The present study used the DISCERN instrument to assess the quality of consumer health information provided by 33 websites of vendors selling cannabis to consumers in Canada. Although online vendors facilitate easier access and convenience to individuals wishing to purchase cannabis, the quality of health information on these websites was generally found to be low. Across the 33 websites, the average of the summed DISCERN scores was 36.83 (SD = 9.73) out of 75, and the mean score for overall quality of the publication (DISCERN question 16) was 2.41 (SD = 0.71) out of 5. Many of these websites failed to discuss uncertainties in research evidence on cannabis, the impact of cannabis use on quality of life, alternatives to cannabis use, risks associated with cannabis use, and lacked references to support claims on effects and benefits of use. These findings warrant concern, considering that patients may use these vendors as primary sources of information prior to purchasing cannabis. Physicians and other healthcare practitioners (HCPs) should be aware of these findings and appropriately caution patients who express interest in cannabis use.
There may be a few reasons why the quality of consumer health information found on most websites are low. First and foremost, the commercial nature of many websites causes them to present information in a sensational rather than objective manner, aiming to persuade consumers to buy their products. This consequently causes much of the information to be biased and unbalanced, with greater focus on the benefits of cannabis use and less focus on the risks, as evident by our results and past studies [20, 21]. Presenting information that could potentially detract consumers from buying products, such as the side-effects of cannabis use, is not in the interest of many online vendors [22]. Additionally, as evident by the aims and goals provided by many of these websites, they are primarily concerned with selling their products as opposed to educating consumers [23]. In our study particularly, websites scored low in many domains as they lacked essential information on cannabis topics including risks of treatment, impact on patient quality of life, and uncertainties in research, among others. Moreover, research on cannabis is an emerging field, therefore clear information on the risks and benefits of use may not be easily found by many website owners. In general, there is concern that website owners may lack the expertise in interpreting and providing health-related information [24–26]. There is also a significant amount of misinformation present online about cannabis [27], and website owners may be influenced by this misinformation without thoroughly reviewing peer-reviewed literature or well-established health information websites.
Comparative Literature
This is the first study to broadly evaluate the quality of health information provided by online cannabis vendors selling to Canadian consumers, with no restrictions on health conditions or purpose of use. Ng et al. recently used the DISCERN instrument to evaluate the quality of web-based consumer health information at the intersection of cannabis and pain [28]. Although the averaged DISCERN scores were found to be higher than those in our study, the quality of health information in this area was still concluded to be poor. However, considering that cannabis is commonly used by consumers for a wide range of conditions other than pain, it is important to broadly evaluate the quality of consumer health information without excluding conditions other than pain. Additionally, this study included health portal, professional, cannabis news, non-profit, and commercial websites from the Netherlands, United States, and Canada, whereas our focus was specifically on commercial websites selling cannabis products to Canadian consumers. As a result, we included and thoroughly evaluated a greater number of websites of this type.
Similar studies have been conducted in jurisdictions other than Canada, primarily in the United States. In 2014, Boatwright et al. evaluated the quality of medical marijuana claims on popular websites determined by online marketing tools, in which they found that 76% of claims made by websites were inaccurate and were based on low-quality evidence [29]. However, this study had significant methodological differences compared to ours, as they only evaluated the top ten most popular websites, and specifically focused on medical marijuana, excluding websites that sold cannabis for recreational purposes. Additionally, this study evaluated the accuracy and quality of only three medical cannabis claims on each website, as opposed to assessing the entire website. Three other studies from the United States (Luc et al.; Cavazos-Rehg et al.; and Kurger et al.) analyzed the content provided by online cannabis retailers, concluding that many dispensaries made unsubstantiated claims about the benefits of cannabis for various conditions, such as nausea, depression, and anxiety [30–32]. Aligning with the findings from our study, Luc et al. and Kruger et al. also reported that there was limited mentioning of potential side effects or risks associated with cannabis use [30, 32]. Moreover, a study from the United Kingdom found that much of the online information on medical cannabis would raise unrealistic expectations of benefits and downplay potential side effects [20]. In other forms of media such as newspapers and online discussion forums, the quality and accuracy of cannabis-related health information was also found to be poor [33, 34]. Furthermore, it is notable to mention that the aforementioned studies did not use the DISCERN instrument, which in addition to quality, assesses the reliability of consumer health information, as opposed to accuracy.
