The lab analysis revealed that the industry offers what it claims in some ways, but potentially misleads or does not inform customers in others. On the one hand, it is clear that the vast majority of products (96.2%) fall within the legal ∆⁹ THC limits established by the 2018 federal Farm Bill (Agriculture Improvement Act of (US) 2018). Additionally, the tests conducted suggest that impurities are not a substantial issue for hemp ∆⁹ THC companies.

However, the investigation also revealed several points which may be concerning from a consumer perspective. Firstly, in many cases, the advertised dosages are substantially higher than would be allowed in any regulated, legal, high-THC cannabis market in the country. Secondly, most of the products (73.6%) do not include the cannabis universal symbol to warn potential consumers of their high ∆⁹ THC content. Despite the fact that most (84.9%) products are backed with a COA, since most of these do not include impurity testing, 75.5% of hemp ∆⁹ THC products are not safety tested. This problem is compounded by the lack of batch labeling for 50.9% of products. Finally, and potentially the largest issue for consumers is the fact that 66.0% of products differ from the ∆⁹ THC dosage stated on their labels by 10% or more, usually providing less than advertised.

Considering the potential risks of excessive amounts of THC (D’Souza et al. 2009), the combination of factors here could be a cause for concern. Customers in adult-use states expect the cannabis universal symbol or some variation thereof on high-THC products, with over 80% of states with medical cannabis requiring this, for instance (Kruger et al. 2022a). In many cases, even the advertised dosages exceed those considered acceptable in adult-use states. Additionally, two out of three products differ from the stated dose by over 10%, which makes it difficult—if not impossible—for customers to know how much THC they will actually consume. The problem from a consumer perspective is one of consent: people may buy under the assumption that the product is “just hemp” or at least not high-THC, and others may get more THC than they wanted or were informed of through labeling. These issues could be solved by the industry, but legislators might also take steps such as setting a total THC cap on hemp products or requiring accurate labeling.

Despite some concern about youth access to intoxicating hemp products (Akingbasote et al. 2022; FDA: Accidental Ingestion by Children of Food Products Containing THC 2022), the analysis revealed many problems with how companies attempt to prevent under-age purchases and access. In particular, only 8 of 53 products (15.1%) required any form of age verification prior to the order being placed. This means that in the remainder of cases, all people had to do prior to making a purchase is either input a date of birth or click a button confirming that he/she is over 21. These systems can be easily circumvented or simply lied to (Williams et al. 2020) and so do not constitute age verification. The remaining purchases had age verification through AgeChecker, which uses public records and images of ID to determine the customer’s true age. Additionally, only 1 product (1.9%) required an adult signature on delivery, and the remainder simply left the delivery in the mailbox.

While getting a valid card to pay may be a challenge, this is essentially the only thing preventing youths from accessing high-THC hemp products such as those in this study. From this point onwards, he or she could easily place an order (simply using an alternative website if they initially chose one of the 15% which performs age checks) and receive the delivery without having to present ID at any stage. In the absence of further laws—requiring a signature on delivery for intoxicating hemp products, for instance—this situation is difficult to rectify.

Finally, the lab analysis showed that 96.2% of hemp ∆⁹ THC products do fall within the limits established by the Farm Bill. This is a positive sign for the industry, noting that current legal opinion suggests that hemp ∆⁹ THC is legal in the absence of further state legislation. However, the finding that 49% of products used ∆⁹ THC converted from CBD may challenge this in some localities. Colorado (CO Department of Public Health and Environment (DPHE): Re 2021), for example, restricts ∆⁸ THC on the basis that, “chemically modifying or converting any naturally occurring cannabinoids from industrial hemp is non-compliant with the statutory definition of ‘industrial hemp product,” and Massachusetts (Commonwealth of Massachusetts: Hemp in Massachusetts: Faqs 2022) has a similar approach. These laws would also apply to ∆⁹ THC products advertised as hemp if levels were increased by conversion from CBD. Maryland is also considering (SB 0788 (Md.) 2022) regulations on THC products that are synthetically or artificially derived.

There are some limitations of the analysis. Firstly, owing to the nature of the market, the majority of products (86.8%) tested were gummies, candies, or other edibles, with relatively few drinks, tinctures, and vaping products. While this represents the market closely, it may be that other product types differ in some important ways (for example, being more likely to exceed the legal limit for ∆⁹). Unfortunately, the sample did not include enough non-edible samples to investigate differences by product type. Only one sample of each product was analyzed, but values could (and likely do) vary by the individual sample and not just by product. There was also some time (less than 2 weeks in all cases) between purchase and testing, which could feasibly have impacted the results through degradation, despite all products being sealed until testing.

The results show the consequences of the legal loophole which hemp ∆⁹ THC companies are currently operating in. With no centralized regulatory body, and very little in the way of state-specific regulations in most cases, intoxicating hemp products are currently allowed to operate with minimal oversight. This is why, for instance, that not all products are accompanied by a COA, why dosages (both advertised and actual) vary so wildly, and why in most cases age verification is essentially absent.

There are three avenues that could help companies and regulators find a potential solution to these issues. Firstly, states with adult-use or medical marijuana programs could incorporate intoxicating hemp products into their existing legal framework for legal cannabis. This would include, for instance, maximum doses of ∆⁹ THC (or indeed other THCs) per serving and per package, as well as lab testing requirements. Secondly, states could improve their hemp legislation to account for intoxicating products, enabling them to implement similar restrictions on THC-rich products and require age verification. Finally, the industry itself could adopt reasonable standards in the absence of official guidance. In the manner most companies now offer COAs for their products, they may (through hemp industry organizations) institute mandatory age checks and standardized testing requirements.