11

Psychosocial

Adolescence and emerging adulthood are the periods where most youths begin to experiment with substances of abuse, including cannabis (Johnston et al., 2015). Exploration for many substances of abuse have maintained historical consistency for the past few decades, with approximately 24.9 percent of youths having used cannabis at least one time by eighth grade to 51.4 percent having tried cannabis by the time they graduate from high school (Johnston et al., 2015). Yet, recent changes in recreational cannabis use laws have been linked to adolescents’ changing perception around accessibility and availability of cannabis and decreased risk of harm from cannabis use, two factors that have been historically connected with rising rates of substance use (Feldstein Ewing et al., 2017;

Schmidt et al., 2016). The result is that we are at the forefront of a changing cannabis landscape for adolescents and young adults.

This is relevant because it is during this precise period of adolescence and young adulthood that the neural substrates that underlie the development of cognition are most active. Indeed, adolescence marks one of the most impressive stretches of neural and behavioral change (Giedd, 2015), with substantial and protracted development in terms of both brain structure and function throughout the teenage years and into the late 20s and early 30s (e.g., Conrod and Nikolaou, 2016). As a result, cannabis and other substance use during this period may incur relatively greater interference in neural, social, and academic functioning as compared to later developmental periods (e.g., adulthood) (Brumback et al., 2016; Jacobus et al., 2015).

However, with the paucity of data on the impact of changes of cannabis policy, coupled with existing limitations in the field of addiction neurodevelopment (e.g., predominance of cross-sectional studies) (Feldstein Ewing et al., 2014), we are still very much at the forefront of beginning to understand how cannabis impacts adolescent through adult cognitive health and broader psychosocial functioning.

COGNITION

Despite what appears, on first glance, to be a very broad existing literature, a surprisingly small number of empirical studies have examined how cannabis impacts the psychosocial domains targeted here. The questions addressed in this section revolve around how cannabis affects three aspects of cognition—memory, learning, and attention—areas that have continued to be prevalent across the self-report, neuropsychological, and magnetic resonance imaging (MRI)/functional magnetic resonance imaging (fMRI) literature since the mid-1970s. Furthermore, these are aspects of cognition that are often explored in other studies. In other words, evaluation of these aspects of cognition increases the potential to compare these findings to other studies, including the 10-year prospective examination of 10,000 youths across 21 sites (the ABCD study; Adolescent Brain Cognitive Development Study, 2016). In terms of the relevance of these aspects of cognition, the domains of learning, memory, and attention are central, as they undergird an individual’s success—or failure—across such areas as academic, employment, and social/relationship functioning. This subsequently renders these three domains of cognition strong proxies for examining interference in functioning, one of the key metrics of cannabis use disorder symptomology according to Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V).

These domains are defined broadly in order to be as inclusive as pos-

sible of how they were measured within the included systematic analyses and component primary manuscripts, and to allow maximal potential for generalization to the broader literature. Thus, within this review, “memory” is defined as the wide array of function that involves the abilities to remember, temporarily store, more extensively store, process, manipulate, recall, and reproduce data (e.g., verbal, auditory, written). In this review, “learning” is defined as the wide array of function that involves the ability to observe, comprehend, absorb, and appropriate new information into an individual’s cognitive repertoire (e.g., verbal, auditory, visual). Finally, in this review, “attention” is defined as an individual’s ability to stay focused on the task at hand without being distracted but also to be cognitively flexible enough to transfer to a different task or set of information when the time requires (e.g., including brain regions relevant to visual, auditory, and verbal processing as well as executive control).

To investigate how cannabis affects these three domains of human cognition (memory, learning, attention), a search was conducted to identify systematic reviews of the existing published literature since the publication of Marijuana and Medicine: Assessing the Science Base, the last Institute of Medicine (IOM) report on marijuana (1999). There were a total of 94 systematic reviews identified that responded to the topic of cannabis and cognition during the period of 2000–2016. Of these, 5 systematic reviews were considered of good quality (Batalla et al., 2013; Broyd et al., 2016; Grant et al., 2003; Martin-Santos et al., 2010; Schreiner and Dunn, 2012). No primary manuscripts were utilized in this section because all study questions were addressed by the systematic reviews.

