Hostname: page-component-6898cfcb55-w5hgh Total loading time: 0 Render date: 2026-11-11T03:22:05.474Z Has data issue: false hasContentIssue false
  • English
  • Français

The association between cannabis and depression: an updated Systematic Review and Meta-analysis

Published online by Cambridge University Press:  12 February 2025

Victoria ChurchillOpen the ORCID record for Victoria Churchill[Opens in a new window],
Caroline Sutton ChubbOpen the ORCID record for Caroline Sutton Chubb[Opens in a new window],
Lucy PopovaOpen the ORCID record for Lucy Popova[Opens in a new window],
Claire A. SpearsOpen the ORCID record for Claire A. Spears[Opens in a new window] and
Terri PigottOpen the ORCID record for Terri Pigott[Opens in a new window]
Victoria Churchill*
Affiliation:
Department of Community Health and Preventive Medicine, Morehouse School of Medicine, Atlanta, GA, USA
Caroline Sutton Chubb
Affiliation:
School of Public Health, Georgia State University, Atlanta, GA, USA
Lucy Popova
Affiliation:
School of Public Health, Georgia State University, Atlanta, GA, USA
Claire A. Spears
Affiliation:
School of Public Health, Georgia State University, Atlanta, GA, USA
Terri Pigott
Affiliation:
School of Public Health, Georgia State University, Atlanta, GA, USA
*
Corresponding author: Victoria Churchill; Email: vchurchill@msm.edu
Rights & Permissions [Opens in a new window]

Abstract

Background

Cannabis is one of the most commonly used drugs globally, although its legal status varies across regions. Public support for its decriminalization has increased, but gaps in our understanding of the health consequences of cannabis use remain, particularly related to its impact on mental health. This article provides an updated systematic review and meta-analysis (previous being Lev-Ran et al., 2014) looking at the relationship between cannabis and depression.

Methods

Literature available before March 2023 was screened for longitudinal studies that included cannabis use and depression. Cross-sectional studies and those only looking at special populations were excluded. Studies must have also controlled for depression at baseline to allow for investigation of a temporal relationship. Extracted data included cannabis measures, depression outcomes, adjusted odds ratios, and study settings. Meta-analysis employed a random effects model with multilevel meta-regression for effect size moderators.

Results

The search yielded 1,599 titles from various databases, resulting in 22 studies for meta-analysis, including 14 from Lev-Ran et al. Eleven studies were US-based, with participants mostly under 18. Meta-analysis showed a higher risk of depression among cannabis users (OR: 1.29, 95% CI: 1.13–1.46). Risk of bias assessment showed medium risk across studies with exposure measurement being a key bias area. The funnel plot and Egger’s Sandwich test did not suggest publication bias.

Conclusions

This study underscores the association between cannabis use and depression but also emphasizes the need for further research, especially in understanding usage patterns, heavy use definitions, and long-term effects on depression risk amidst changing cannabis trends.

Keywords

cannabisdepressionmeta-analysissystematic review

Information

Type
Original Article
Information
Psychological Medicine , Volume 55 , 2025, e44
Check for updates
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2025. Published by Cambridge University Press

Introduction

Cannabis is one of the most widely used drugs, with approximately 2.5% of people worldwide currently using cannabis in some form (World Drug Report 2019; World Health Organization, n.d.). As of 2019 in the United States (US), about 18% of Americans aged 12 and older reported having ever used cannabis in 2019 (Centers for Disease Control and Prevention, 2021), and in 2020, there were 2.8 million new cannabis users in the US, of which about 1 million were adolescents aged 12–17 (Substance Abuse and Mental Health Services Administration, 2020). Laws and practices surrounding cannabis have evolved significantly over the years, varying both on the state and country levels. In the US, cannabis was initially classified as a Schedule I drug under the Controlled Substances Act of 1970, indicating it had “a high potential for abuse” and “no currently accepted medical use” (Eddy, Reference Eddy2010). However, in the following decades, individual states enacted policies regarding the criminalization of cannabis for both personal and medical use. Recently, there has been a growing push for the decriminalization of medical and recreational cannabis (Grucza et al., Reference Grucza, Vuolo, Krauss, Plunk, Agrawal, Chaloupka and Bierut2018; Mahabir et al, Reference Mahabir, Merchant, Smith and Garibaldi2020). Public support for cannabis decriminalization has increased sharply, and as of 2022 only 10% of adults in the US believe all cannabis should be illegal (Van Green, Reference Van Green2022). Nonetheless, there remain significant gaps in our understanding of the potential consequences and benefits of this drug.

One research gap is in understanding the relationship between cannabis use and mental health, specifically whether cannabis use leads to common psychological disorders such as depression. In 2014, Lev-Ran et al. published a meta-analysis investigating the development of depression among cannabis users who were not depressed at baseline (Lev-Ran et al., Reference Lev-Ran, Roerecke, Foll, George, McKenzie and Rehm2014). Their systematic review of the literature concluded in December 2012 and was not limited to any country, language, or age group. They included 14 studies that either assessed the clinical diagnosis or symptoms of major depressive disorder (MDD) or dysthymia (mild chronic depression) (Patel & Rose, Reference Patel and Rose2021). They found that cannabis use was associated with higher odds of developing depression among adults (Odds Ratio [OR]: 1.17, 95% CI: 1.05, 1.30). Lev-Ran et al. also conducted meta-regressions on the age of cannabis onset (aged 18 years or younger compared to greater than 18 years) and on heavy cannabis use but found that both were not significant (ps > 0.05) and underpowered due to the small number of studies included. Sensitivity analyses showed that the OR for developing depression among heavy cannabis users compared to non-users was also not significant (OR: 1.34, 95% CI: 0.96–1.87). Overall, this meta-analysis indicated a positive relationship between cannabis use and the development of depression over time. No differences were found by age of onset or heaviness of cannabis use, although the latter analyses were underpowered.

Since Lev-Ran et al. (Reference Lev-Ran, Roerecke, Foll, George, McKenzie and Rehm2014), there have been other systematic reviews and meta-analyses that explored the relationship between cannabis use and mental health. In 2018, Mammen and co-authors published a systematic analysis that looked at mental health outcomes among people with anxiety or other mood disorders at baseline. They found that recent cannabis use (within the prior 6 months) was associated with increased levels of mental health symptoms (Mammen et al., Reference Mammen, Rueda, Roerecke, Bonato, Lev-Ran and Rehm2018). This offers evidence of the relationship between cannabis and poor mental health outcomes, although all participants in the Mammen et al. article had a mood disorder at baseline. Another meta-analysis published in 2019 explored cannabis use in adolescence and mental health outcomes, including depression, in young adulthood and found a pooled OR of 1.37 (95% CI: 1.16, 1.62) (Gobbi et al., Reference Gobbi, Atkin, Zytynski, Wang, Askari, Boruff and Mayo2019). This is further evidence that there is a relationship between cannabis and depression, even though the authors limited their review to only include studies with adolescent cannabis users.

