Cannabinoid Profile Test Results

Washington Growers are the Best on the Planet – according to at least some of our accredited Labs

(Author’s note:  I initially considered titling this post “Setting Appropriate Retail Price Levels for Cannabis:  What the Feds (and, apparently, some of the Labs) Don’t Want Your Customers To Know”. … I decided that was not appropriately uplifting and “nice”, given the Season we are in).

Welcome to the third installment in my series assessing the results being reported by the 14 labs serving the testing needs of Washington’s State-legal Cannabis industry during the June/July/August period of this year.  This will close out my summarization of test results required by the WSLCB by discussing Foreign Matter screening  and Cannabinoid Profile (Potency) characterization.

Foreign Matter:

I don’t generally use traditional pie charts, as I believe their use without proper context can be misleading and that, even when used properly, they are very inefficient at communicating precise information when compared to a simple bar chart.

However, when the population represented by the pie is unambiguous and meaningful (as it is in this case – namely all of the screens conducted for Foreign Matter during June, July, and August of this year), and the slices of the pie are mutually exclusive and exhaustive (in this case passes vs failures), it can be a useful tool.  It is still an inefficient use of pixels, but pixels are cheap these days.

I chose to use a pie chart primarily so that I could look back one day and say that I did, but also because it actually does a good job of illustrating the primary takeaway regarding Foreign Matter screening.

After in-depth qualitative review, I have come to the conclusion that our bud is likely very very free of foreign matter.  I’m guessing if I came up with a statistic to use here, it would tell us that we are probabilistically very unlikely to see a failure (let alone lots of them).  It is almost as if Foreign Matter does not exist in Washington Bud.  Taken to it’s extreme this might lead to the conclusion that the only thing that exists in the Foreign Matter testing universe is pure, clean bud.  This is not surprising if the samples being tested are non-random, non-representative, non-PRAGUE in form).  It would have to be an inattentive grower that submits a sample for testing that contains enough foreign matter to merit a failure.  Either that, or a very attentive grower that is executing an objective, robust sampling protocol..

I do wonder about the utility of requiring a test that is almost never failed.  Of course, I also wonder about awarding trophies for participation, so take my value-laden opinions with a grain of salt.

I know that my gut and years of relevant education, training, and experience lead me to doubt whether non-random/non-representative sampling is compatible with having the ability to meaningfully generalize test results to the product actually finding it’s way onto retail shelves.  Every ounce of my being tells me it is not.

Since beginning my investigation of Washington’s Lab-reported Quality Assurance and Potency testing proficiency, I have increasingly found myself repeating the following mantra before going to bed:

“Washington Cannabis farmers are THE BEST and display extraordinary hygiene and quality assurance before their product is ever tested.  They make virtually no mistakes.  They are THAT GOOD.  Every single one of them.  Almost every single time.  I am so lucky to live in Washington – Land of the World’s Best Bud.”

The more I say it to myself, the more I come to almost believe it.  The more I come to believe it the lower my blood pressure becomes.

The reason it is important for me to believe is because I WANT it to be true.  I’m glad that the data appear to be largely supportive.  If, however, it is not really true, then some things that are showing up in the data must be errors.

Systematic errors.  Systematic errors that one might reasonably classify as being self-serving.

If what I repeat in the mantra every night is not true, then there appears to be something fundamentally wrong with at least some of the data being reported by at least some of the Labs.

To be frank, I don’t particularly care about Foreign Matter.  Perhaps I would if there was any evidence of it’s existence.  But it is a catch-22.  The data say it is almost never present.  Should I feel confident in the validity of this statement?  I don’t know.  But I do know that I keep repeating that mantra.  Perhaps I believe that the data are lying to me.

It is important that we all believe in the labels that assert the quality of our regulated Cannabis.  It is important because, if those labels are not accurate, then what benefit does that bagged/bottled/taxed stuff that one can’t even smell before purchase have over the stuff one can get from that now-out-of-work-priority 3 ex-dispensary owner forced out of business in the name of SSB 5052?  Forced out of business in the name of the people … in the name of the patients.

I would LIKE to believe that Foreign Matter does not exist in State-legal Cannabis.  I’d like even more to have a pattern of lab results that makes sense in a non-biased logical world.  That does not appear to be what I’m seeing now that I have summarized all of the test results.

No matter how often I repeat the mantra, I just don’t feel it to be completely true.