In contrast to other sections of this report, given the diversity of the metrics and constructs in learning, memory, and attention, and the different coverage of these domains within the 5 different systematic reviews, we present summaries from each of the systematic reviews in these domains rather than only presenting one representative systematic review for the topic area of cognition. Furthermore, reflective of the field of cognition at this time, the presented systematic reviews reflect data from the fields of neuropsychology, computer-administered cognitive tests, as well as brain structure and function (e.g., MRI/fMRI). The latter represent some of our most contemporary, sensitive, and specific metrics of cognitive function at this time.

It should be noted that Chapter 12 (Mental Health) highlights the multidirectional and complex relationship between cannabis use and cannabis use disorder and cognitive performance among individuals with psychotic disorders. For further information on this topic, please refer to Chapter 12.

The collection of systematic reviews used in this chapter represents a large body of work. The Broyd et al. (2016) systematic review is the most

recent, evaluating 3,021 total manuscripts, yielding a final number of 105 manuscripts in their review. Within their systematic review, they evaluated cannabis’s interference with cognition across a number of assessment methodologies. Furthermore, they evaluated the impact of these cognitive domains across developmental periods, including adolescence, emerging adulthood, and adulthood (for additional information about developmental implications among adolescents, see Box 11-1). Batalla and colleagues began with 142 studies, which they narrowed to 43 manuscripts. As with the Broyd et al. (2016) team, Batalla et al. (2013) included studies

across the age span, including adolescents and adults. One of the older systematic reviews, Grant et al. (2003), commenced their review with 1,830 manuscripts, which were reduced to a group of 117 papers in their final evaluation. Martin-Santos et al. (2010) began their examination with 66 manuscripts, which resulted in a final set of 41 studies of cannabis’s effect on cognition. Schreiner and Dunn (2012) started with more than 800 studies, which they narrowed to a final set of 13 studies.

In these systematic reviews, “acute” generally reflects cognitive domains assessed within a short window (often within several hours) immediately after cannabis use. The individual may or may not still be intoxicated during this examination. In contrast, “sustained” generally reflects cognitive domains assessed after a period of abstinence from cannabis. Within the reviewed studies, that ranges from several hours to months after discontinuing cannabis use.

Is There an Association Between Cannabis Use and Learning?

Systematic Reviews

Of our final set of five systematic reviews, three addressed cannabis use on the cognitive domain of learning (Broyd et al., 2016; Grant et al., 2003; Schreiner and Dunn, 2012).

In terms of acute impact of cannabis use on learning, primarily relying on word list learning, data from 11 manuscripts within the Broyd et al. (2016) systematic review contributed to “strong” support of acute cannabis use on interference in learning. However, in terms of sustained effects, Broyd et al. (2016) only showed “mixed” support. Grant et al. (2003) assessed sustained impact of cannabis use on learning via neuropsychological tests (e.g., California Verbal Learning Test–Learning Trials; Rey Auditory Verbal Learning Test–Learning Trial). Across nine component studies, Grant et al. (2003) found a small negative effect size (ES) of −0.21 (99% confidence interval [CI] = −0.39 to −0.022) for the sustained impact of cannabis on learning. Schreiner and Dunn (2012) also examined sustained impact on learning, with component studies also relying on neuropsychological metrics (e.g., California Verbal Learning Test–Learning Trials; Rey Auditory Verbal Learning Test−Learning Trials; VIG−Visual Learning). Using the criteria of cannabis abstinence for at least 1 month (measured as ≥25 days) within their 13 examined studies, they found a very small ES of −0.16 (95% CI = −0.33–0.02).