Several explanations for the cannabis-depression link have been proposed. One mechanism focuses on the role of cannabis use, especially during formative periods, in changing brain structure and regulations, (Langlois et al, Reference Langlois, Potvin, Khullar and Tourjman2021) with some evidence from animal models (Bambico, Nguyen, Katz, & Gobbi, Reference Bambico, Nguyen, Katz and Gobbi2010). Another implicates shared genetic factors that might be responsible for both depression and cannabis use (Hodgson et al., Reference Hodgson, Almasy, Knowles, Kent, Curran, Dyer and Glahn2017). Others focused on shared social or environmental factors (Degenhardt et al, Reference Degenhardt, Hall and Lynskey2003). Evidence supporting these mechanisms indicates that multiple pathways might be in play in explaining the relationship between cannabis and depression.

Considering the changing landscape of cannabis in the US and globally, an update to the Lev-Ran et al. meta-analysis is warranted. Besides growing social and legal acceptance of cannabis, over the years, the potency of cannabis products has been increasing (ElSohly et al, Reference ElSohly, Chandra, Radwan, Majumdar and Church2021; ElSohly et al., Reference ElSohly, Mehmedic, Foster, Gon, Chandra and Church2016). Modes of administration are also changing, with vaping cannabis becoming more popular, as well as using concentrated cannabis products (e.g., shatter, wax), all of which differ in terms of potency and delivery of the psychoactive substance (Prince & Conner, Reference Prince and Conner2019). Furthermore, it is particularly important to address the issue of mental health in the US. Studies have shown that rates of depression in the US have increased in recent years, especially since the start of the COVID-19 pandemic (Niedzwiedz et al., Reference Niedzwiedz, Green, Benzeval, Campbell, Craig, Demou and Katikireddi2021; Zheng et al., Reference Zheng, Morstead, Sin, Klaiber, Umberson, Kamble and DeLongis2021). The purpose of this article was to update the Lev-Ran et al. meta-analysis by including additional studies published since 2013 and to further explore the relationship between cannabis use and depression. In addition, we conducted meta-analyses looking at the nuanced relationships between early-onset cannabis use (i.e., initiating use in adolescence) and depression, as well as heavy cannabis use and depression. We hypothesized that those who start using cannabis at a younger age would be more likely to develop depression, compared to those who initiate use as adults, which may be due to the psychosocial consequences of early cannabis use that have been demonstrated in the literature (Meier, Reference Meier2021; Pacheco-Colón et al., Reference Pacheco-Colón, Ramirez and Gonzalez2019). We also expected that heavy cannabis users would be more likely to develop depression, compared to light/occasional users due to the potential for a dose–response relationship between cannabis and depression.

Methods

This review follows the PRISMA 2020 reporting guidelines (Page et al., Reference Page, McKenzie, Bossuyt, Boutron, Hoffmann, Mulrow and Moher2021). Although this review closely adhered to systematic review methodologies, the timing of the project did not allow for registration in the International Prospective Register of Systematic Reviews (PROSPERO).

To guide this review, we applied the PICO framework: Population (non-institutionalized youth and adults worldwide); Intervention (cannabis use); Comparison (non-cannabis users); and Outcome (depression).

Eligibility criteria

The literature search included literature from inception through March 31, 2023. Initial eligibility criteria were established a priori, but additional criteria were added after the literature search began. Articles could originate from any country as long as they were available in English. The inclusion criteria established prior to the literature search were: (1) studies had to report on cannabis use separate from other drug use; (2) studies had to report on depression, separate from any other mental health outcomes; and (3) studies had to be longitudinal, with cannabis use measured prior to the outcome of depression or controlled for at baseline. Exclusion criteria were: (1) cross-sectional studies and (2) other systematic review and meta-analysis articles. After the preliminary search, we decided to exclude studies that only focused on special populations that were likely to have a higher risk of both cannabis use and depression, to reduce confounders. For example, LGBTQ-only studies were excluded (Dyar et al., Reference Dyar, Bradley, Morgan, Sullivan and Mustanski2020) as it has been shown that gender and sexual minority populations have higher rates of depression than cisgender and heterosexual populations (Ferlatte et al., Reference Ferlatte, Salway, Rice, Oliffe, Knight and Ogrodniczuk2020) as well as higher rates of problematic cannabis use (Dyar et al., Reference Dyar, Feinstein, Newcomb and Whitton2021). In addition, studies that focused on war Veteran-only populations were excluded (Gunn et al., Reference Gunn, Stevens, Micalizzi, Jackson, Borsari and Metrik2020) because this population also has unique experiences that may lead to differing rates of cannabis use (Davis et al., Reference Davis, Lin, Ilgen and Bohnert2018) and depression (Liu et al., Reference Liu, Collins, Wang, Xie and Bie2019) compared to the general population. We excluded studies that only included participants who were seeking treatment for a mental health issue at baseline (Bahorik et al., Reference Bahorik, Sterling, Campbell, Weisner, Ramo and Satre2018).

Search strategies

Systematic literature searches were conducted by VC and CC. The initial search started with articles that cited the Lev-Ran et al. meta-analysis. Searches were then conducted in PubMed, Ovid Medline, and Google Scholar. Gray literature was searched using ProQuest for unpublished theses and dissertations. An ancestor search of the Lev-Ran articles was conducted by reviewing all the “Cited By” articles in PubMed. The full search strategy, including keywords and limits for each database, can be found in Supplemental Materials I.

Screening

The screening was conducted in three stages (initial screening of titles; title and abstract screening; and full-text screening) on Rayyan (Ouzzani et al., Reference Ouzzani, Hammady, Fedorowicz and Elmagarmid2016) using the eligibility criteria specified above. VC conducted the original screening of potential studies published through March 31, 2022, with no specified beginning date. Both VC and CC conducted the screening of potential studies published between April 1, 2022, and March 31, 2023. Disagreements between VC and CC were discussed and TP acted as the tie-breaking vote for seven articles when agreement could not be reached. Detailed screening results, including citations of excluded articles and reasons for exclusion, can be found in Supplement II.

Extraction

Study information was independently extracted by VC, with TP serving to assist with extracting data when clarity was needed. The coding sheet can be found in Supplement III. The final codes are detailed below.

Cannabis Use Measure. Cannabis use in any form was the exposure of interest. Codes specific to cannabis included: how cannabis use was measured (for example, self-reported frequency, or a validated measure such as the Composite International Diagnostic Interview [CIDI]). Specific definitions of exposure were collected. These ranged from ever use to cannabis use disorder. Most of the articles included analyses for different levels of exposure, so these were extracted as separate effect sizes. For example, Pedersen (Reference Pedersen2008) reported on participants who had used cannabis 1–10 times in the past 12 months, as well as those who used cannabis 11 or more times in the past 12 months.

Although we intended to code for the method of cannabis use (i.e., edibles, inhaled), this was not reported consistently across eligible studies. However, we were able to code for early onset/adolescent use which we defined as cannabis use prior to age 18, and heavy cannabis use. While heavy use was not universally defined, we determined that studies reporting outcomes for “cannabis dependence,” “cannabis use disorder,” “diagnosis of cannabis abuse,” “chronic cannabis use,” “persistent use” and “daily use” met the criteria for heavy use.

