Lower-risk substance use guidelines accessible by youth

To identify LRSUGs accessible to youth, we conducted a digital environmental scan and assessment of existing LRSUGs’ available to youth. We defined LRSUGs as any set of recommendations designed to help substance users identify and modify behavior to manage the negative effects of using substances [6, 8‐11]. Table 1 provides a list of the LRSUG definitions that we used for this study. To accomplish the digital environmental scan, we devised a novel method of searching the grey literature which was based on a rapid review framework [17, 18]. We streamlined our review methods by restricting our search to the first five pages of Google. This decision was made because we wanted a search process that would return results that youth might realistically encounter while looking for health information about substances. The use of the first five pages in Google was supported by previous research showing that the first few search results within search engines are a widely used source of health information among youth and has been used in previous studies [19]. This study was also designed to focus on information available through an internet search. We excluded other platforms (e.g., social media and physical distribution) primarily due to the difficulty in finding reliable predicable information on these platforms. Our review consisted of five steps (See Fig. 1).

In Step 1, we conducted a series of Google searches, using the search terms and phrases listed in Supplemental Table 1. Search terms were generated by using a smaller scale preliminary search, with search terms that lead to the largest number of relevant LRSUGs being selected for inclusion in the study. Given that Google’s search algorithm incorporates geographic location and search histories, this search was conducted using Incognito mode and with cookies turned off to limit any potential impact our specific geographical location on the search results [20]. We included LRSUGs from other jurisdictions due to using Google as our search engine. In Step 2, we extracted the returned search results from the first five pages of Google search, resulting in fifty links for each search term.

In Step 3, the titles of each search result were reviewed, and documents not related to using the substance were excluded (e.g., cultivation guidelines). In Step 4, we visited the web pages for the returned results and applied our exclusion criteria. LRSUGs were excluded if they [a] were not written in English, [b] did not have any LRSUGs; [c] did not contain any information related to the low-risk substances we identified for this study; [d] were not accessible (e.g., due to a paywall); or [e] if they contained abstinence only recommendations. We did not exclude LRSUGs on the sole basis of not containing youth specific information because youth can still access and use the information in these LRSUGs.

In Step 5, we coded and extracted information from the identified LRSUGs. Variable codes were identified a priori. All codes were designed as categorical or binary. The codes were generated from the information contained within the LRSUGs identified in a preliminary search. There were 15 codes in total capturing information about [1] the organization publishing the LRSUGs (i.e., Health [e.g., public health units], Government, Provincial, Municipal, Academic, Third-Sector [e.g., a retailer or drug user union]), [2] the drugs included in the LRSUGs, [3] whether the guideline was tailored, or partially tailored, for youth, [4] whether the guideline cited evidence in support of its recommendations, [5, 6] whether the information contained tailored information for key populations and by gender, [7‐9] whether the LRSUGs provided recommendations about dosing, frequency of use, or timing of use, [10‐13] whether the LRSUGs discussed the legality of substances, positive effects of use, negative short term effects and negative long term effects, [14] whether additional resources (e.g., link/phone number to a treatment service) were provided or recommended, and [15] whether the LRSUGs recommended abstinence. Extracted data was stored in an excel sheet, with each LRSUG representing one row and each code representing one column. Using these data, the number and proportion of LRSUGs that included each code were numerically calculated.