Implications and Future Directions
Patients may bring up their interest or use of cannabis to HCPs, who should be aware of the low quality of cannabis-related health information provided by online vendors as highlighted by our findings. HCPs should appropriately caution patients about these findings and refer them to sources of high-quality information, such as the National Center for Complementary and Integrative Health [35]. This will ensure that patients are adequately informed prior to purchasing cannabis online and may ultimately guide purchasing practices. Unfortunately, another concern relates to HCPs training and education on the topic of cannabis, both in Canada and across other jurisdictions. As many HCPs lack the necessary knowledge to effectively counsel patients on the safe use of cannabis, they are often reluctant to discuss this as a therapeutic option with patients [36–38]. With the known and rapid increase in cannabis vendors and the low-quality information provided by them online, it is of urgent importance to adequately train HCPs and HCP students so that patients have a reliable source of information to turn to for guidance. Public health agencies and those involved in cannabis-specific health policy may consider using the present study as a resource to inform healthcare providers and patients alike of the high frequency of low-quality information provided by online cannabis vendors.
Although our study evaluated the quality of cannabis-related health information provided by Canadian vendors, and similar research has been conducted in the United States, it would be beneficial for further research to replicate this study in other jurisdictions where cannabis has been legalized such as the Netherlands or Uruguay [39, 40]. Moreover, one study identified that source credibility had no significant effect on consumers’ interpretation of the quality of online health information [41]. Future research should also serve to evaluate the accuracy of information provided by online cannabis vendors, similar to the approach used by Boatwright et al. for medical cannabis, as this was not possible using the DISCERN instrument [29]. When doing this, a special focus should be placed on distinguishing between the type of sources used to support information (e.g., peer-reviewed literature versus blog posts), allowing for low-quality information and inaccuracies to be identified. Lastly, one study examined the implementation and effectiveness of online responsible vendor training for recreational marijuana in the United States, revealing that most employees were satisfied with the training and found it user-friendly [42]. Although this training was mainly focused on regulatory practices such as using the state’s inventory tracking system or checking for valid identifications, the authors suggest that cannabis-specific training on topics such as safety and dosing is a crucial future step. Such training may be beneficial for owners of online cannabis vendors in Canada as well as other jurisdictions, allowing them to incorporate important topics related to cannabis safety in online descriptions.
Strengths and Limitations
Our study had several strengths and limitations. One strength was the use of a validated and reliable instrument, DISCERN, to assess the quality of consumer health information on cannabis products. Another strength was the use of six different search terms on Google.ca, of which the first 40 search results were viewed for each (totalling 240 webpages), ensuring that the most frequently visited online cannabis vendors were captured and assessed. Moreover, an inherent strength of our study is that it is the first to assess the quality of health information provided by online cannabis vendors in Canada, with no restrictions to information on certain medical conditions.
One limitation to our methodology was that only websites with English-language content were eligible, potentially excluding websites in French (Canada’s second national language) and other languages. However, it should be noted that there were no non-English websites identified through our search. Additionally, considering that the internet is a constantly changing tool, we were only able to identify and assess websites at a certain snapshot of time. Therefore, if our methods are reapplied now, the content on many of these websites may have changed, and different websites may appear in the search results. Further, an inherent limitation of the DISCERN tool is that while it can be used to assess whether references and additional sources of information are provided by websites, it does not distinguish between the types of sources, such as peer-reviewed scientific literature versus blog posts, where the former would be deemed more credible in most cases and increase the quality of information provided by the website.