Depression Measure. The outcome was any type of clinical or self-reported depression, including major depressive disorder (MDD), major depressive episode (MDE), and general depressive symptoms. We coded for the specific measure that was used (e.g., Clinical Interview Schedule [CIS]-Revised). We also coded if the depression measure was obtained via a self-report survey, structured interview, or indeterminate. Other mental health issues, such as anxiety (Feingold, et al., Reference Feingold, Weiser, Rehm and Lev-Ran2015; Kedzior & Laeber, Reference Kedzior and Laeber2014) and schizophrenia, were not coded due to the complexity of additional codes while considering the limited resources of the project.

Results from Individual Studies. Adjusted odds ratios (aORs) for the depression outcomes were extracted from all studies that controlled for depression at baseline. If studies reported multiple outcomes for different exposure categories, we extracted those into separate effect sizes. For example, Feingold et al. (Reference Feingold, Weiser, Rehm and Lev-Ran2015) reported aORs for those who used cannabis less than weekly, weekly to less than daily, and daily, resulting in three effects for this meta-analysis (Feingold et al., Reference Feingold, Weiser, Rehm and Lev-Ran2015). Overall, we extracted 1–5 effects from all the articles (median: 2 effects per study). When studies reported multiple aORs for a single effect, we chose the one that was adjusted for the most variables, to maintain consistency across studies. Control variables for each of the studies can be found in Table 3. We converted aOR to log-odds ratios using the R program metafor based on extant literature (Colditz et al., Reference Colditz, Brewer, Berkey, Wilson, Burdick, Fineberg and Mosteller1994).

Other Coded Data. For each study, we collected data on setting (years that the study took place, country [US vs non-US], and US state of participants), age of participants at baseline (under 18, over 18, or mix), length of follow-up period, and any adjustments made in the analyses. We also attempted to code for the percentages of gender identities in the sample, but that variable was inconsistently reported in the primary studies included in this meta-analysis.

Risk of Bias. The risk of bias was assessed using an adapted version of the “Tool to Assess Risk of Bias in Cohort Studies” available from Cochrane (Supplement IV) (Sterne et al., Reference Sterne, Hernan, McAleenan, Reeves and Higgins2022). We utilized 7 out of the 8 items that applied to our systematic review, including “Can we be confident in the assessment of exposure?” We did not use the item that assessed co-interventions, since it was not appropriate for this systematic review (Supplement V). As suggested by the literature, multiple tests were used to triangulate evidence of publication bias (Vevea et al., Reference Vevea, Coburn and Sutton2019). The risk of bias was assessed by testing funnel plot asymmetry (Pustejovsky & Tipton, Reference Pustejovsky and Tipton2022) and conducting an Egger’s regression test adapted for dependent effect sizes (Duval & Tweedie, Reference Duval and Tweedie2000; Egger et al., Reference Egger, Davey Smith, Schneider and Minder1997).

Meta-analytic Approach. Rstudio version 1.2 was used to conduct all analyses (RStudio Team, 2018). We used a random effects model to synthesize the results given the likely variation across studies in population and context. The random effects models were estimated using REML. Many studies reported multiple effect sizes due to subgroup analyses (e.g., different levels of cannabis use, different ages of onset of use); thus, given the presence of dependent effect sizes, we used robust variance estimation for the analysis using the package clubSandwich (Pustejovsky & Tipton, Reference Pustejovsky and Tipton2022). We estimated effect size models using a hierarchical and correlated effects model assuming a correlation of 0.6 among dependent effect sizes. This allowed us to make more conservative assumptions about statistical significance. For the meta-analyses of early onset, heavy use, and country, we examined effect size moderators using multilevel meta-regression and RVE adjustment using metafor and clubSandwich. As suggested by Williams et al. (2021), these models are exploratory in nature (Williams et al, Reference Williams, Citkowicz, Miller, Lindsay and Walters2022).

Results

Search results

Overall, 1,599 titles were initially screened across PubMed (1372), MedLine (222), and ProQuest (5). From the ancestor search of the Lev-Ran articles, we screened an additional 969 articles. After duplicates were removed (n = 249) there were 2,319 articles that were title screened. We excluded 2,123 articles based on title screening, leaving 94 articles for abstract/full-text screen. The most common reasons for exclusion were review articles, included only special populations (i.e., Veterans (Gunn et al., Reference Gunn, Stevens, Micalizzi, Jackson, Borsari and Metrik2020), LGBT (London-Nadeau et al., Reference London-Nadeau, Rioux, Parent, Vitaro, Côté, Boivin and Castellanos-Ryan2021)), study design (i.e., cross-sectional (Hawke et al, Reference Hawke, Wilkins and Henderson2020)), or did not look at depression separately from other mental health outcomes (Spineli & Pandis, Reference Spineli and Pandis2020). In the end, 22 studies were included in the meta-analysis, which included the 14 original articles from the Lev-Ran meta-analysis. Summaries of the included studies can be found in Tables 1 and 2. Study selection is visualized in the PRISMA diagram (Figure 1).

Table 1.

Summary of studies included in the meta-analysis

Notes: *indicates article was included in the Lev-Ran, Reference Lev-Ran, Roerecke, Foll, George, McKenzie and Rehm2014 meta-analysis.

Table 2.

Measurements used in included studies

Notes: * Indicates article was included in the Lev-Ran, Reference Lev-Ran, Roerecke, Foll, George, McKenzie and Rehm2014 meta-analysis.

Figure 1.

PRISMA diagram.

Study characteristics

Eleven of the studies took place in the United States, three of which were national surveys. The others were based in Sweden (n = 2), Australia (n = 2); New Zealand (n = 2), The Netherlands (n = 1), United Kingdom (n = 1), Canada (n = 1), Switzerland (n = 1), and Norway (n = 1). The majority (n = 12) enrolled participants who were under 18 at the first time point. The median follow-up time was 7 years (interquartile range: 3–16 years).

Meta-analysis

Extracted effects can be found in Table 3.

Table 3.

Results from the systematic review

Notes: *included in the initial Lev-Ran et al. article.

a adjusted for variables in column 4, unless none specified.

b sub-analysis for early onset.

c sub-analysis for heavy/problematic use.

Mean Effect Size and Heterogeneity. We estimated the mean effect size using a correlated and hierarchical effects model (CHE) with robust variance estimation using clubSandwich. The estimated mean log-odds of depression was 0.25 (SE = 0.06, df = 18, p < .001). This corresponds to an OR of 1.29, with a 95% CI [1.13, 1.46]. This would suggest a higher risk of developing depression among people who used cannabis at baseline.

We utilized the model to create a 95% prediction interval of 0.75 and 2.20. This is the likely range of effect size values where we would expect a randomly selected new study for the meta-analysis to fall. This is a wide interval, suggesting a high degree of uncertainty around future predictions.

