Data Fabrication Could Be Rife In Drugs Trials

| June 30, 2017

These days, the gold standard of evidence-based medicine is the double-blind, randomised, placebo-controlled trial. If a drug or natural compound doesn’t show a positive effect in such a trial, it is unlikely to be approved as a conventional treatment. But a bombshell paper now claims the results and data of such trials may often be falsified.1

Dr. JB Carlisle, consultant anaesthetist at Torbay Hospital in Devon, is also something of an expert in statistics. And it must be his passion, since he undertook the Herculean task of trawling through 5,087 randomised, controlled trials, published in respected scientific journals between January 2000 and December 2015.

Dr. Carlisle used a statistical test to check whether these trials really were ‘randomised’ – in other words, whether participants were allocated to treatment or placebo groups entirely at random. Provided the trial is large enough, this process of randomisation evens out variable factors between the groups that could affect the results.

The findings were shocking. More than 15 per cent of trials that had been published, and not retracted later, had abnormal data distribution, showing they weren’t properly randomised. And, when Dr. Carlisle applied his test only to trials that had been retracted after publication, that figure jumped to 43 per cent.

It seems that quality scientific journals, like the New England Journal of Medicine or the Journal of the American Medical Association, are quite good at spotting dodgy trial data and (probably very politely) suggesting the authors retract their paper. But there are still the 15 per cent that slip through the net. And, when it comes to those “pay-to-publish” journals that are less rigorous, the figures could be a lot higher.

Big Pharma’s bad practices

So, does this mean that in all the trials that failed Dr. Carlisle’s test, participants were deliberately allocated to particular groups or that data was deliberately falsified? Possibly not, because errors can creep in, for instance when figures are entered onto a computer spreadsheet – a decimal point in the wrong place can make a big difference! And a poorly-designed trial, or one with too few participants, is likely to show the same kind of anomalies.

This is not the first time that Big Pharma has been accused of rigging the results of clinical trials. As I pointed out here, previous investigations have revealed the extent to which trial data may be invented, altered or hidden, in order to show a favourable result and so promote drug sales. Last year, the Chinese State Food and Drug Administration investigated 1,622 clinical trials for new pharmaceutical drugs and found that an astonishing 80 per cent of the trial data was suspect, incomplete, or totally non-existent!

In fact, corruption, malpractice and fraud are just part of normal business for Big Pharma, according to a report published last year.2 These dirty dealings run right through the process of drug development, testing, approval and prescribing. And according to the same report, national governments, regulatory bodies and Big Pharma’s bosses are all turning a blind eye. But really, it’s not surprising that when a profit-driven, fiercely competitive and morally bankrupt industry is put in charge of looking after our health, things can go horribly wrong.

Wishing you the best of health,

Martin Hum
PhD DHD Nutritionist
for Real Diabetes Truth

Did you find this information useful?


Sign up now to get instant access to 10 Diet Secrets for Healthy Blood Sugar along with daily updates on how to enjoy good health, naturally.

No spam pledge: We’ll never rent, misuse or abuse your email address. By signing up you are subscribing to the free Daily Health e-letter and you will occasionally receive carefully selected advertising messages from us and third parties. By signing up, you are consenting to receive these promotions. You can unsubscribe at any time. See our privacy policy


Bear in mind we are not addressing anyone’s personal situation and you should rely on this for informational purposes only. Please consult with your own physician before acting on any recommendations contained herein.

  1. Carlisle JB. Data fabrication and other reasons for non-random sampling in 5087 randomised, controlled trials in anaesthetic and general medical journals. Anaesthesia. 2017 Jun 4 (Online ahead of print).
  2. Corruption in the pharmaceutical sector: diagnosing the challenges. Transparency International UK, 2016.
Print Friendly

Tags: , ,

Category: Diabetes News Views

Comments are closed.