One example study of the component studies within this section includes a study by Hanson et al. (2010). In this study, 19 adolescent marijuana users (mean age = 18 years) with limited other alcohol and/ or other substance use were compared with 21 demographically similar

non-using controls (mean age = 17.4 years). Participants completed neuropsychological batteries assessing learning and other cognitive domains at several points post-cessation (e.g., 3 days; 2 weeks; 3 weeks). Abstinence was verified via decreasing tetrahydrocannabinol metabolite values assessed via serial urine drug screens. Marijuana users showed initial poorer performance on learning as compared with non-using controls in acute assessments (at 3 days; p <0.01). However, they showed significant improvements with cessation, with no differences observed on learning between the cannabis-using and non-cannabis-using groups at either of the sustained time points (e.g., 2 weeks; 3 weeks).

Primary Literature

In this review of cannabis, the primary literature was searched when the systematic review content did not fully cover or address study questions. Given the breadth and scope of the existing systematic reviews in this domain, additional primary literature was not searched for the domain of learning.

Is There an Association Between Cannabis Use and Memory?

Systematic Reviews

Of the final set of five systematic reviews, three addressed cannabis use on the cognitive domain of memory (Batalla et al., 2013; Broyd et al., 2016; Martin-Santos et al., 2010).

In terms of acute impact of cannabis use on memory, the Broyd et al. (2016) systematic review was the only one to address this question. In this review, 22 studies assessed memory, including working memory and other memory function using various neuropsychological tests such as the Sternberg task, Trails B, n-back, and Wechsler tests, including spatial working memory, digit span, and digit recall. These studies showed moderate to strong evidence for acute interference of cannabis on memory. In terms of a long-term sustained relationship between cannabis use and learning following abstinence, the 11 studies examined by Broyd et al. (2016) showed mixed to no evidence for interference in memory functioning after cessation from cannabis use. Similarly, Batalla et al. (2013) examined memory using seven MRI/fMRI studies. The range in mean days of abstinence in these studies extended from 7 days to 201 days post-cannabis cessation. Batalla et al. (2013) found that although there was no difference in task performance between cannabis users and cannabis nonusers, cannabis users engaged slightly different parts of their brains as compared to nonusers to accomplish the task, often described in

the neuroimaging literature as the utilization of “compensatory” efforts. Similar to Batalla et al. (2013), Martin-Santos et al. (2010) examined five empirical MRI/fMRI studies. Individuals in these studies had abstained from using cannabis for an average of 24 hours to 26 days. As with Batalla et al. (2013), cannabis users showed equivalent performance across the neuroimaging tasks to the nonusers, but they could have engaged in compensatory efforts to achieve these outcomes.

One example study in the memory systematic analyses includes a recent study by Roten and colleagues (2015). This is a pharmacotherapy trial of 78 youth ages 15 to 21 years seeking treatment for cannabis dependence. These youths were evaluated to ensure their abstinence from cannabis use via urine cannabinoid testing. They received a computer-administered battery of tests, including verbal memory, visual memory, and composite memory. Youths who were recently abstinent and continuously abstinent for 4 weeks showed significantly better memory performance as compared to youths who were still using cannabis at the 4 week measurement (difference [d] = 7.2 ± 2.1, p <0.001 and d = 7.5 ± 2.4, p = 0.002, respectively).

Primary Literature

In this review of cannabis, the primary literature was searched when the systematic review content did not fully cover or address study questions. Given the breadth and scope of the existing systematic reviews in this domain, additional primary literature was not searched for the domain of memory.

Is There an Association Between Cannabis Use and Attention?

Systematic Reviews

Of our final set of five systematic reviews, four addressed cannabis use on the cognitive domain of attention (Batalla et al., 2013; Broyd et al., 2016; Grant et al., 2003; Schreiner and Dunn, 2012).

To determine the acute impact of cannabis use on attention, Broyd et al. (2016) reviewed 17 studies that assessed attention using several approaches, including using neuropsychological metrics of continuous task performance, divided attention tasks, reaction time, and attention control tasks. The synthesized findings from studies showed strong evidence for acute interference of cannabis on attention, as reported by the authors.

