Breaking news: Study finds astonishing results about teens, marijuana, and health issues!

Which of the following is true if you started smoking marijuana as a teen:

1. You will have more substance abuse problems later in life
2. It will increase the chance that you will experience a psychotic illness
3. You will not be any more anxious or depressed than when you started smoking marijuana
4. As you grow older you may have fewer allergies

Psychological research studies have become a hot topic in the media. Every month hundreds of journals publish a new set of articles, and journalists from the major daily papers, magazines, health blogs, and other sources scour the titles looking for a scoop. Academics and their respective research institutes have caught on, and now provide new releases complete with “plain language summaries”. Within literally minutes from being released, a high interest study will appear in news feeds, often containing huge portions of the original summaries paraphrased liberally, with other parts overlooked or simply omitted. It can all make for interesting reading and helps to get new research to the masses. It can also make for some pretty eye catching tag lines like the one that got you reading this post – the kind people write to get your attention by picking out surprising conclusions that may not be quite accurate but sure sound interesting and get lots of traction. Unfortunately, all too often the more important message can get lost in translation.

Believe it or not, I’m actually going to use a recent study on marijuana to make my point. It was a study published in March last year by Bechtold et al. tracking marijuana users from their teens until they were well into their 30s – a not too easy feat – while simultaneously following a wide variety of physical outcomes, including stuff like heart problems, inability to carry out activities, and allergies as well as mental health outcomes like depression, anxiety, and psychotic illness. This is all pretty relevant given that the US government is holding its course in the war against drugs while several states de-criminalize its use. Just the sort of thing that gets both sides on the debate very, very interested. And boy does that ever make for good headlines and lots of short-term traction in the media.

In fact, Bechtel et al.’s publication was responsible for a flurry of eye catching headlines and a ton of attention. It also was followed by the publication of a “clarification” that hardly got any traction at all. In case you are wondering, a clarification is something that researchers are required to do when other researchers read their study, have a fit, and write a letter to the editor insisting that the authors made a mistake and must re-analyse the data in order to correct the mistake.

So what got everyone’s attention? And why the clarification? And how did it all sort out? Let’s start with a quick recap of the study. Like many researchers in the field of drug use, Bechtel et al. realized that quite a few studies had been published on the supposed harms of pot use. The most common negative outcomes included things like increased depression and anxiety, and possibly even psychosis, which is about as nasty and serious as mental health symptoms get. But Bechtel’s group of researchers were not looking to simply replicate these studies, they wanted to address a glaring problem that most of the prior efforts shared – they had only asked people to reflect on their past experience. None had actually followed people over a reasonably long period of time, nor had anyone actually interviewed their subjects from the start and all along the way. Nor had most researchers accounted for things like a person’s age, education level, or many other features that might actually account for negative health outcomes. In other words, they carried out a pretty thorough study that took a lot of time and a lot of work.

In the end, they got a surprise. Based on their results, it seemed that the only thing early marijuana use predicted was a higher incidence of asthma later in life, and a lower incidence of allergies. In other words, rather than making a person anxious, depressed or psychotic, marijuana use just made it more likely that you would have asthma and and your hayfever might clear up.

This was picked up by the media in a rousing “marijuana is not bad for you” sort of way, with most articles focusing on the lack of negative physical and mental health outcomes. It was also picked up by the anti-drug research community, which is as you might guess, pretty huge and well funded, in a rousing “Bechtel et al. are wrong” sort of way. Of particular note was concern that the commonly accepted idea that marijuana use caused anxiety, depression, and most notably psychosis was being questioned. I won’t get too far into the details, other than to let you know that in the end, Bechtel et al. were compelled to write their “clarification”, which from what I can guess is their term for “we have a gun to our head so will tell you what they want us to say without actually admitting we were wrong”. And what were they feeling so compelled to say? This is where it gets really interesting, and I will paraphrase given that Bechtel et al. likely slaved over the wording of this:

After combining groups to create a single marijuana use group… “the group difference for allergies remained statistically significant (p<.04) but the difference for asthma did not (p=.10)…The group difference on psychotic disorder approached statistical significance and would have been significant if a more liberal test (i.e., one-tailed) was used.”

What is so important about this? From the standpoint of the original findings, pretty well nothing. The re-analysis, which simply combined three different types of marijuana user (early starter, life long, and late starters) into one group instead of looking at them separately, did not result in much change in the results. But it says a lot about the war on drugs, and the force of politics in research. This is because despite very meticulous reanalysis of more than a dozen different health outcomes, none showed any significant negative outcomes, and only the variable “psychotic disorder” managed to show even a weak association. And this association would only be considered meaningful if we used one of the weakest tests possible in research. For anyone with a bit of research experience, this kind of attention to detail is absurd. The entire point of doing statistical tests is to be rigorous, and the cut-off that determines what is an acceptable level of significance is well-established. If researchers were to start reporting every non-significant but “potentially significant” result, the world would be awash in trivial, misleading information.

So what is going on here that gets me so excited? Well, as the Queen in Hamlet once said “the lady doth protest too much, methinks”. Bechtold et al.’s findings were a shock to a contingent of researchers who believe that marijuana use is bad categorically. It posed a huge threat to entire federal agencies that are determined to press upon the public the harms of drug use. This article was a shot across the bow that suggested perhaps our assumptions about harm and drug use might in some cases be misplaced. Bechtold et al. stressed that the findings warranted further investigation. But the fact of the matter is that when a good study that cautiously posits modest results is forced to conduct a reanalysis, and then puts out a clarification that cites non-significant findings, we are not working for the betterment of science, but instead are revealing how politics and bias are just as much a part of research as they are a part of all the rest of our lives.

If we flash forward to the present, there are signs that Bechtold’s study heralded some important advances in marijuana research. For example, the commonly accepted view that pot use causes anxiety and depression is weakening, though its use does appear to increase the chances that substance problems will rear their head down the road (See Blanco, 2016 for more on this). Furthermore, along with several other researchers around the nation, Steven Lavoilette at Western University has been hard at work pulling apart the issue of psychosis and cannabis use, and his results suggest that while the THC component that gets you high may worsen psychosis, the cannabinoid component appears to improve psychosis. But wait, Steven has a vested interest in his work, seeing as he is part owner of a company that produces the cannabinoid producing strain of pot.

As with so many research questions, we have some ways to go before we can make any definitive conclusions about the effects of pot on your mind. With public sentiment shifting towards the legalization of marijuana, we shouldn’t be surprised to see some results emerging that appear to debunk the bold faced, black and white demonizing of a substance that for the most part has become nearly as popular as alcohol in our society. In my clinical work, pot seems no different from another drug that we all love to take – alcohol – in that both give us a lot of pleasure in moderation and can cause a lot of grief when our use of the substance spirals out of control. I see a trend with my physician colleagues to jump on the medicinal pot bandwagon, and already I have patients being switched to cannabis for the treatment of chronic pain. And on this issue of pot use and psychosis? Well, I just spent a year trying to convince one of my pot smoking, meth abusing schizophrenic patients to clean up his act, which he sort of did. Now he only smokes pot, and to my surprise and delight his psychosis cleared up entirely.

As for that multiple choice question I started off with? I’m afraid I can’t figure out the answer myself, but I do have a neat tagline for another eye catching article:

Miracle cure for allergies? Find out what researchers are saying now about pot use.

Study: smoking pot doesn’t make you anxious or depressed. 2