Marijuana & The Association Game

As an American I believe there are two things we get away with consistently in this great nation.   Subtle, non-deliberate prejudice and getting high.  From the ’60’s onward it has been difficult to ignore the cultural impact marijuana as made upon this country and its successive youth generations.

From parks, to classrooms, to hospitals, to your boss’ private stash that he or she still thinks no one knows about, we as Americans have already kind of given this potent plant a collective sigh and shrug as to what to do with it.  With legalization on the tip of every youth’s tongue, and states such as Colorado and Washington seeing enough tax revenue to pay for regulation, you’d think the health aspects of one of America’s oldest imports would be settled upon.

Your boss and I hung out recently. It was chill.

Except it is far from it.  The true issue is that marijuana is a lot more complex than we originally thought, especially due to its psychoactive component and the fact that it really isn’t recognized as either a stimulant, depressant, or hallucinogen.  It’s kind of a mix of all three,and even that depends on your individual reaction.

So how does that complicate things?  Well, mental health is a little tougher to parse than physical health at times, simply because science isn’t good at figuring it out yet.  Heck we’ve known depression has existed for the longest time, yet we still can’t make heads or tails of what really causes it.   So you can imagine how difficult it is to craft a really good study around marijuana addiction or its possible contribution to depression.

And that’s kind of the main issue.  We’re now stuck in this vicious cycle of trying to prove that anything can be coupled with marijuana.   Bladder Cancer?  Yup.  Parkinson’s?  Uh-Huh.  Heck, we even looked at how it affects the rate our bones heal. What’s next? How marijuana will help us jump 3 inches higher, when there’s a full moon, while watching Late Night with Stephen Colbert? How cannabis could freshen a dog’s breath?
Today I offer a glimpse into the mile wide breadth of marijuana research and offer a criticism as to why this approach could hurt efforts and ultimately lack the cost effectiveness of clinical, experimental studies.

A Brief Background

Cannabis and cancer have long been the metaphorical chariot of fire for the Pro-Cannabis crowd in the clinical field, while the placement of cannabis as a schedule 1 drug under 1970’s Control Substance Act  is considered a politically charged decision.  The studies that followed were far from it as the Investigational New Drug program from 1978-1992 gave American scientists and doctors the ability to experiment with and study cannabis on a case by case basis.

“For Science…”

From it, we learned of two mainstay chemicals of the plant delta-9-THC and cannabidiol, two cannabanoids.  Those aren’t the only two active compounds in the plant, but rather the two we’ve try to study the most, and have isolated to make into clinical drugs such as dronabinol.   In layman’s terms, the THC is the psychoactive component of the plant (gets you high, man), whilst the cannabidiol is considered more of the therapeutic component due to its ability to block cell growth, reduce inflammation, and act as an antivirus.

So why the hubbub?  What gives with the confusion as to what it does, and why do I see a marijuana related cure for cancer every other month on the Huffington Post?!

To understand this, we must first look at…

How the heck do we do science on this?!

For those of you apt readers who enjoy digging through pages of material and policy, here is the FDA’s official link towards getting a drug approved.  For the rest of you, I’m just going to give this the Buzzfeed treatment and cut out any contextual information for pure content to stuff into your brains like the Nike logo with any developing athlete.

“Look mom! We’re good at sports now!”

The only difference between a drug and a toxin is dosage.  The name of the game is essentially figuring out what is the appropriate dosage, and for that you need three things.  People, Animals, and Petri dish full of whatever cell you plan to test your drug on.   Let’s say we’re testing to see if Marijuana makes your armpits smell like a welfare queen burnout.

You begin by testing on armpit cells in a petri dish to see whether or not it produces that burnout scent.  You know, the one that smells like cold pizza and art magazine subscriptions? If that holds, then you’d want to move onto animals or animal cells.  Typically, you’d try two different studies with these animals (most likely mice), comparing animals not treated with cannabis with those treated and seeing if there is a difference in effect (case-control study) or checking to see what levels of dosage will correspond to what level of burnout smell (dose-response study).  These are known as Pre-Clinical Studies.

Oh look, an actual infographic

You take all that, and then write up a very nice application for an ethics committee , then proceed to human/clinical trials.   Funny enough, according to USA’s very own, we still have “No clinical trials of Cannabis as a treatment for cancer in humans have been found in the CAM on PubMed database maintained by the National Institutes of Health.” That isn’t to say that we haven’t done clinical trials of whether or not there is a correlation between cancer , but rather there hasn’t been one with cannabis as the treatment for cancer.  That’s because we are currently stuck in what I like to call…

The Association Game

What we’re stuck with is a very funny system of studying what I personally call the association game.   It’s essentially the matching of very odd effects through a statistical correlation.  For example, Kim Kardashian’s name being mentioned is commonly associated with eye-rolls and long drawn out sighs with social commentary between each breathe.   In the case of cannabis, and in my experience, in the case of other chemical-health studies, the literature is saturated with one particular type of study:  Cross-Sectional.

