseline characteristics, alongside results that were significantly different to the collated findings of other published works.

The Schietroma group papers had been used to inform, among other things, the World Health Organisation (WHO) guidelines on the use of liberal peri-operative inspired oxygen, but have now been retracted. The WHO has already revised its guidelines, but the wider implications of the Carlisle Method are profound.

Professor Andrew Klein, the Editor-in-Chief of the journal Anaesthesia said: “In light of the latest spate of fabricated data discovered by the application of the Carlisle Method to Schietroma’s work, it is now apparent that this tool is not only robust in nature, but also fundamental to any and all medical papers being published to measure accuracy and integrity; two cornerstones of research.

“If further investigation confirms that dozens more medical papers contain false or fabricated data, it is possible that certain treatments may need to be withdrawn from use, and currently accepted techniques may need to change – further evidence of the efficacy of the tool.

“It is the belief of Anaesthesia and the Association of Anaesthetists that the Carlisle screening tool should be used worldwide by medical publications, ensuring we do everything we can to commit to complete accuracy in the publication of research.”

You can join the debate on Twitter on 18 February 2019 at 19:00 (GMT).
You can also read our FAQs on the Myles, Carlisle and Scarr paper here.

Please note that it may take up to 5 minutes for your comments to appear.