In terms of the long-term sustained relationship between cannabis use and attention following abstinence, 10 studies examined by Broyd et

al. (2016) showed mixed evidence for impairment in attention functioning after cessation from cannabis use. Likewise, using a series of MRI and fMRI measures (e.g., attention network task, functional connectivity via Multi-Source Interference Task) with three studies, Batalla et al. (2013) showed limited evidence of differences in task performance, but as with the other domains, they found evidence that cannabis users may be engaging a different neural network to achieve similar outcomes during the task (e.g., compensatory efforts). In a review of 11 studies, Grant et al. (2003) also examined the long-term sustained relationship between cannabis use and attention following abstinence. In their study, Grant et al. (2003) examined attention primarily using neuropsychological measures, finding a small ES for the influence of cannabis use on attention (ES, −0.083; 99% CI = –0.32–0.15). Finally, Schreiner and Dunn (2012) primarily examined neuropsychological test performance to determine any sustained impact of cannabis on attention performance, including the Continuous Performance Task and the Iowa Gambling Task (IGT). With the 13 component studies, the authors found a small ES for the sustained impact of cannabis on attention (ES, –0.20; 95% CI = −0.49–0.09).

An example of a component study from this section includes Crane et al. (2013). This study included 69 cannabis using 18- to 24-year-olds (mean age = 21 years). Attention was measured with four neuropsychological measures, including the IGT, the Balloon Analogue Risk Task, the Monetary Choice Questionnaire, and the GoStop Task. Interestingly, cannabis use was only associated with a significant difference on one measure (IGT and past year cannabis use, p <0.03; IGT and past-month cannabis use, p <0.003). There were no significant sustained associations between cannabis use on the other three measures of inhibition (for past year cannabis use and past-month cannabis use for the Balloon Analogue Risk Task, the Monetary Choice Questionnaire, and the GoStop Task).

Primary Literature

In this review of cannabis, the primary literature was searched when the systematic review content did not fully cover or address study questions. Given the breadth and scope of the existing systematic reviews in this domain, additional primary literature was not searched for the domain of attention.

Discussion of Findings

In sum, within the domain of learning, the Broyd et al. (2016) systematic review and the component study highlighted within that review showed strong data for the acute (immediate) impact of cannabis use on

learning. However, results from three systematic reviews (Batalla et al., 2013; Broyd et al., 2016; Martin-Santos et al., 2010) reflected limited to no support for the association between the sustained effects of cannabis use after cessation and the cognitive domain of learning. Similarly, for the domain of memory, the Broyd et al. (2016) systematic review and the component study within it showed moderate to strong evidence for the acute (immediate) impact of cannabis use on memory. However, as with learning, there were limited to no data to support the association between the sustained effects of cannabis use after cessation and the cognitive domain of memory in the three systematic reviews that addressed this question (Batalla et al., 2013; Broyd et al., 2016; Martin-Santos et al., 2010). Of interest, the neuroimaging studies reflected that while there was no difference in terms of performance on memory tasks, cannabis users may recruit different parts of their brain to achieve equivalent performance to control subjects on these tasks, suggesting the need to examine how cannabis may impact the neural regions that drive the processing of memory in future research. Finally, for the domain of attention, the Broyd et al. (2016) systematic review showed strong evidence for the acute (immediate) impact of cannabis on attention. However, as with the other domains, the evidence from other systematic reviews (Batalla et al., 2013; Broyd et al., 2016; Grant et al., 2003; Martin-Santos et al., 2010; Schreiner and Dunn, 2012) suggest that there were limited to no data to support the association between the sustained effects of cannabis use after cessation and the cognitive domain of attention.

ACADEMIC ACHIEVEMENT

Is There an Association Between Cannabis Use and Academic Achievement and Education?

For the psychosocial areas that go beyond cognition, there was one systematic review (Macleod et al., 2004) that examined the effects of can-

nabis on a number of psychosocial outcomes as reported in longitudinal studies of general population samples. Specifically, this review contributed to our evaluation of the research literature related to the effects of cannabis on academic achievement as well as social relationships and other social roles. There was no systematic review of the research literature on the effects of cannabis on employment and income.