A cross-sectional study is taking a snapshot of a population and seeing what affects what outcome.  It’s one of the purest forms of checking for associations across a population.  For example, you want to do a study of a group of people in a town and whether or not they sport Donald Trump’s haircut.  You go out, get the basics about your study population, see whether or not they have what looks like a dead possum on their head, and boom.   You have a study that was cheap, quick, and gives insight into the town.

Oh wow…two infographics in a row. Psh, Try-hard.

Why is this a problem though?  See, with cannabis research expecting to receive almost $70 million dollars in the next 5 years through the federal government alone, that’s a lot of grant funding suddenly available for the NIH.   Lets also couple our rise in college graduates and rise in those pursuing degrees beyond a bachelor’s and we see a cohort of really hungry researchers looking for experience and a quick publication to stack their resumes.  Oh gee, I wonder what’s a hot button topic that can easily be given a cross sectional study and can be done relatively quickly.  Hmmmmmmm.

So now we have a lot of cross sectional studies floating about our literature.  I mean, a LOT of them.   No one has compiled the statistics behind it but I went ahead and searched up “cannabis” and “Meta-Analysis”  along with “cannabis” and “Review” from 2014 onward to see just how many cited studies were in fact cross sectional ones.   And guess what?  There are a lot of them.

This article on psychosis had every single source coming from cross-sectional analysis, 30 of them to be exact.  This one on fetal development?  Over half.  It took about three pages of digging on WorldCat to find one that included an actual meta-analysis on clinical trials for pain perception.   I’ll admit that my investigation into the issue wasn’t done the most effectively, but due to time constraints it’s the best I could muster.  Just take a look yourself and see how wide the breadth is for what people have looked into what marijuana can affect.  It’s practically everything.

I wasn’t kidding…

What is the problem with this though?  What is so bad about the association game when you’re trying to figure out the full spectrum of what a drug does?  Dr. Nina H. Fefferman at Rutgers University phrased this issue pretty eloquently.   They simply aren’t good at producing new data.  Drawing associations is great, seeing the strength of the associations is also good, but our statistical tests ultimately share a fatal flaw in all of these associationd and that’s why no one can definitively say whether or not marijuana causes anything.  All they can say with their observations and data sets is that there is a trend.  That’s it.  And you wonder why we haven’t really gotten anywhere with our research.

Slow and Steady wins the race

We’ve all heard of the story of the rabbit and the tortoise right?  Now, imagine that the rabbit is ‘cross-sectional observational studies’ whereas the tortoise is ‘clinical experimental studies’.  If they were to race towards the goal of gaining knowledge valuable to public health policy, the rabbit would lose, not because of its ego, but because it’s running in a circle.

Whoa whoa whoa…no need to get political here

Clinical trials and experiments are indeed more expensive.   They are also difficult to execute well due to how crucial screening out other non-factors are to the accuracy of the study.  There are studies on how to actually do proper screening in itself, which only goes to show how little room for error there is for these types of sciences.

However, like every season of Game of Thrones, you need to wade through that waist deep pond of monotony and hardship for that sweet, juicy finale. Let’s return to that statistic from I used in the beginning of this article.  There are NO trials for cannabis treating cancer. None. Forget the politics behind this plant, if the science isn’t there then how can we expect our representatives to put their weight behind utilizing some of the purported healing properties of marijuana?

These trials are hard, long, expensive, tedious, and other adjectives that can describe both clinical trials and your neighborhood gigolo.  But the payoff is solid: good science and concrete data that actually progresses our true understanding of cannabis.

The best way I like to view this is in a courtroom.  Imagine you’re trying to prove to a jury that marijuana is good, but all you can cite are studies that say “we’re 95% confident that in our sample of 300 people who have diabetes, 100 of them may have benefited from smoking marijuana.”  Not really convincing right?  Versus citing “a clinical trial looking at 30 patients in a controlled environment showed that smoking cannabis helped control back spasms. Here are the blood tests and EKG’s that show why this seems to be the case.” See the difference? One is merely drawing upon a statistical association, the other has a myriad of useful data and controls in place to give good insight as to why.

So don’t be surprised…

…if this whole worldwide legalization thing takes a while.  It won’t when we’re still so invested in the association game – not to say this is entirely a bad thing.  Association gives us an idea of what to look into in depth.  If there is an abundance of cross-sectional studies between weed and depression, then the topic certainly merits grant funding for a full fledged experimental study.   There are of course cohort studies that also shed valuable light on relationships between ingesting cannabanoids and other symptoms.  But, we cannot let the prospects of a quick publication draw our efforts away from establishing good science.  It’s hurting the constituents who have been patiently waiting for data that may help them with judicial or political cases.  Or high school drug dealers wanting to show off how much of a marijuana guru they are.   Mostly the latter.

Dude, that’s a sativa bro, get your facts straight

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s