A total of 136 Google searches were conducted, returning 6800 results. After removing duplicates, 2,634 unique documents were identified. Of these, 257 were identified as LRSUGs. Among the 257 identified LRSUGs, 127 were excluded because they did not meet our criteria, resulting in a final inclusion of 130 LRSUGs (See Lower-Risk Substance Use Guidelines for Youth—Guideline Coding Table—Additional File 1). Table 2 provides a summary of results. Most LRSUGs were published by health (n = 51, 39%) and third-sector organizations (n = 41, 32%), followed by provincial/state (n = 18, 14%), government (n = 14, 11%), municipal (n = 4, 3%), and academic (n = 2, 2%) sources. The most common LRSUGs related to alcohol (n = 40, 31%), cannabis (n = 30, 23%), and caffeine (n = 21, 16%). Only 2 (2%) of LRSUGs focused on prescription stimulants and only 5 (4%) focused on nicotine. Less than a fifth (n = 21, 16%) of the LRSUGs were youth-specific; while almost half (n = 58, 45%) were only partially tailored to youth (i.e., contains information specifically for youth, despite the public being the target demographic). Over half (n = 84, 65%) of the LRSUGs included recommendations for key sub-populations (i.e., pregnant women, people with a predisposition to mental health complication, people on medication, and people with physical conditions that may be made worse with substance use), but only 25% (n = 32) contained sex and gender specific recommendations. Recommendations included in the LRSUGs considered dosages (83% n = 108,), frequency of use (n = 72, 55%), and when to consume (n = 86, 66%). Less than one-fifth (n = 23, 13%) of LRSUGs recommended abstinence as a first line response. Just one-quarter (n = 32, 25%) of the LRSUGs provided information about the legality of substances, 43% (n = 56) discussed positive effects of using the drug, 58% (n = 76) discussed short-term negative effects, and 53% (n = 69) discussed long-term negative effects. Finally, 38% (n = 50) of the LRSUGs were supported with appropriately cited evidence and 39% (n = 51) provided links to external resources to support youth who use drugs. Example recommendations for cannabis use are displayed in Table 3, demonstrating significant variability and diversity across recommendations (e.g., dosing, timing of use, age of initiation).

We conducted a digital assessment to identify and characterize LRSUGs for widely used substances (i.e., cannabis, alcohol, caffeine, LSD, psilocybin mushrooms, nicotine, prescription opioids, and prescription stimulants). In doing so, a number of key observations were made about existing LRSUGs. First, we observed that a variety of organizations were involved in the creation and publishing of LRSUGs for youth, with health organizations and third-parties accounting for the largest number of LRSUGs. Little research has examined what organizations and sources youth who use substances rely on the most when seeking information about substance use, nor if the presentation and format of LRSUGs from the organizations and institutions that produce these LRSUGs are acceptable and useful to youth. This is especially important with the emergence of platforms such as Tik-Tok and Twitter that may be underleveraged in existing knowledge translation strategies. Moreover, young people who use substances and access social media platforms for their health information may be exposed to various recommendations that come from other users. This information may be incorrect and could also be harmful to users. Future research should examine how both the publisher/platform and format affect the quality of information available and how these factors affect both the accessibility and uptake of LRSUGs.

Second, we note that the lack of recommendations tailored for age (the focus of our analysis), gender, physical disability, and mental health complications, reflects the need for multi-stakeholder approaches in setting standards for the development and promotion of these harm reduction tools that account for these specific populations of youth. Indeed, untailored LRSUGs can cause harm. Moore et al. [21] examined how well teenage drinkers follow low risk drinking guidelines, which recommended teens abstain for as long as possible, but that they should follow the adult guidelines if they do decide to drink. The authors found that low-risk teenage drinkers suffered a number of negative effects associated with high-risk alcohol use, despite their adherence to the lower-risk alcohol guidelines [21]. Similarly, given that substance use can affect specific populations (e.g., people predisposed to a psychiatric condition) more intensely, LRSUGs must consider their target audiences carefully [6, 7]. A particularly relevant finding was that only a few LRSUGs provided sex and gender-specific recommendations. Many substances will naturally affect people differently according to their biological sex and, in some cases, gender identification [13, 14]. Clearly it is important that future LRSUGs are made with specific populations in mind as one-size-fits-all approaches fail to accommodate the diverse needs of particular groups of youth.

To minimize the potential harms experienced by young people who use drugs, it is important to develop evidence-based LRSUGs that are specifically tailored to youth. Jenkins et al. [22], conducted a study investigating youth perceptions, experiences, and harm reduction strategies as they relate to cannabis; one major finding was the difference in these domains based upon geographic, cultural, peer, and political contexts. This is unsurprising considering that certain cultures may have traditions relating to certain substances. Given these realities, future research should evaluate how accessible and appropriate general LRSUGs are for different populations and demographic groups. It is necessary that stakeholders are engaged in the development of public health resources relevant to them by recruiting them for a variety of roles in the project, such as recruiting youth to be co-investigators on projects aimed at youth [16, 23‐25]. This will lead to better evidence-based LRSUGs which are tailored to the specific needs of youth and will avoid any harms from unsuitable or non-evidence-based LRSUGs.