Meta-regression

To explore the heterogeneity across studies, we used the CHE model with robust variance estimation to fit a meta-regression using Country (US versus non-US), Cannabis Use (Non-heavy vs Heavy), and Onset (Late vs Early) as moderators. Table 4 presents the results of the meta-regression model with all three moderators and the associated Satterthwaite degrees of freedom. The meta-regression results test the difference between the levels of each moderator, controlling for the other moderators in the model. Only Cannabis Use was statistically significant at the p < 0.10 level, indicating that controlling for Country and Onset, heavy cannabis users had higher log odds of depression than non-heavy users. Table 4 also presents the adjusted means for each level of the three moderators, adjusted for the other moderators in the model. With the exception of US studies, the adjusted sub-group mean odds ratios were all significantly different from 1. The US studies had an average odds ratio that did not differ from 1.

Table 4.

Moderator results from mixed-effects meta-regression model

Note: *significant at p < 0.10, ** significant at p < .01; ^ indicates reference group;

The first column reports the conditional means. The conditional means are the predicted values (OR) from a multivariable meta-regression model that simultaneously controlled for all the listed moderators (e.g., the odds ratio for U.S. studies when all other moderators are fixed at their observed mean). The standard errors (SE) were adjusted for effect size dependencies using robust variance estimation. The p values assess whether the levels of a moderator are statistically significantly different from one another, controlling for all other moderators in the model.

Risk of Bias. Results from this risk of bias tool can be found in Table 5. Overall, there was a medium risk of bias across all studies with a mean of 2.17 (sd = 0.57) on a scale from 1 (low risk of bias) to 5 (high risk of bias). All studies were assigned a “1” for “Was the selection of exposed and non-exposed cohorts drawn from the same population?” as they all selected cannabis users and non-users from the same population. The next lowest score was for whether there was adequate control for prognostic variables related to the outcome of depression (mean = 2.14), but this one also had the highest variability (sd = 1.03). The category that had the highest risk of bias was in the measure of the exposure, with a mean of 2.82 and the lowest variability (sd = 0.45). All of the studies relied on self-report from the participants to classify cannabis use. The question that assessed the measurement of the outcome had a moderate level of bias (mean = 2.34, sd = 0.45). Of note, only one study looked at clinical records to make a determination for depression) (Haroz et al., Reference Haroz, Ritchey, Bass, Kohrt, Augustinavicius, Michalopoulos and Bolton2017).

Table 5.

Risk of Bias assessment, average scores ranging from 1 (low risk of bias) to 5 (high risk of bias)

Notes: *indicates article was included in the Lev-Ran, Reference Lev-Ran, Roerecke, Foll, George, McKenzie and Rehm2014 meta-analysis. Scores were based on the following scale: 1 = “Definitely yes (low risk of bias)”; 2 = “Probably yes”; 3 = “Probably no”; 4 = “Definitely no (high risk of bias).” Half-points were given if the study was deemed to be between two levels in the risk of bias scale. aWas selection of exposed and non-exposed cohorts drawn from the same population? bCan we be confident in the assessment of the exposure? cCan we be confident that the outcome of interest was not present at start of the study? If depression at the start of the study was controlled for, we assigned a score of 1.75. dDid the study match exposed and unexposed for all variables that are associated with the outcome of interest or did the statistical analysis adjust for these prognostic variables? eCan we be confident in the assessment of the presence or abssence of prognostic factors? fCan we be confident in the assessment of outcome? gWas the follow up of cohorts adequate? hAssessed using the modified “Tool to Assess Risk of Bias in Cohort Studies” see Supplement III.

A meta-regression using items on the Risk of Bias assessment tool as predictors was conducted. We recorded each study on the Cochrane Tool metric as either high or low risk of bias, where a score of above 2.5 was considered high. We focused on two of the questions: (3) Can we be confident that the outcome of interest was not present at the start of the study?; and (4) Can we be confident in the assessment of the presence or absence of prognostic factors? These were run in separate meta-regression models as a single predictor for developing depression. Neither item was related to effect size heterogeneity.

Selection Bias/Publication Bias. We also produced a funnel plot (Figure 2) and conducted Egger’s sandwich regression test to check for the risk of publication bias in the presence of dependent effect sizes. The Egger’s test produced a score of 0.07, which was not significant (p = .89). This, in addition to the non-symmetric funnel plot, does not provide evidence of bias; however, it is not possible to completely rule out any risk of either selection or publication bias.

Figure 2.

Funnel plot for publication bias.

Discussion

This systematic review and meta-analysis was designed to explore the longitudinal relationship between cannabis use and depression, building upon a prior meta-analysis by Lev-Ran et al. (2013). In updating this work, identified eight additional studies (Arseneault, Reference Arseneault2002; Blanco et al., Reference Blanco, Hasin, Wall, Flórez-Salamanca, Hoertel, Wang and Olfson2016; Copeland et al., Reference Copeland, Hill and Shanahan2022; Danielsson et al., Reference Danielsson, Lundin, Agardh, Allebeck and Forsell2016; Feingold et al., Reference Feingold, Weiser, Rehm and Lev-Ran2015; Gage et al., Reference Gage, Hickman, Heron, Munafò, Lewis, Macleod and Zammit2015; Hengartner et al., Reference Hengartner, Angst, Ajdacic-Gross and Rössler2020; Schaefer et al., Reference Schaefer, Hamdi, Malone, Vrieze, Wilson, McGue and Iacono2021) that we added to the original studies from the Lev-Ran et al. analysis (Bovasso, Reference Bovasso2001; Brook et al., Reference Brook, Brook, Zhang, Cohen and Whiteman2002; Brook et al., Reference Brook, Lee, Brown, Finch and Brook2011; Degenhardt et al., Reference Degenhardt, Coffey, Romaniuk, Swift, Carlin, Hall and Patton2013; Fergusson & Horwood, Reference Fergusson and Horwood1997; Georgiades & Boyle, Reference Georgiades and Boyle2007; Harder et al., Reference Harder, Stuart and Anthony2008; Manrique-Garcia et al., Reference Manrique-Garcia, Zammit, Dalman, Hemmingsson and Allebeck2012; Marmorstein & Iacono, Reference Marmorstein and Iacono2011; Paton et al., Reference Paton, Kessler and Kandel1977; Patton et al., Reference Patton, Coffey, Carlin, Degenhardt, Lynskey and Hall2002; Pedersen, Reference Pedersen2008; van Laar et al, Reference van Laar, van Dorsselaer, Monshouwer and de Graaf2007).

Our findings indicated an odds ratio of 1.29 (95% CI: 1.13, 1.46) compared to non-users, consistent with Lev-Ran et al.’s 1.17 (95% CI: 1.05–1.30) but with a high degree of heterogeneity (95% PI: 0.75, 2.20). Similar to Lev-Ran et al., we found that studies with heavy cannabis users reported a higher odds ratio than those on non-heavy users and that the effect estimate did not significantly differ between early- and late-onset cannabis users. This lack of statistically significant difference may reflect the high heterogeneity in exposure and outcome measurements across studies, making it difficult to infer robust conclusions.