Because only one systematic review was available, we also focused on the primary literature to address questions related to the effect of cannabis on (1) academic achievement; (2) employment and income; and (3) social relationships and other social roles. The primary literature to be reviewed and summarized is based on studies published subsequent to 1999. In selecting that literature, we focused on studies that met criteria derived from the Newcastle–Ottawa quality assessment scale. In particular, (1) prospective studies in which cannabis use occurred prior to the outcomes of interest; (2) multiple assessments of the variables of interest over time; (3) samples that are representative, either of the nation or a major subgroup; (4) multiple measures of cannabis use, involving frequency and/or quantity of use; (5) a relatively large sample size; and (6) consideration of relevant sociodemographic control variables such as sex/gender, age, family income, ethnicity/race, and/or history related to the outcome of interest.

Systematic Reviews

In their systematic review, Macleod et al. (2004) identified 16 high-quality longitudinal studies of the general population in which the effects of cannabis use on psychosocial outcomes, including educational attainment, were examined. The authors reported that cannabis use was consistently related to negative educational outcomes (measured primarily by drop-out rates), but they also noted that the strength of the association varied across the studies reviewed. In addition, including the appropriate control variables in the analyses typically resulted in a substantial decrease in the strength of the association. There was no evidence of a causal relationship between cannabis use and lower educational attainment.

Primary Literature

The primary literature published subsequent to Macleod et al.’s 2004 review continues to show that it is difficult to document a direct link between cannabis use and negative educational outcomes because other variables play a role. At best, indirect relationships have been reported. For example, Arria et al. (2013) used longitudinal growth curve modeling to analyze cannabis use and grade point average (GPA) data across 4

years of university education. They found no direct links from cannabis to GPA, but they did report an indirect path in which increased cannabis use led to increased skipping of classes, which resulted in lower GPA. Using data from the Coronary Artery Risk Development in Young Adults study, Braun et al. (2000) initially found an inverse relationship between past-month cannabis use and becoming a college graduate. When analyses were adjusted for variables such as age and parental education, this relationship disappeared, so that cannabis use was unrelated to college graduation.

There is some evidence to suggest that a higher frequency and persistence of cannabis use are associated with some negative educational outcomes. Using data from the Victoria Adolescent Health Cohort (1992–2003), Degenhardt et al. (2010) examined a cohort of a representative sample of Australian students (n = 1,943) from an average age of 14.9 years through an average of 24.1 years. Individuals who were persistent or weekly users of cannabis in adolescence and young adulthood had poorer post-school outcomes at age 24 years (adjusted odds ratio [aOR], 0.84; 95% CI = 0.55–1.3; n = 190)¹ compared with individuals who never used cannabis. Adjustment for background factors and cigarette smoking reduced this association.

The age at which cannabis use is initiated may be important in determining negative educational outcomes. Using data from three Australian cohort studies involving more than 6,000 participants, Horwood et al. (2010) reported that individuals who began to use cannabis before age 15 years experienced significantly greater negative educational outcomes, even after reductions in odds ratios (ORs) based on an adjustment for confounding variables. Pooled odds ratios (pOR) estimates indicated that the educational achievement of those who never used cannabis by age 18 years were 1.9 to 2.9 times greater than for those who used cannabis before the age of 15 years. The researchers found that individuals who had not used cannabis by age 18 were more likely to complete high school (pOR, 2.9; 95% CI = 1.8–4.6; p <0.001), enroll in university (pOR, 1.9; 95% CI = 1.5–2.4; p <0.001), and earn a university degree (pOR, 2.5; 95% CI = 1.8–3.5; p <0.001) compared to individuals who had used cannabis before age 18. In related findings, Brook et al. (2002) reported that minority youths ages 10 to 19 years who used cannabis had higher rates of being suspended or expelled from school (aOR, 2.68; 95% CI = 1.73–4.14; p <0.001).

Educational outcomes related to cannabis use tend to be confounded with the use of other substances, particularly tobacco/smoking cigarettes. Mokrysz et al. (2016) analyzed data from the Avon Longitudinal Study of

___________________

¹ Adjusted for non-Australian birth, symptoms of depression and anxiety in adolescence, high-risk alcohol use, and maximum level of cigarette smoking in adolescence.

