Clinical trials are tricky – one can say one thing, one can say another. But what if people with a financial interest could hide the ones you didn’t like, and only use the ones you did like?
A series of studies in the British Medical Journal suggest that clinical trials are missing – trials that may say something quite different than what the public is hearing.
In the United States, the government is trying to tighten up on requirements for clinical trials. One of these regulations is that all eligible clinical trials must be reported (on ClinicalTrials.gov) within one year of the study’s conclusion. This came into effect in 2007.
This database has an impressive list of studies – currently 119,213 trials from 178 countries. So whether or not trials are reported truly has a global impact.
So has the new requirement solved the problem? Not a chance. One study found that only 22% of the eligible trials had been reported after 12 months! (see Compliance with mandatory reporting of clinical trial results on ClinicalTrials.gov) Another study found that 50% of trials funded by the National Institutes of Health were published after 30 months, and a third of them were still unpublished after 51 months. (see Publication of NIH funded trials registered in ClinicalTrials.gov) Yes, that’s over 4 years!
Another study found that even when studies were reported on Medline (an online database of trials), they weren’t properly tagged and indexed. (see Understanding why evidence from randomised clinical trials may not be retrieved from Medline)
–>>> This means basically that someone could do a well-designed clinical trial related to cluster headache or migraine, then not report the results for a few years. Even when they do report it, they can leave off tags so that no one can find it. Even if they do find it, many will miss studies that are so old.
Why is this happening? Don’t scientists want to publish their work?
Admittedly, some of this could come from a misunderstanding of the rules, laziness, simple mistakes, and so forth. There’s also time and expense to consider – it may be easier for a large company to publish than a smaller one with limited resources.
But the reality is that many of these studies are sponsored by the companies that sell you the medication you use. If a study says something the drug company doesn’t like – the study may just disappear. And that’s a scary thought.
Often researchers will publish a meta-analysis of the available studies. For example they could take the data from 12 trials of a certain drug, and tell us that, according to 8 of the studies, the drug is very effective. That sounds good until you realize there could be 12 more trials out there somewhere that say the opposite. One study found just that – the missing data often contradicted what the meta-analysis studies were telling us. (see Effect of reporting bias on meta-analyses of drug trials: reanalysis of meta-analyses)
The answer isn’t easy. Clinical trials can be tricky to regulate. And it’s even harder if you’re a small fish playing in the same pond with the huge drug companies.
But if the rules are there, they need to be enforced. And the public needs to know when they’re not being enforced.
The world of clinical trials is not as pure, easy to understand, and efficient as many people believe. It’s more important than ever for the buyer to beware – do your own research, be cautious, and don’t believe everything you read – even if it’s in a medical journal. What’s missing may change a whole lot.