There was no significant difference in odds of depression after cannabis use when US and non-US countries were compared directly and holding the heaviness of cannabis use and onset constant. However, our interpretation of this insignificant finding remains cautious as dichotomizing US vs non-US results may not be inappropriate for the outcome of depression. For instance, a systematic review of qualitative literature exploring the definition of depression globally found that the Diagnostic and Statistical Manual of Mental Disorders (DSM) commonly used in Western countries does not adequately reflect how depression is expressed by individuals in other countries (Haroz et al., Reference Haroz, Ritchey, Bass, Kohrt, Augustinavicius, Michalopoulos and Bolton2017; Kessler & Bromet, Reference Kessler and Bromet2013).

Limitations

A few key limitations should be noted. First, this review was not registered in advance with PROSPERO or a similar systematic review registry, which could have enhanced transparency and methodological rigor by establishing a priori protocols for study selection and data extraction. As such, some risk of selection bias exists, given the absence of a public pre-registration outlining the study design.

Second, we did not search all possible databases and instead relied heavily on PubMed as our primary data source. While this was in line with the previous methodology of Lev-Ran et al., it may have limited the comprehensiveness of our search and increased the risk of missing relevant studies published in other databases. Cannabis use and mental health studies, in particular, are multidisciplinary and may be published in journals indexed in databases like PsycINFO or Scopus, potentially omitting relevant literature.

Furthermore, we excluded studies focusing on specific subpopulations, such as LGBTQ+ communities, or individuals with specific psychiatric comorbidities. While this decision was made to maintain a more generalizable sample, it may have inadvertently reduced the applicability of our findings to these groups.

In summary, these limitations underscore the need for a cautious interpretation of our results, as they may not fully reflect the broader body of literature on cannabis use and depression. Future research should prioritize protocol registration, utilize a wider range of databases, and include diverse populations to provide a more comprehensive understanding of this complex relationship.

Suggestions for future research

Throughout the process, we identified several opportunities for future research to more clearly determine the relationship between cannabis use and depression.

The definition of who is a cannabis user is not consistent across studies. For example, some studies dichotomize people into cannabis “users” and “non-users.” However, even this is not consistent across studies. The article by van Laar et al. (Reference van Laar, van Dorsselaer, Monshouwer and de Graaf2007) defined a cannabis user as anyone who used it more than 5 times in their lifetime, whereas Manrique-Garcia et al. (Reference Manrique-Garcia, Zammit, Dalman, Hemmingsson and Allebeck2012) defined a cannabis user as anyone who used it at least once. Manrique-Garcia et al. had tried to categorize into more groups, but were unable to due to “the small number of cases.” Attempts to label cannabis users as “heavy,” “moderate,” and “light” were also not consistent. Furthermore, some studies looked at the past year to determine the heaviness of use, while others looked at lifetime use.

Accordingly, another limitation of the studies is that many of them were not able to consider changes in cannabis use over time. Measuring cannabis use at a single time point may not account for the waxing and waning of use across the lifecycle of an individual. While many cannabis users start in adolescence or young adulthood, there is evidence that some may discontinue the behavior as they “mature out” (Arora et al., Reference Arora, Qualls, Bobitt, Milavetz and Kaskie2021; Kosty et al., Reference Kosty, Seeley, Farmer, Stevens and Lewinsohn2017). It may be beneficial to explore whether quitting cannabis as an adult reduces the risk of depression to the levels of non-users or if youth cannabis exposure is enough to elevate the risk of depression despite subsequent non-use. This could have major implications for interventions that aim to delay the onset of cannabis use. Furthermore, due to the nature of the articles included in the meta-analysis, determining an average length of follow-up is not plausible. This limits our interpretation of the role of time in the relationship between cannabis use and depression.

Similar to cannabis use, depression was inconsistently measured across studies. While “major depressive disorder” and “major depressive episode” were commonly cited, other definitions such as “depressive disorder” and “depressive symptoms” were used. Additionally, there were several diagnostic tools and scales used to measure depression across the studies; the most predominant was the DSM, with different versions (III, III-R, IV, and V) being used across studies. It is not guaranteed that an individual who met the criteria for depression on one measure would meet it on another; therefore, the heterogeneity of the outcome measure is an area of concern that future research needs to consider in order to gain a better understanding of the relationship between cannabis and depression.

Finally, we were unable to analyze differences in the relationship between cannabis use and depression across genders, racial/ethnic groups, and age groups due to insufficient reporting in the studies reviewed. This limitation highlights an important gap, as understanding how these relationships vary by gender, race/ethnicity, and age is crucial to comprehensively addressing the potential disproportionate burden of cannabis on specific demographics. Future research should prioritize collecting and reporting these demographic details to allow for more nuanced analysis and to support targeted interventions.

Conclusions

There is evidence that cannabis use is associated with the onset of depression over time. While we updated a previous meta-analysis from 2012 and found similar results, gaps in our understanding of how cannabis affects mental health remain. The current state of the literature has not kept up with the changing landscape of cannabis use in the US and the world. Although this is partly due to the nature of longitudinal research, we recommend that researchers focus on the following: 1) consider the mode of cannabis use (e.g., edibles, concentrates); 2) find ways to better define cannabis use, specifically definitions of heavy versus light use; and 3) conduct analyses to investigate how cannabis use changes over time relates to the risk of depression.

Lastly, although our findings indicate an association, it is likely that confounding factors, which were not fully captured in our analysis, contribute to this relationship. Therefore, while our study explores a temporal, longitudinal connection, it cannot definitively establish causation.

Supplementary material

The supplementary material for this article can be found at http://doi.org/10.1017/S0033291724003143.

Funding statement

This research received no specific grant from any funding agency, commercial or not-for-profit sectors.

Competing interest

There are no conflicts of interest to declare.