Third, we found that most LRSUGs related to cannabis, alcohol, and caffeine. Relatively few were available for prescription medications, LSD, psilocybin, and nicotine. These drugs are used by a sizable population of Canadian youth; approximately 10–30% use e-cigarettes, up to 7.7% use psilocybin mushrooms, 7% use prescription medications, and up to 4.3% use LSD [3, 26]. These substances also pose considerable risk if misused, most notably tobacco and prescription opioids [27, 28]. Marshall et al. [23] found that there are limited harm reduction strategies targeted towards young people who use prescription opioids. Similarly, Faraone et al. [29] found that most of the research regarding non-medical prescription stimulant use primarily focuses on college aged users, with very little information for youth users. LRSUGs are also needed for less normative/more stigmatized and less commonly used substances, given that finding information about these drugs from unofficial lay sources might be difficult for some users, and that information may not be evidence-based.

Fourth, the LRSUGs we reviewed presented inconsistent information. Recommendations about abstinence, dosing, timing of use, and frequency of use were common, but not always considered. For example, Table 1 provides a characterization of the recommendations found within various cannabis guidelines and demonstrates significant variation in recommendations regarding dosage, timing, age of onset, and consumption methods. Similarly, there was a lack of consistency in reviewing the positive and negative effects of substance use – including information about the criminality and legal risks associated with using drugs. These differences may arise from the ideological and evidentiary underpinnings of the LRSUGs and organizations developing them. Indeed, evidence from reliable academic sources was cited in less than half of the included LRSUGs. This is problematic given that youth may find it difficult to evaluate or trust the accuracy of information found online [30]. Providing solid scientific evidence and youth support for recommendations is critical to maintaining the integrity of LRSUGs while addressing the needs of youth.

Finally, we documented a clear missed opportunity for facilitating treatment and care, with less than half of LRSUGs providing links to external sources that could support youth who use drugs. Youth can take advantage of these links to get referrals to additional support, information, and treatment. Despite the advantages of these resources, youth may not be able to fully utilize them due to issues related to legal consent, program times, cost, access, and the appropriateness of the program for the individual. Presently, there is little research into if and how people who use substances understand the self-referral sections of drug education literature (e.g., brochures, pamphlets, LRSUGs designed to provide information on harm reduction) in search of treatment. Despite this, other harm reduction interventions have been found to increase the willingness for users to change their habits and increase the number of users entering other treatment programs [31, 32].

This study is limited by the difficulty of conducting searches within the grey literature. To conduct our study, we used a novel method of searching this area of literature. Indeed, there is a lack of a databases for LRSUGs for both youth and the general population. As such, our approach was designed to replicate the way an English-speaking youth would access this information and therefore may not be inclusive of all LRSUGs available on substance use. Therefore, a formal scoping review of all the LRSUGs available to youth was not undertaken as our review was meant to characterize and analyze the information ecosystem youth are accessing for knowledge on safe drug use. Given that most youth who search for information do so at a surface level and spend little time evaluating the information for its relevancy and accuracy, instead preferring rapid information [31], our approach to review the first few pages of each Google search aligns with the search patterns that a youth might typically engage in while still being comprehensive enough to generalize the content of these LRSUGs [21]. The language and search terms we used were also non-exhaustive and were limited to possible search terms a youth would use. We rationalized this decision with the large number of duplicates we encountered during our search. If a more comprehensive search terms were used, it is likely the percentage of duplicates to unique articles would be skewed even further due to the nature of Google.

We did not investigate the health information eco-system that exists on social media platforms. We rationalized this decision due to the constantly changing nature of social media, the difficulty in searching for specific information on the various platforms, the retrievability issues of prior found information, the known unreliability of information on these platforms, and the fact that visibility on these platforms is determined through popularity. Similarly, we did not investigate the health information and LRSUGs provided by small scale platforms and formats such as schools, religious groups, and community centers. Future studies should investigate the health information eco-system that exists on social media and the smaller scale platforms (e.g., school LRSUGs), and the differences, disadvantages, and advantages between the various platforms.

Our study also focused on the documents that were relevant to youth who currently use substances; thus, we excluded any documents that did not provide any recommendations to make substance use safer, such as those focused on abstinence only messaging. While this may be a limitation for our study, we justified this decision because of the anticipated difficulty in identifying the numerous sources that discourage youth from substance abuse, as well as the lack of useful information within these documents that pertain to youth substance users.