As there have been no previous literature reviews describing risk and protective factors for negative consequences of drugs other than alcohol, this review builds on extant literature documenting the complexity of interactions that contribute to risk for negative consequences associated with cannabis use behavior. In line with previous reviews, ²⁸ there was a modest, yet varied, association between frequency or quantity of cannabis use and experiences of related negative consequences—a finding also observed with alcohol use consequences. ¹¹⁶ It is likely that the associations between cannabis use frequency and a sum of predefined number of self-reported negative consequence items represents an artificial upper boundary of associations between cannabis use and adverse outcomes ²⁸ or that self-report assessments do not sufficiently capture individual differences in patterns of substance misuse (ie, patterns of consequences may be unique to individuals or there are adverse outcomes that are not being assessed in existing measures leading to underreporting). In examining negative events resulting from cannabis use as independent outcomes, compared with a sum of experiences, researchers may be better positioned to identify person-centered patterns of CRNCs and develop effective secondary prevention programs to reduce immediate risk of harm and chances of escalating to the threshold of CUD. In addition, researchers can characterize cannabis misuse with greater specificity by ascertaining multiple measures of use—including frequency, quantity, and duration of use—that contextualize the patterns of use amongst those experiencing CRNCs.

Similar to reviews on alcohol-related negative consequences, ¹⁹ most studies included in the present review investigated intrapersonal (ie, psychological) predictors of CRNCs. Mental health issues, such as anxiety and psychological distress, appear to be important correlates of CRNCs and should be addressed in substance misuse prevention programing. Some individuals undergoing treatment for a substance use disorder might also require services that address concurrent mental illness, ¹¹⁷ a finding that underscores the need to address ongoing mental health concerns among individuals reporting substance misuse. ¹¹⁸ Personality traits might also play an important role in our understanding of who does and does not experience CRNCs as they do in research on alcohol consequences. ¹⁹ A review of controlled laboratory studies ¹¹⁹ identified impulsivity as a facilitator and consequence of drug use which suggests that some personality traits, like impulsivity, are exacerbated by drug use and might contribute to more frequent, and perhaps more severe consequences as use escalates. In fact, low levels of impulsivity has been described as a predictor of treatment success. ¹²⁰ The next logical step in this line of inquiry is to explore whether interventions to lower impulsivity would predict success in secondary prevention interventions to reduce CRNCs among those who do not have a clinical disorder (ie, CUD).

Such associations may not be limited to impulsivity alone as we found evidence that sensation- and risk-seeking might engender problematic cannabis use behaviors as well. As demonstrated previously, sensation-seeking is a quantitatively distinct phenomenon from impulsivity ⁶⁸ and more work is needed to explore how these unique traits influence problematic cannabis use. To accomplish this, longitudinal studies that can establish baseline personality traits and examine trends over time as individuals initiate and escalate their cannabis use are needed. Motives, but not expectancies, were investigated frequently as a predictor of CRNCs with coping motives emerging as a robust risk factor.^38,73,79 As suggested by mediational findings, individuals using substances to cope with underlying mental health issues are likely at risk for problematic substance use. The only protective factor identified in this review that would be more amenable to modification were protective behavioral strategies (PBS)—a trend that aligns with alcohol research. PBS were generally defined as the approaches individuals used to reduce the quantity and frequency of cannabis use in order to avoid adverse outcomes (ie, CRNCs). Of note, it appears no single PBS reduces the odds of experiencing CRNCs, but that the use of multiple PBS can attenuate the risk of experiencing adverse outcomes following cannabis use. As such, interventions should prioritize the teaching of multiple methods or consider developing individualized plans for users based on their perceived helpfulness of a particular strategy, self-efficacy to employ a specific strategy, and patterns of use. Researchers should also consider investigating other factors (eg, religiosity) that have been identified as protective against alcohol-related consequences,^19,121 but have not been examined in relation to cannabis.

While the literature has documented the influence of peers and family members on substance use initiation and frequency/quantity of use,^{122bibr123-11782218231166622-124} there is significantly less research exploring the role of peer/family influence on experiences of CRNCs. Given the clear association between peer/family use and alcohol consequences^19,125,126 more work is needed to understand the importance of peer and family influences to determine whether family-based or social network interventions could facilitate a reduction in CRNCs among cannabis users. Previous research has described the impact of social and cultural norms in initiation of drug use,^127,128 and findings from this review suggest that social norms also contribute to CRNCs. Injunctive norms may be particularly important for youth who have not fully established their personal and social identity and turn to peers for approval of their beliefs and actions. Conversely, norms-focused interventions (eg, normative feedback) have shown effectiveness at reducing alcohol use and related consequences ¹²⁹ as well as general substance use ¹³⁰ and should continue being used to offset pro-drug use social and cultural norms that might influence experiences of CRNCs.

Education and income were 2 important social determinants that had an inverse association with CRNCs. That is, the odds or frequency of CRNCs decreased as education and income increased. Policies addressing these social determinants can have sweeping, downstream effects on health and continued efforts to address ongoing economic disparities can reduce the economic and social costs associated with problematic cannabis use. At the community level, efforts are needed to reduce pro-drug cultures (eg, alcohol and drug use are commonplace in college culture) that promote heavy or more frequent use that can lead for example, to more college students experiencing CRNCs. Similarly, initiatives to protect vulnerable communities, such as sexual and gender diverse populations and veterans from experiences of trauma and stigma, important risk factors for SRNCs identified in the present review, may help mitigate risk for CRNCs. As policies challenging the legality of cannabis and new state-level policies across the US decriminalize or legalize recreational cannabis use; more comparative work is needed to understand the role of policy in the severity of problematic use.

We also investigated demographic differences in CRNCs. Overall, males appear to be at a greater risk of experiencing CRNCs relative to females which aligns with findings that clinical populations in treatment for substance use disorder are disproportionately male. ¹³¹ Males, particularly adolescent or young adult males, should be considered a priority population for cannabis misuse interventions aimed at reducing CRNCs. Far more research focusing on sexual and gender minority populations is needed as the evidence to date suggests that correlates and experiences of SRNCs among sexual and gender minority youth and young adults have not been adequately studied. Similarly, there is a dearth of literature examining racial/ethnic differences in CRNCs. The samples for many studies included in this review were majority non-Hispanic White and female therefore more attention should be given to identifying risk and protective factors for CRNCs in racially and ethnically diverse samples.