References

Arora, K., Qualls, S. H., Bobitt, J., Milavetz, G., & Kaskie, B. (2021). Older cannabis users are not all alike: Lifespan cannabis use patterns. Journal of Applied Gerontology, 40(1), 87–94. https://doi.org/10.1177/0733464819894922CrossRefGoogle Scholar
Arseneault, L. (2002). Cannabis use in adolescence and risk for adult psychosis: Longitudinal prospective study. British Medical Journal 325(7374), 1212–1213. https://doi.org/10.1136/bmj.325.7374.1212CrossRefGoogle ScholarPubMed
Bahorik, A. L., Sterling, S. A., Campbell, C. I., Weisner, C., Ramo, D., & Satre, D. D. (2018). Medical and non-medical marijuana use in depression: Longitudinal associations with suicidal ideation, everyday functioning, and psychiatry service utilization. Journal of Affective Disorders, 241, 8–14. https://doi.org/10.1016/j.jad.2018.05.065CrossRefGoogle ScholarPubMed
Bambico, F. R., Nguyen, N.-T., Katz, N., & Gobbi, G. (2010). Chronic exposure to cannabinoids during adolescence but not during adulthood impairs emotional behaviour and monoaminergic neurotransmission. Neurobiology of Disease, 37(3), 641–655. https://doi.org/10.1016/j.nbd.2009.11.020CrossRefGoogle Scholar
Blanco, C., Hasin, D. S., Wall, M. M., Flórez-Salamanca, L., Hoertel, N., Wang, S., … Olfson, M. (2016). Cannabis use and risk of psychiatric disorders: Prospective evidence from a US National longitudinal study. JAMA Psychiatry, 73(4), 388–395. https://doi.org/10.1001/jamapsychiatry.2015.3229CrossRefGoogle ScholarPubMed
Bovasso, G. B. (2001). Cannabis abuse as a risk factor for depressive symptoms. The American Journal of Psychiatry, 158(12), 2033–2037. https://doi.org/10.1176/appi.ajp.158.12.2033CrossRefGoogle ScholarPubMed
Brook, D. W., Brook, J. S., Zhang, C., Cohen, P., & Whiteman, M. (2002). Drug use and the risk of major depressive disorder, alcohol dependence, and substance use disorders. Archives of General Psychiatry, 59(11), 1039–1044. https://doi.org/10.1001/archpsyc.59.11.1039CrossRefGoogle ScholarPubMed
Brook, J. S., Lee, J. Y., Brown, E. N., Finch, S. J., & Brook, D. W. (2011). Developmental trajectories of marijuana use from adolescence to adulthood: Personality and social role outcomes. Psychological Reports, 108(2), 339–357. https://doi.org/10.2466/10.18.PR0.108.2.339-357CrossRefGoogle ScholarPubMed
Centers for Disease Control and Prevention. (2021, June 8). Data and Statistics | Marijuana | CDC. Retrieved October 6, 2021, from Marijuana and Public Health website. https://www.cdc.gov/marijuana/data-statistics.htmGoogle Scholar
Colditz, G. A., Brewer, T. F., Berkey, C. S., Wilson, M. E., Burdick, E., Fineberg, H. V., & Mosteller, F. (1994). Efficacy of BCG vaccine in the prevention of tuberculosis: Meta-analysis of the published literature. JAMA, 271(9), 698–702. https://doi.org/10.1001/jama.1994.03510330076038CrossRefGoogle ScholarPubMed
Copeland, W. E., Hill, S. N., & Shanahan, L. (2022). Adult psychiatric, substance, and functional outcomes of different definitions of early cannabis use. Journal of the American Academy of Child & Adolescent Psychiatry, 61(4), 533–543. https://doi.org/10.1016/j.jaac.2021.07.824CrossRefGoogle ScholarPubMed
Danielsson, A.-K., Lundin, A., Agardh, E., Allebeck, P., & Forsell, Y. (2016). Cannabis use, depression and anxiety: A 3-year prospective population-based study. Journal of Affective Disorders, 193, 103–108. https://doi.org/10.1016/j.jad.2015.12.045CrossRefGoogle ScholarPubMed
Davis, A. K., Lin, L. A., Ilgen, M. A., & Bohnert, K. M. (2018). Recent cannabis use among Veterans in the United States: Results from a national sample. Addictive Behaviors, 76, 223–228. https://doi.org/10.1016/j.addbeh.2017.08.010CrossRefGoogle ScholarPubMed
Degenhardt, L., Coffey, C., Romaniuk, H., Swift, W., Carlin, J. B., Hall, W. D., & Patton, G. C. (2013). The persistence of the association between adolescent cannabis use and common mental disorders into young adulthood. Addiction (Abingdon, England), 108(1), 124–133. https://doi.org/10.1111/j.1360-0443.2012.04015.xCrossRefGoogle ScholarPubMed
Degenhardt, L., Hall, W., & Lynskey, M. (2003). Exploring the association between cannabis use and depression. Addiction (Abingdon, England), 98(11), 1493–1504. https://doi.org/10.1046/j.1360-0443.2003.00437.xCrossRefGoogle ScholarPubMed
Duval, S., & Tweedie, R. (2000). Trim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics, 56(2), 455–463. https://doi.org/10.1111/j.0006-341x.2000.00455.xCrossRefGoogle ScholarPubMed
Dyar, C., Bradley, H., Morgan, E., Sullivan, P. S., & Mustanski, B. (2020). Reduction and cessation of alcohol, cannabis, and stimulant use: Prospective associations with changes in depressive symptoms across two cohort studies of sexual and gender minorities. Journal of Studies on Alcohol and Drugs, 81(6), 790–797.CrossRefGoogle ScholarPubMed
Dyar, C., Feinstein, B. A., Newcomb, M. E., & Whitton, S. W. (2021). Cannabis use companions’ gender and sexual orientation: Associations with problematic cannabis use in a sample of sexual minorities assigned female at birth. Addictive Behaviors, 118, 106878. https://doi.org/10.1016/j.addbeh.2021.106878CrossRefGoogle Scholar
Eddy, M. (2010). Medical marijuana: Review and Analysis of federal and state policies (No. 7–5700). https://medicalmarijuana.procon.org/wp-content/uploads/sites/37/medicalmarjiuanastatepolicies040210.pdfGoogle Scholar
Egger, M., Davey Smith, G., Schneider, M., & Minder, C. (1997). Bias in meta-analysis detected by a simple, graphical test. BMJ (Clinical Research Ed.), 315(7109), 629–634. https://doi.org/10.1136/bmj.315.7109.629CrossRefGoogle ScholarPubMed
ElSohly, M. A., Chandra, S., Radwan, M., Majumdar, C. G., & Church, J. C. (2021). A comprehensive review of cannabis potency in the United States in the last decade. Biological Psychiatry. Cognitive Neuroscience and Neuroimaging, 6(6), 603–606. https://doi.org/10.1016/j.bpsc.2020.12.016CrossRefGoogle ScholarPubMed
ElSohly, M. A., Mehmedic, Z., Foster, S., Gon, C., Chandra, S., & Church, J. C. (2016). Changes in cannabis potency over the last 2 decades (1995–2014): Analysis of current data in the United States. Biological Psychiatry, 79(7), 613–619. https://doi.org/10.1016/j.biopsych.2016.01.004CrossRefGoogle ScholarPubMed
Feingold, D., Weiser, M., Rehm, J., & Lev-Ran, S. (2015). The association between cannabis use and mood disorders: A longitudinal study. Journal of Affective Disorders, 172, 211–218. https://doi.org/10.1016/j.jad.2014.10.006CrossRefGoogle ScholarPubMed
Fergusson, D. M., & Horwood, L. J. (1997). Early onset cannabis use and psychosocial adjustment in young adults. Addiction (Abingdon, England), 92(3), 279–296.CrossRefGoogle ScholarPubMed