What is clear is that failures relating to the presence of large amounts of foreign matter are very rare and have been reported by only 3 of the 14 labs (in spite of the fact that ALL labs have conducted foreign matter tests).  I am, however, seeing an overall “pattern” of results across the different tests that seems oddly consistent.  That pattern will be summarized in the lab-level report card that I expect to publish later this month.

As it stands, Washington Bud is reported to be CLEAN (free of foreign matter).  I cannot say the same about Oregon’s bud.   Due to the fact that they do not tax bud in the unfair-competition market environment they have created down there, I find it difficult to believe that they have the same impeccable degree of oversight from the entire system (from growers through sampling through screening/testing through reporting) that we enjoy here in the true Northwest.  In spite of the attractive short-term economics of buying down there (state lines notwithstanding), I’d be hard-pressed to trust their relatively unregulated bud.  There is probably pizza mixed up in some of it.  They like their pizza down there in that relatively untaxed industry.  The farmers there can AFFORD pizza.

Cannabinoid Profiles (Potency):

On to that part of this series which I anticipate being of most interest to many readers:  Cannabinoid (Potency) testing.  It is becoming clear that the ability of our growers to acquire the pizza beloved by many of them is strongly linked to the reported Cannabinoid profile of their products.  If their product has Cannabinoid levels that are “too low”, it may not move at the price-points and/or frequency necessary to sustain their businesses – or appetites.

All of the tests up until this point have involved decisions of pass/fail for product based on quality attributes that relate to the safety of the product.  If the product fails, it is unsafe.  Unsafe product is not sale-able as such.

Tests conducted in a PRAGUE manner lend confidence to potential consumers that this relatively safe product has not been treated in a way that makes it less so.  These tests, and their validity, are crucial to public health.  Failing to maintain and expect consistently high proficiency from the labs is tantamount to putting consumers and patients at risk.  It is, effectively, trading short-term cash-flow and profit (and market-share for the labs) with the strategic viability (and health) of the regulated Cannabis market in this state … and the health of it’s consumers and patients.

The bottom line is that any failure to maintain proficiency in the functioning and reporting of the labs puts consumers (and patients) at risk.  Any lack of confidence in the quality of regulated Cannabis compromises the regulated Cannabis market.  It removes one of it’s primary advantages over the Dark Side.

Cannabinoid tests are different.  They are currently the primary window through which processors, retailers, and consumers can assess and grasp the “potency” and “medicinal-ness vs. recreational-ness” of any given product.  Cannabinoid results are central to knowing what psychoactive compounds are in the product and how much are in there.  They are crucial to knowing how much CBD is in the product.  They are crucial to knowing the CBD/THC ratio of the product.  They are crucial to not “pulling a Maureen Dowd”.

They also seem to be crucial in determining both the prices commanded by some products and the relative rapidity with which those products move through the system (this is an initial impression that I have developed while creating some tools intended to assist my consulting practice).

On to the results.  The pattern of change in average reported cannabinoid levels over time is dramatic and is worth detailing.  As such, I’ve decided to look at all I-502 potency testing performed on Flower between June of 2014 and August 2015.  (I’ve looked at all inventory types, but will only report on Flower in this series).

For each sample tested, values are reported for the % CBD, THC, THCA and “TOTAL” observed.  A consistent definition of  “TOTAL” was, apparently, unclear to the Labs until about May or June of this year.  Around May of this year, the Labs were directed to begin using THCmax when reporting “TOTAL”.  THCmax is the total amount of THC one would expect in the product if the THCA had been totally decarboxylated into THC.

THCmax is the metric that the AHP monograph (American Herbal Pharmacopoeia Cannabis Inflorescence monograph, ver 2014) suggests should be used to describe “THC” in situations where differing levels of decarboxylation will be present.  It is a way of keeping the reporting of “THC levels” consistent across testing methods.

Normally, most of the THC in dried/cured bud will be in the form of THCA.  Over time, it will degrade into (among other things) THC.

As it turns out, some of the methods used to assess the cannabinoid profile by the Labs appear to result in the decarboxylation of the bud in the process of measuring cannabinoid levels

Here is a graphic of the average levels of THC and THCA reported by each lab for flower across the 5 quarters for which I have data.

Comparing these two graphs, it is clear that Labs K and M are consistently reporting much higher levels of THC than their peers.  They are, at the same time, reporting much LOWER levels of THCA.  I’m guessing that these two labs exclusively use a decarboxylating test (likely gas chromatography).

