Lower-Risk Cannabis Use Guidelines: Will Users Listen?

In a highly cited and influential article, Fischer et al. proposed Lower-Risk Cannabis Use Guidelines (LRCUG) in the form of 10 recommendations based on a systematic review of evidence in the literature concerning health harms.¹ They recognize that harms increase with dose, and so counsel against higher potencies (recommendation 3) and higher frequency of use (recommendation 7), and observe that abstinence is the most effective way to avoid risks (recommendation 1), particularly for pregnant women and those with family histories of mental health disorders (recommendation 9). Other recommendations warn against particular practices such as driving while impaired (recommendation 8), using synthetic cannabinoids (recommendation 4), and smoking in general (recommendation 5) and in particular ways such as inhaling deeply (recommendation 6). The remaining two caution that early initiation is associated with worse outcomes (recommendation 2), and combinations of risky behaviors might magnify the risks (recommendation 10).

We agree that cannabis use would be safer if the LRCUG were followed, but they are not. For example, recommendation 7 in Fischer et al. states that “Users should be aware and vigilant to keep their own cannabis use—and that of friends, peers, and fellow users—occasional (e.g., use only 1 day/week, weekend use only, etc.) at most.”^1(pe4) However, adults using 50 or fewer times in the past year (equivalent to 1 day/week) accounted for just 3.6% of the past-year days of cannabis use reported to the 2017 US household survey (0.18 billion out of the 5.07 billion recorded by variable IRMJFY in the Public-Use Data Analysis System available at https://pdas.samhsa.gov/#). Broadening the notion of occasional use to 100 times in the past year (a proxy for weekend-only use) increases that share only to 7.3%. Because infrequent users consume less per day of use, their share of consumption is even smaller than their share of use-days. So, approximately 95% of current consumption violates that recommendation. Most users may follow LRCUG, but because heavy users account for a disproportionate share of consumption, most of the use does not.

Likewise, Fischer et al. warn that “Early initiation of cannabis use (i.e., most clearly that which begins before age 16 years) is associated with multiple subsequent adverse health and social effects.”^1(pe4) Yet 62% of current cannabis users in the United States report initiating by age 16 years, a proportion that rises to 74% among the 5.8 million who report using every day in the last month.

Fischer et al. seem optimistic that LRCUG could become more effective at altering consumption patterns after legalization. We are less sanguine. It is not as if safer use messages are entirely new. A “start low, go slow” approach is already promoted by Health Canada among other organizations and governments. Yet those messages have not prevented the use patterns just described.

Furthermore, commercial legalization of the sort pursued by Canada and the United States creates organized opposition with a moneyed interest in promoting greater consumption. Even as health authorities urge safer use, industry urges greater and sometimes riskier use. This suggests the need for meta-guidelines such as “If you hear procannabis information, check its source and ask whether the researcher or author has ties to industry.”

For example, Fischer et al. recommend that some populations, including pregnant women, should refrain from using cannabis (recommendation 9), and Colorado requires packaging to carry a warning about extra risks for women who are pregnant, breastfeeding, or planning on becoming pregnant. However, when Dickson et al. called 400 Colorado cannabis dispensaries, the majority (69%) recommended treating morning sickness with cannabis products.² After the study came out, the industry magazine High Times published an article entitled “Why I turned to cannabis for morning sickness—and why you shouldn’t judge.” The article’s subtitle was “Is cannabis the new wonder drug?” (https://hightimes.com/health/turned-cannabis-morning-sickness-shouldnt-judge).

Attempts to cajole consumers into lower-risk practices have to overcome not only contrary messaging from industry but also changes in price, potency, product form, and availability that may encourage higher-risk use. Fischer et al. warn that “High THC [tetrahydrocannabinol]-content products are generally associated with higher risks”^1(pe4) but legalization has brought sharply higher potency for flower products and a proliferation of edibles, vaping, and dabbing products whose potencies often exceed 60%.³ Legalization has also brought much lower prices per unit of THC^3,4; the prevalence and intensity of use tend to rise when price falls.

Likewise, Fischer et al. observe that “The use of both cannabis and alcohol results in multiply increased impairment and risks for driving, and categorically should be avoided.”^1(pe4) However, in the wake of Canada’s legalization, Anheuser-Busch, Constellation Brands, and Molson-Coors have all invested in or partnered with Canadian cannabis companies (Tilray, Canopy, and Hexo, respectively). These ventures have not yet announced intentions to market THC-bearing alcoholic beverages, but they would be poised to do so if regulations change to permit it.

Another question is whether guidelines for individual users are sufficient, or whether governments also have an obligation to regulate in ways that promote lower-risk patterns of use. The Ottawa Charter emphasizes that health promotion requires support for policy, systems, and environmental solutions, not just a focus on individual behavior change. Legalization forces policymakers to confront thorny issues concerning where and when can cannabis be sold, what products retailers can sell, and where those products can be used.⁵ The public health community needs to develop guidelines for how policymakers can answer such questions with regulations that promote lower-risk use.⁶

Government, higher education, and nonprofits also need guidelines for policies to prevent industry from influencing cannabis research. Article 5.3 of the WHO Framework Convention on Tobacco Control states that “Parties shall act to protect these [tobacco control] policies from commercial and other vested interests” (https://www.who.int/tobacco/wntd/2012/article_5_3_fctc/en). Something similar may be needed vis-à-vis cannabis. Harvard and the Massachusetts Institute of Technology accepted $9 million to study cannabis’s health benefits from a cannabis industry investor, and Thomas Jefferson University’s troubled Lambert Center has been accused of being too close to the cannabis industry.⁷

More generally, we believe that a more effective population-level way of preventing high-risk use would be keeping for-profit industry out of the business of supplying cannabis altogether; there may be other, safer models of legalization, such as restricting supply to government agencies or not-for-profit organizations.

We close by noting the potential tension that LRCUG can create for abstinence supporters. Some believe that people are going to use cannabis regardless of what the public health community says, so it makes sense to advise how to do so more safely. Others believe that there is no safe level of use and LRCUG may normalize use. For example, when somebody uses cannabis outdoors near a school, that might normalize cannabis use in the eyes of children even if that person were following all of the guidelines of Fischer et al.

Overall, LRCUG can be an important strategy for reducing health risks for those who consume cannabis, and the contribution by Fischer et al. is highly valuable in this regard. However, there remains a gap between publishing 10 abstract recommendations and actual behavior change, and there may be limits to how effective individual-level guidelines can be. Therefore, the existence of LRCUG should not be seen as obviating the need for other, complementary strategies for discouraging risky use.