Ferlatte, O., Salway, T., Rice, S. M., Oliffe, J. L., Knight, R., & Ogrodniczuk, J. S. (2020). Inequities in depression within a population of sexual and gender minorities. Journal of Mental Health (Abingdon, England), 29(5), 573–580. https://doi.org/10.1080/09638237.2019.1581345CrossRefGoogle ScholarPubMed
Gage, S. H., Hickman, M., Heron, J., Munafò, M. R., Lewis, G., Macleod, J., & Zammit, S. (2015). Associations of cannabis and cigarette use with depression and anxiety at age 18: Findings from the Avon longitudinal study of parents and children. PLOS ONE, 10(4), e0122896. https://doi.org/10.1371/journal.pone.0122896CrossRefGoogle ScholarPubMed
Georgiades, K., & Boyle, M. H. (2007). Adolescent tobacco and cannabis use: Young adult outcomes from the Ontario Child Health Study. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 48(7), 724–731. https://doi.org/10.1111/j.1469-7610.2007.01740.xCrossRefGoogle ScholarPubMed
Gobbi, G., Atkin, T., Zytynski, T., Wang, S., Askari, S., Boruff, J., … Mayo, N. (2019). Association of cannabis use in adolescence and risk of depression, anxiety, and suicidality in young adulthood: A systematic review and meta-analysis. JAMA Psychiatry, 76(4), 426–434. https://doi.org/10.1001/jamapsychiatry.2018.4500CrossRefGoogle Scholar
Grucza, R. A., Vuolo, M., Krauss, M. J., Plunk, A. D., Agrawal, A., Chaloupka, F. J., & Bierut, L. J. (2018). Cannabis decriminalization: A study of recent policy change in five U.S. states. International Journal of Drug Policy, 59, 67–75. https://doi.org/10.1016/j.drugpo.2018.06.016CrossRefGoogle ScholarPubMed
Gunn, R. L., Stevens, A. K., Micalizzi, L., Jackson, K. M., Borsari, B., & Metrik, J. (2020). Longitudinal associations between negative urgency, symptoms of depression, cannabis and alcohol use in veterans. Experimental and Clinical Psychopharmacology, 28(4), 426–437. https://doi.org/10.1037/pha0000357CrossRefGoogle ScholarPubMed
Harder, V. S., Stuart, E. A., & Anthony, J. C. (2008). Adolescent cannabis problems and young adult depression: Male-female stratified propensity score analyses. American Journal of Epidemiology, 168(6), 592–601. https://doi.org/10.1093/aje/kwn184CrossRefGoogle Scholar
Haroz, E. E., Ritchey, M., Bass, J. K., Kohrt, B. A., Augustinavicius, J., Michalopoulos, L., … Bolton, P. (2017). How is depression experienced around the world? A systematic review of qualitative literature. Social Science & Medicine (1982), 183, 151–162. https://doi.org/10.1016/j.socscimed.2016.12.030CrossRefGoogle Scholar
Hawke, L. D., Wilkins, L., & Henderson, J. (2020). Early cannabis initiation: Substance use and mental health profiles of service-seeking youth. Journal of Adolescence, 83, 112–121. https://doi.org/10.1016/j.adolescence.2020.06.004CrossRefGoogle ScholarPubMed
Hengartner, M. P., Angst, J., Ajdacic-Gross, V., & Rössler, W. (2020). Cannabis use during adolescence and the occurrence of depression, suicidality and anxiety disorder across adulthood: Findings from a longitudinal cohort study over 30 years. Journal of Affective Disorders, 272, 98–103. https://doi.org/10.1016/j.jad.2020.03.126CrossRefGoogle ScholarPubMed
Hodgson, K., Almasy, L., Knowles, E. E. M., Kent, J. W., Curran, J. E., Dyer, T. D., … Glahn, D. C. (2017). The genetic basis of the comorbidity between cannabis use and major depression. Addiction (Abingdon, England), 112(1), 113–123. https://doi.org/10.1111/add.13558CrossRefGoogle ScholarPubMed
Kedzior, K. K., & Laeber, L. T. (2014). A positive association between anxiety disorders and cannabis use or cannabis use disorders in the general population—A meta-analysis of 31 studies. BMC Psychiatry, 14, 136.CrossRefGoogle ScholarPubMed
Kessler, R. C., & Bromet, E. J. (2013). The epidemiology of depression across cultures. Annual Review of Public Health, 34, 119–138. https://doi.org/10.1146/annurev-publhealth-031912-114409CrossRefGoogle ScholarPubMed
Kosty, D. B., Seeley, J. R., Farmer, R. F., Stevens, J. J., & Lewinsohn, P. M. (2017). Trajectories of cannabis use disorder: Risk factors, clinical characteristics and outcomes. Addiction, 112(2), 279–287. https://doi.org/10.1111/add.13557CrossRefGoogle ScholarPubMed
Langlois, C., Potvin, S., Khullar, A., & Tourjman, S. V. (2021). Down and high: Reflections regarding depression and cannabis. Frontiers in Psychiatry, 12, 625158. https://doi.org/10.3389/fpsyt.2021.625158CrossRefGoogle ScholarPubMed
Lev-Ran, S., Roerecke, M., Foll, B. L., George, T. P., McKenzie, K., & Rehm, J. (2014). The association between cannabis use and depression: A systematic review and meta-analysis of longitudinal studies. Psychological Medicine, 44(4), 797–810. https://doi.org/10.1017/S0033291713001438CrossRefGoogle ScholarPubMed
Liu, Y., Collins, C., Wang, K., Xie, X., & Bie, R. (2019). The prevalence and trend of depression among veterans in the United States. Journal of Affective Disorders, 245, 724–727. https://doi.org/10.1016/j.jad.2018.11.031CrossRefGoogle ScholarPubMed
London-Nadeau, K., Rioux, C., Parent, S., Vitaro, F., Côté, S. M., Boivin, M., … Castellanos-Ryan, N. (2021). Longitudinal associations of cannabis, depression, and anxiety in heterosexual and LGB adolescents. Journal of Abnormal Psychology, 130(4), 333–345. https://doi.org/10.1037/abn0000542CrossRefGoogle ScholarPubMed
Mahabir, V. K., Merchant, J. J., Smith, C., & Garibaldi, A. (2020). Medical cannabis use in the United States: A retrospective database study. Journal of Cannabis Research, 2(1), 32. https://doi.org/10.1186/s42238-020-00038-wCrossRefGoogle ScholarPubMed
Mammen, G., Rueda, S., Roerecke, M., Bonato, S., Lev-Ran, S., & Rehm, J. (2018). Association of cannabis with long-term clinical symptoms in anxiety and mood disorders: A systematic review of prospective studies. The Journal of Clinical Psychiatry, 79(4), 17r11839. https://doi.org/10.4088/JCP.17r11839CrossRefGoogle ScholarPubMed
Manrique-Garcia, E., Zammit, S., Dalman, C., Hemmingsson, T., & Allebeck, P. (2012). Cannabis use and depression: A longitudinal study of a national cohort of Swedish conscripts. BMC Psychiatry, 12, 112. https://doi.org/10.1186/1471-244X-12-112CrossRefGoogle ScholarPubMed
Marmorstein, N. R., & Iacono, W. G. (2011). Explaining associations between cannabis use disorders in adolescence and later major depression: A test of the psychosocial failure model. Addictive Behaviors, 36(7), 773–776. https://doi.org/10.1016/j.addbeh.2011.02.006CrossRefGoogle ScholarPubMed
Meier, M. H. (2021). Cannabis use and psychosocial functioning: Evidence from prospective longitudinal studies. Current Opinion in Psychology, 38, 19–24. https://doi.org/10.1016/j.copsyc.2020.07.001CrossRefGoogle ScholarPubMed