Parents and Children, a propective study of 2,235 adolescents, 24 percent of whom reported using cannabis by the age of 15 years. When analyses included appropriate confounding variables (particularly tobacco use) even heavy (≥50 times) cannabis users (mean educational performance,² 69.2 percent; 95% CI = 65.0–73.3) did not significantly differ from never users in their educational performance at age 16 (mean educational performance, 80.8 percent; 95% CI = 80.2–81.4).

Similarly, McCaffrey et al. (2010) followed 4,500 adolescents for 4 years through high school and reported a positive association between cannabis use and dropout rates (OR, 5.6; risk ratio [RR] = 3.8). However, the remaining association (OR, 2.4; RR = 1.7) became statistically insignificant when the data were adjusted for cigarette use. Degenhendt et al.’s 2010 study found that occasional cannabis use was linked to lower educational outcomes (i.e., dropping out of school), but that the initial relationship was attenuated by tobacco use, which was relatively high in their sample. Green and Ensminger (2006) found that heavy use of cannabis during adolescence was associated with dropping out of school.

Discussion of Findings

Researchers have hypothesized and some studies have reported that cannabis use is linked to negative educational outcomes. However, the relationships among these variables are complex as are the ways in which the specific variables of interest are measured. In addition, all such research requires the careful consideration of a wide range of control variables that include sociodemographic confounders (e.g., gender/ sex, family socioeconomic status [SES]) and educational confounds (e.g., parental education, intelligence quotient [IQ], student’s cognitive ability) (Fergusson and Boden, 2008; Horwood et al., 2010). This complexity requires that researchers use sophisticated data-analytic techniques (e.g., propensity scoring to reduce confounding by measured factors) (McCaffrey et al., 2010). Use of less sophisticated approaches (e.g., correlations, logistic regression) can lead to an overestimation of the association between cannabis use and negative educational outcomes (McCaffrey et al., 2010). Typically, the primary literature cannot elucidate the mechanisms through which cannabis use may produce negative educational outcomes, although some have speculated that these outcomes may be related to cannabis’s effects on the brain, including cognitive impairment.

In all of the primary research literature reviewed on the effects of cannabis on academic achievement, employment and income, as well as social relationships and other social roles, there were a number of

___________________

² Measured in percentage of General Certificate of Secondary Education points.

limitations. Below, we summarize aspects of various studies that make it difficult to draw definitive conclusions regarding the causal relationships among cannabis use and the different psychosocial outcomes that we examined. They include the following:

1. Sample heterogeneity (e.g., differences related to sample’s SES, age, gender, ethnicity).
2. Inconsistent measures of cannabis use (yes/no; cross-sectional reports of frequency and/or quantity/amount; categories based on history of use).
3. Inconsistent/varying measures of the duration of cannabis use and outcome variables.
4. Even in longitudinal studies, the measures of interest often are cross-sectional snapshots.
5. The history and persistence of cannabis use is not always considered. In adolescence through adulthood, patterns of cannabis use can vary (groupings include consistent never users, occasional users, persistent heavy users, and so on).
6. In almost every study, the measure of cannabis use is based only on self-report, which cannot be validated.
7. Failure to consider individual characteristics (e.g., attitudes related to the outcomes of interest).
8. Multiple substances being used; it is difficult to separate out effects of cannabis relative to use of other drugs, including alcohol and smoking tobacco. Often cannabis effects are less strongly related to outcomes of interest.
9. The complexity of the relationships means that confounds must be considered and statistical analyses must be sophisticated. Many studies meet criteria for design and samples, but report outcomes based on less sophisticated analyses (e.g., correlations, logistic regressions).

EMPLOYMENT AND INCOME

Is There an Association Between Cannabis Use and Employment and Income?

Systematic Reviews

The committee did not identify a good- or fair-quality systematic review that reported on the association between cannabis use and employment and income.

