Testing labs: How Far We Still Need To Go


Testing laboratories are essential to the future of cannabis use in the United States. Today’s cannabis laboratory testing does not meet current needs and it’s questionable whether it ever will, without significant changes. Those statements are a huge oversimplification of the present and future for laboratories, producers, manufacturers, distributors, and retailers—but they ring true. When we compare cannabis to other markets that rely on laboratory services, it’s clear that we are not providing/receiving what we should. This is true even if we only look at the capacity of existing laboratories and completely discount the willfully fraudulent analysis that is known to occur in our industry.

So much is brand new in cannabis. Many experienced hands in cannabis must permit, comply, and qualify in technical areas far outside their previous experience to remain viable and sustainable. Past resistance to seeking outside help for cannabis problem solving is on the wane, but transparency in laboratory practice is nowhere near what is found in comparable markets.  

Most for-profit testing laboratories support science, engineering, industry, or medicine. They use accepted, published methods and disclose their quality assurance procedures and quality control data as a normal course of business. Laboratory analytical services is a mature industry. If we look at laboratory services in agriculture, drinking water, engineering, pharmaceuticals, and foods, the following are general expectations, if not required by law or regulation. The laboratory:

1.Will generally be accredited through inspection, auditing, and performance testing by an independent third party and/or a regulatory agency, and specifically approved for each analysis they perform.

2.Has a written, functional, data-driven, and audited quality assurance and quality control program.

3.Uses methods that are in the public domain or makes the methods used and performance expected available to the client.

4.Will have a formal qualifications package that presents summaries of items 1–3 above, along with resumes/qualifications of principal scientists and laboratory organizational information.

5.Provides auditable data that conforms to the method requirements, laboratory performance as presented in the qualification package, and QA/QC program. Calculates and reports errors in the data based on uncertainty in the analysis. 

The expectations above are not outlandish.  

1.We should expect that a cannabis lab can prove to an outside agency that it is doing what it says it is doing at any time. We expect the same for the folks that make and test our other pharmaceuticals.

2.You can’t “wing it” on quality. Useful quality processes are formal, written, and audited. Quality goals must have metrics to measure achievement. Things go wrong in labs; you need a plan on what to do before out-of-control events happen. Quality problems are communicated to clients.

3.Providing lab services using strictly published and/or publically disclosed methods is only controversial and complex for the cannabis industry. In all other professional lab service markets, it is standard practice to use standard methods. In lab services involving public health, engineering, or the environment, it is unheard of to use “black box” methods. Transparency in cannabis analysis starts with publishing your methods or using a published method.

4.Lab clients deserve to know the who, what, where, and why of laboratory services. Analytical laboratories are operated and staffed by qualified, degreed scientists. Equipment and methods used should be completely transparent to the client. Commercial laboratories often market their services on how well quality is maintained and the numerical standards that the laboratory can achieve. They publish the standards used to determine whether an analysis is in control or out of control. Supplying a qualifications document to clients before services are rendered is an industry standard for environmental and engineering analytical laboratories.

5.Analytical tests can only produce an approximation of the “true” value of the component analyzed. No analysis is ever 100% accurate; there is always error. Responsible use of data requires that the error level be known. Analysts and quality professionals know that every analysis can be characterized by its precision, accuracy, completeness, and comparability. Despite that, the current industry standard in cannabis is not to present run-specific precision, accuracy, or other quality information to the client. Providing just a value is only half the answer,  particularly when it’s already known by the laboratory that the value is always an approximation.       

The author is unfamiliar with any laboratory in the cannabis industry that meets the five goals above—goals met daily by thousands of service laboratories in other industries. Cannabis labs are not wholly to blame—far from it. Other labs are not federally illegal. Other markets have had decades of methods development and governmental support. These issues create a unique set of problems not faced by others.  