If a customer’s attention can be focused on THC, the advantage of using GC to test Cannabinoid levels is clear.  This is part of the reason, no doubt, that the value of TOTAL is also required to be reported for tested product.  That is the good news.  The bad news is that what is being reported as TOTAL is clearly not THCmax (at least not consistently).

This chart shows the overall average levels (across labs) reported for TOTAL and CBD —- and the average level of THCmax derived from reported values of THC and THCA.   As CBD averages are much lower, they are plotted against the right vertical axis.

The primary takeaways from this chart are that average reported Cannabinoid levels are increasing rapidly and that what is being reported as TOTAL has historically been higher than THCmax.  During the most recent quarter, it would appear that the labs are increasingly following the new direction from the WSLCB to use THCmax when reporting TOTAL.

I’ve pained for the better part of a week about how to best illustrate my primary points on Cannabinoid profile characterization.  These points are that there is a surprisingly high degree of variability across the labs in their reporting, and that when ambiguity apparently existed as to what should be reported as “Total”, the choices made by many labs display a clear bias towards reporting values that are higher than the objective and appropriate norm.

I believe that the following chart does this effectively.

This chart displays the average difference between what was actually reported as TOTAL and what “should” have been reported (per the LCB’s recent direction to the Labs) as THCmax.

This value is reported as an average, per lab, for each of the (up to) 5 quarters each lab has been servicing the Industry.

Values greater than zero imply an “inflation” of reported potency levels for “THC” by the labs.

Values less than zero imply an “under-reporting” of potency levels by the labs.

Four take-aways from this chart:

When bias appears to exist (which is often), it has been overwhelmingly to over-report THC-based potency levels (compared to the THCmax standard).

Some labs (D, G, and J — and to a degree, A, K, and M appear to have been using THCmax all along ….or shortly after dialing in their processes over their first months of operation.

Some labs (N, L, I, H, and E) have been consistently over-reporting Total Potency.

Some labs apparently did not receive (or understand) the recent direction to begin using THCmax (Labs B and C) … either that or they are staffed by anarchistic types that do not appreciate being told how to do their jobs by a quasi-governmental authority.  Lab C, in particular, catches my eye, as their rapidly increasing over-reporting of potency almost suggests that they have been looking at these data and trying to emulate Lab N (which showed increasing over-reporting bias over the first 4 quarters but  rapidly came in-line with LCB direction in the most recent quarter).

This is an odd pattern of results and clearly demonstrates a lack of consistency across labs in what they are reporting.  Until recently, there was apparently no clear mandate (other than what is suggested by the AHP and good science and objectivity) regarding the definition of what, exactly, was required when reporting TOTAL.

With that said and done — I have had many questions from people that basically reduce to “which labs give the highest results”?

I am not going to say that here, as I have no intention of driving business to those whose data suggest that they may be inflating the results that they report.

I do, however, want to get this information out in a way that increases the PRAGUE-ness of reporting by the Labs.  Increased PRAGUE-ness would increase the quality of our cannabis.  It would increase the confidence of current and potential consumers in the quality of the product offered in regulated stores. It would increase the safety of the product being sold in those regulated stores.

There appears to be a vicious circle out there in which labs that report higher potency results (and consistently fail to fail product as often as their peers) tend to have developed a greater share of the testing market.  I am not implying any cause-and-effect here.  I do not know if this is a chicken or if this is an egg (or, frankly, why it crossed and/or rolled across the street).  I do not even know if it has gotten to the other side, or if it ever will.

I welcome your comments and suggestions regarding how best to handle the release of this information with the goal of increasing the validity (and consistency) of results being reported by our Labs going forward.

My initial thoughts involve letting Retailers know about these results (please point any Retailers that you do business with to this blog) and, perhaps, letting them know that I’ll eventually unblind the Labs so that we can all identify our favorite(s) amongst Labs A through N.

That way, the Retailers can apply pressure to the Producer/Processor/Lab consortia in a way that increases the accuracy (and PRAGUE-ness) of the results that are reported and placed on product labels.

In the meanwhile … I know that I’ll be asking which labs conducted the testing for any product that I’m thinking of buying in our regulated stores.  My purchasing decisions will, in no small part, be based on the lab used to describe the product.

Of course, I have the advantage of having seen my report card.  You will, also, if you check back here in a week or two.

In the meanwhile, enjoy your very clean–world’s-most-potent and optimally-dried-bud.

I hope that I have not compromised your illusions as to what is in your bowl too severely.

Happy Holidays to all.