Niedzwiedz, C. L., Green, M. J., Benzeval, M., Campbell, D., Craig, P., Demou, E., … Katikireddi, S. V. (2021). Mental health and health behaviours before and during the initial phase of the COVID-19 lockdown: Longitudinal analyses of the UK Household Longitudinal Study. J Epidemiol Community Health, 75(3), 224–231. https://doi.org/10.1136/jech-2020-215060CrossRefGoogle ScholarPubMed
Ouzzani, M., Hammady, H., Fedorowicz, Z., & Elmagarmid, A. (2016). Rayyan—A web and mobile app for systematic reviews. Systematic Reviews, 5(1), 210. https://doi.org/10.1186/s13643-016-0384-4CrossRefGoogle ScholarPubMed
Pacheco-Colón, I., Ramirez, A. R., & Gonzalez, R. (2019). Effects of adolescent cannabis use on motivation and depression: A systematic review. Current Addiction Reports, 6(4), 532–546. https://doi.org/10.1007/s40429-019-00274-yCrossRefGoogle ScholarPubMed
Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., … Moher, D. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ, 372, n71. https://doi.org/10.1136/bmj.n71CrossRefGoogle ScholarPubMed
Patel, R. K., & Rose, G. M. (2021). Persistent depressive disorder. In StatPearls. Treasure Island (FL): StatPearls Publishing. http://www.ncbi.nlm.nih.gov/books/NBK541052/Google Scholar
Paton, S., Kessler, R., & Kandel, D. (1977). Depressive mood and adolescent illicit drug use: A longitudinal analysis. The Journal of Genetic Psychology, 131(2), 267–289. https://doi.org/10.1080/00221325.1977.10533299CrossRefGoogle ScholarPubMed
Patton, G. C., Coffey, C., Carlin, J. B., Degenhardt, L., Lynskey, M., & Hall, W. (2002). Cannabis use and mental health in young people: Cohort study. BMJ, 325(7374), 1195–1198. https://doi.org/10.1136/bmj.325.7374.1195CrossRefGoogle Scholar
Pedersen, W. (2008). Does cannabis use lead to depression and suicidal behaviours? A population-based longitudinal study. Acta Psychiatrica Scandinavica, 118(5), 395–403. https://doi.org/10.1111/j.1600-0447.2008.01259.xCrossRefGoogle ScholarPubMed
Prince, M. A., & Conner, B. T. (2019). Examining links between cannabis potency and mental and physical health outcomes. Behaviour Research and Therapy, 115, 111–120. https://doi.org/10.1016/j.brat.2018.11.008CrossRefGoogle ScholarPubMed
Pustejovsky, J. E., & Tipton, E. (2022). Meta-analysis with robust variance estimation: Expanding the range of working models. Prevention Science: The Official Journal of the Society for Prevention Research, 23(3), 425–438. https://doi.org/10.1007/s11121-021-01246-3CrossRefGoogle ScholarPubMed
RStudio Team. (2018). RStudio: Integrated development for R. Boston, MA: RStudio, Inc. http://www.rstudio.comGoogle Scholar
Schaefer, J. D., Hamdi, N. R., Malone, S. M., Vrieze, S., Wilson, S., McGue, M., & Iacono, W. G. (2021). Associations between adolescent cannabis use and young-adult functioning in three longitudinal twin studies. Proceedings of the National Academy of Sciences of the United States of America, 118(14), e2013180118. https://doi.org/10.1073/pnas.2013180118CrossRefGoogle ScholarPubMed
Spineli, L. M., & Pandis, N. (2020). Prediction interval in random-effects meta-analysis. American Journal of Orthodontics and Dentofacial Orthopedics, 157(4), 586–588. https://doi.org/10.1016/j.ajodo.2019.12.011CrossRefGoogle ScholarPubMed
Sterne, J. A., Hernan, M. A., McAleenan, A., Reeves, B. C., & Higgins, J. P. T. (2022). Chapter 25: Assessing risk of bias in a non-randomized study. In Cochrane handbook for systematic reviews of interventions. https://training.cochrane.org/handbook/current/chapter-25Google Scholar
Substance Abuse and Mental Health Services Administration. (2020). Key substance use and mental health indicators in the United States: Results from the 2020 national survey on drug use and health (HHS Publication No. PEP21–07–01-003, NSDUH Series H-56). 156.Google Scholar
Van Green, T. (2022). Americans overwhelmingly say marijuana should be legal for recreational or medical use. Retrieved November 22, 2022, from Pew Research Center website: https://www.pewresearch.org/fact-tank/2021/04/16/americans-overwhelmingly-say-marijuana-should-be-legal-for-recreational-or-medical-use/Google Scholar
van Laar, M., van Dorsselaer, S., Monshouwer, K., & de Graaf, R. (2007). Does cannabis use predict the first incidence of mood and anxiety disorders in the adult population? Addiction (Abingdon, England), 102(8), 1251–1260. https://doi.org/10.1111/j.1360-0443.2007.01875.xCrossRefGoogle ScholarPubMed
Vevea, J., Coburn, K., & Sutton, A. (2019). Publication bias. In The handbook of research synthesis and meta-analysis (3rd ed., pp. 383–432). New York: Russell Sage Foundation.CrossRefGoogle Scholar
Williams, R., Citkowicz, M., Miller, D. I., Lindsay, J., & Walters, K. (2022). Heterogeneity in mathematics intervention effects: Evidence from a meta-analysis of 191 randomized experiments. Journal of Research on Educational Effectiveness, 15(3), 584–634. https://doi.org/10.1080/19345747.2021.2009072CrossRefGoogle Scholar
World Drug Report 2019: 35 million people worldwide suffer from drug use disorders while only 1 in 7 people receive treatment. (2019). Retrieved October 12, 2021, from United Nations: Office on Drugs and Crime website: www.unodc.org/unodc/en/frontpage/2019/June/world-drug-report-2019_-35-million-people-worldwide-suffer-from-drug-use-disorders-while-only-1-in-7-people-receive-treatment.htmlGoogle Scholar
World Health Organization. (n.d.). Cannabis. Retrieved September 22, 2021, from https://www.who.int/teams/mental-health-and-substance-use/alcohol-drugs-and-addictive-behaviours/drugs-psychoactive/cannabisGoogle Scholar
Zheng, J., Morstead, T., Sin, N., Klaiber, P., Umberson, D., Kamble, S., & DeLongis, A. (2021). Psychological distress in North America during COVID-19: The role of pandemic-related stressors. Social Science & Medicine (1982), 270, 113687. https://doi.org/10.1016/j.socscimed.2021.113687CrossRefGoogle ScholarPubMed
Figure 0

Table 1.Summary of studies included in the meta-analysis

Figure 1

Table 2.Measurements used in included studies

Figure 2

Figure 1.PRISMA diagram.

Figure 3

Table 3.Results from the systematic review

Figure 4

Table 4.Moderator results from mixed-effects meta-regression model

Figure 5

Table 5.Risk of Bias assessment, average scores ranging from 1 (low risk of bias) to 5 (high risk of bias)

Figure 6

Figure 2.Funnel plot for publication bias.

Supplementary material: File

Churchill et al. supplementary material

Churchill et al. supplementary material
Download Churchill et al. supplementary material(File)
File 65.5 KB