Cannabis testing is not currently accomplished at the level found in mature service labs in other industries, but measurable progress is being made. Laboratories are seeking accreditation and developing and enforcing QA/QC programs. Some states are taking a leadership role in requiring accreditation, but there isn’t consensus yet among states on how to best regulate cannabis labs. Industry organizations such as the American Association of Oil Chemists and American Society for Testing and Materials are using academic and industry experts along with consumer consensus to develop and publish standards and methods. Performance programs such as Emerald Test operate within states with performance standards and cross-laboratory validation of results. These efforts are in their infancy and they need support.

Method transparency in other laboratory markets exists to encourage trust and provide for data comparability. The opaque methods currently used by most cannabis labs are untrustworthy due to lack of transparency. The ultimate goal of any lab should be to provide data that are trustworthy, reproducible, and comparable. The cannabis industry will never foster trust with the public, producers, manufacturing, or government without continued diligence and challenging work from these institutions.

On many fronts cannabis laboratories are seeking the performance levels found in other laboratory testing markets. There are, however, overarching issues that can’t be solved at the laboratory or even at the corporate level. One of those missing competencies is for data comparability. Commercial laboratories have long relied on programs where performance samples are distributed among labs to identify unknown quantities of materials. The laboratory can evaluate its performance against the standard values. To perform this work, high-concentration, ultra-high purity standards are needed, in addition to performance samples. Concentration standards and blind-testing samples relied on for performance testing in other markets are unavailable to cannabis laboratories.  

An analytical standard is a verified solution of known concentration that can be used to validate an analytical instruments response. The highest concentration of analytical standard that can be shipped across state lines is 0.3% THC. Performance standards suffer from the same low limit. There are fundamental limits on how accurate testing can be with low-concentration standards. Samples must be diluted to the concentration of the standards, decreasing data quality. Not all labs can analyze the same performance samples due to restrictions on shipping, and high-concentration standards must be manufactured in-house or obtained in-state. This creates a fundamental scientific problem. Without cross-laboratory analysis of differences between method performance, none of the opaque methods used are comparable. Without a laboratory cross-comparison performance testing method, performance is invisible to all but the lab doing the analysis—if then. Research-grade, high concentration standards are not broadly available, and no calibration can be better than your standard. Most of our broadly available cannabis standards have only one or two significant figures, therefore all of our analysis can have no better than two-digit precision. That translates directly to the dose accuracy of any cannabis-derived product, or mass balance at any cannabis manufacturing facility. All are reported with unknown accuracy because method accuracy or error rate is not provided with values in cannabis lab reporting.

Is it too bold to say that the general public cannot rely on any current cannabis potency analysis? The low concentrations of our analytical standards make for imprecise and inaccurate results, and the uniform use of proprietary or unpublished methods eliminates the inter-laboratory performance analysis that other industries count on—make determining a labs level of precision impossible. Without accuracy or precision, what do we really have in a test result?

The need for transparency and trustworthiness of laboratory-generated cannabis data cannot be overstated. Few decisions in our industry can be made without relying on laboratory-generated data. The public trust in cannabis product quality is directly linked to laboratory transparency, accreditation, and accuracy. Other laboratory analysis markets outperform the cannabis laboratory industry in all quality metrics. Overall, cannabis laboratory data are currently untrustworthy when evaluated against similar laboratory data from comparable markets. The lack of inter-comparability testing of methods, opaque testing methods, and suitable standards are the greatest current hindrances to increasing the trustworthiness of cannabis products.

A great number of challenges and hurdles are in the way of a reliable cannabis testing industry, but they are not insurmountable. There are bad actors in this space but there are also people who are doing everything right. As the cannabis testing sector matures and people at organizations like American Association of Oil Chemistry, American Society for Testing and Materials, Association of Official Agricultural Chemists, and the American Chemical Society’s Cannabis Chemistry Subdivision continue their excellent work, the hurdles will become lesser and the challenges will become fewer.


greg miller.jpg

Gregory P. Miller, PhD, is the owner/head chemist of X-Ray Pharms in Socorro, New Mexico. Miller is an applied chemist with over 25 years experience solving complex technical problems. He founded X-Ray Pharms in 2015 to manufacture cannabis-derived products. For more information about X-Ray Pharms, visit xraypharms.com.

Scholarship half page ad - low.jpg