In July Scott Gavura wrote a review of cannabidiol (CBD, one of the active ingredients in cannabis) as an herbal drug and correctly pointed out that the hype is greatly outpacing the science. His main points were that, generally, the quality control of CBD products is poor (as with many herbal products) and this makes it difficult to extrapolate from scientific studies to actual use. Further, the evidence for efficacy for many of the claimed uses, such as pain, movement disorders, and psychiatric disorders, is actually quite poor and unconvincing (the exceptions being the very narrow uses in Dravet syndrome and Lennox-Gastaut syndrome). Some claims, like treating cancer, are pure pseudoscience. And finally there are side effects and drug-drug interactions with CBD, and so safety cannot be assumed.

A recent study emphasized Scott’s concerns. This is an observational study looking at older adults with COPD (chronic obstructive pulmonary disease) and their use of cannabinoids. The study included 4,000 individuals, and they found:

Researchers observed particularly worse health outcomes among patients with COPD who were using higher doses of cannabinoids. Compared to non-users, new higher-dose cannabinoid users had a 178 per cent relative increase in hospitalizations for COPD or pneumonia, and a 231 per cent relative increase in all-cause death.

That is a significant effect size, especially for mortality as an outcome. This is the kind of data that could kill a new drug in development. The study also found that those using cannabinoids were no better off than patients using opioids for chronic pain. The usual caveats to an observational study apply – this is not randomized, which means that people may have chosen to use cannabinoids because they were sicker, rather than the use making them sicker. Or there may be some other confounding variable. This will need to be sorted out with randomized controlled trials.

And that is the point. I am not saying (and Scott was not saying) that CBD specifically or cannabinoids in general do not work or are unsafe. Right now we mostly have preliminary data. What we are saying is essentially two things. First, the preliminary data is not that impressive. It certainly isn’t a home run, especially for the most common uses of these agents. This does not rule out that eventually we will find a formulation that has a benefit for specific indications. It means we don’t know.

Second, and the point that I most want to emphasize here, is that we really do need high quality research in order to know what the net health effects are of any new proposed treatment. Long-term anecdotal use means very little. People were extolling the virtues of radioactive tonics while they were killing themselves. Native Americans used tobacco as a medicinal. Bloodletting was all the rage for two millennia. Uncontrolled observations are misleading for many reasons, such as our poor grasp of statistics (such as regression to the mean) and the vagaries of perception and memory that lead to confirmation bias and subjective validation.

There is another factor when it comes to products like CBD or other cannabinoids – the public has been programmed by literally centuries of marketing to accept the health halo around things that are deemed “natural“. This is a meaningless distinction, however. CBD is a drug, nothing more and nothing less.

In the end what matters is good clinical decision-making backed by solid scientific evidence. Clinical decision-making generally revolves around considerations of risk vs. benefit, personalized to the individual patient’s priorities and values. In order to make a risk vs. benefit determination, you need objective data (not vague reassurances). We need to know the quality and consistency of the product, so that we can determine exactly what the patient is getting and in what dose. We need to know, as much as possible, what the body does to the drug and what the drug does to the body, how the drug is eliminated, the half life, and any drug-drug interactions. We then also need to know the effect of the drug, both beneficial and negative, in different populations and conditions.

Yes – you really do need to know all of that before you can make any reasonable risk-benefit decision. One piece of missing information can make a dramatic difference in outcome. CBD is no different. The existing evidence is showing that there are serious drug-drug interaction (mostly because of changes to liver enzymes), and there are potentially serious side effects in certain populations. In addition, beneficial effects are mostly small and/or dubious.

I am not saying this to fearmonger about CBD – all of this is also true of most new drugs. The point is that we should not give CBD special treatment, because of its mystique, or because of the appeal to nature fallacy. We need to complete the normal process of studying CBD as the drug that it is. There is no justification for short-circuiting this process. If we do, then we risk causing far more harm than any realized benefit, as the new study of increased death in older COPD patients suggests. We are currently at the point analogous to when radioactive tonics were in their heyday – all hype, little science, and marketed more with feelings than with facts.

Although it’s probably too late, we should hit the pause button on cannabinoids until proper research has a chance to play itself out. A “what’s the harm” approach here is not appropriate. Appeal to anecdote is more likely to cause harm than good. Let the science happen.


Posted by Steven Novella

Founder and currently Executive Editor of Science-Based Medicine Steven Novella, MD is an academic clinical neurologist at the Yale University School of Medicine. He is also the host and producer of the popular weekly science podcast, The Skeptics’ Guide to the Universe, and the author of the NeuroLogicaBlog, a daily blog that covers news and issues in neuroscience, but also general science, scientific skepticism, philosophy of science, critical thinking, and the intersection of science with the media and society. Dr. Novella also has produced two courses with The Great Courses, and published a book on critical thinking - also called The Skeptics Guide to the Universe.