Primary Literature

The primary literature to be reviewed and summarized is based on studies published subsequent to 1999. In selecting that literature, the committee focused on studies that met criteria derived from the Newcastle–Ottawa criteria (Wells et al., 2014), as described in the previous section.

Popovici and French (2014) analyzed two waves of panel data from the nationally representative National Epidemiologic Survey on Alcohol and Related Conditions. Initial analyses suggested a significant association between cannabis and employment status (implying poorer labor market outcomes; also see Fergusson and Boden, 2008). However, more sophisticated (fixed-effect) data analyses that considered individual sources of heterogeneity resulted in smaller and less significant relationships between cannabis and unemployment for men and women (OR, 0.813; 95% CI = 0.237–2.791 and OR, 0.777, 95% CI = 0.269–2.239, respectively). The researchers concluded that cannabis use is less detrimental to labor market participation than was suggested in previous research. A similar conclusion was reached by Lee et al. (2015a), who found that cannabis use was not related to unemployment (OR, 0.96; 95% CI = 0.91–1.01), but rather that it is confounded with the use of other substances such as drinking alcohol and tobacco use, which are associated with unemployment.

There are some studies that suggest that the persistent use of cannabis over longer periods of time is associated with unemployment. Zhang et al. (2016) reported that chronic cannabis users (who started in adolescence) were more likely to be unemployed at age 43 (across three decades) than non/experimental users (aOR, 3.51; 95% CI = 1.13–10.91). Braun et al. (2000) also found cannabis users to be less likely to be employed than nonusers. Those who were employed tended to have lower prestige occupations (measured by the Occupational Prestige Score [OPS]; across 10 years) compared to nonusers. Some of this may be related to lessened commitment to work among those who use cannabis over time. Hyggen (2012) found low work commitment (as measured by the Work Involve-

ment Scale) among cannabis users compared to abstainers, starting from young adulthood (ages 17 to 20 years) through to middle age (early to mid-40s).

Some of the negative effects of cannabis use on unemployment may be exacerbated among those from low SES backgrounds (Lee et al., 2015a). Other studies of low SES and minority samples also report that chronic cannabis use is related to increased unemployment (Green and Ensminger, 2006; Lee et al., 2015b). Disentangling the effects of cannabis use from other variables related to having a low SES and/or a disadvantaged background may be fruitful areas for future research.

Discussion of Findings

All of the committee’s conclusions are based on primary literature. In some cases, especially with more sophisticated data analyses, cannabis use has not been linked to outcomes such as labor market participation and unemployment. In other cases, a longer duration of cannabis use has been associated with unemployment. A lower SES may exacerbate these negative outcomes. Along with the limitations described on page 280, our examination of the literature on the relationship between cannabis use and employment was limited by the difficulty in determining causality. Because employment status is not static, it is possible that the relationships may be cyclical (e.g., depending on context, unemployment could contribute to the use of cannabis and other substances [Lee et al., 2015a] and cannabis/substance use could contribute to unemployment).

SOCIAL RELATIONSHIPS AND OTHER SOCIAL ROLES

Is There an Association Between Cannabis Use and Social Functioning and Social Roles?

Systematic Reviews

There was one systematic review that examined the effects of cannabis on social functioning as one of a number of outcomes in longitudinal studies of general population samples. In their systematic review, Macleod et al. (2004) identified 16 high-quality longitudinal studies of the general population in which the effects of cannabis use on psychosocial outcomes, including social functioning, were examined. The authors found that can-

nabis use was inconsistently related to social functioning as manifested by antisocial behaviors such as conduct disorder or delinquency, offending, and contact with police. Associations related to an individual’s gender and ethnicity also produced inconsistent findings. Using data from the Christchurch Health and Development Study (n = 1,265), Fergusson et al. (1996) reported that cannabis use at younger ages (<15 years) was consistently associated with antisocial behavior (aOR, 1.0; 95% CI = 0.5–2.1). Interestingly, the use of tobacco and alcohol showed similar associations.

Primary Literature

The primary literature has shown that there is a statistical association between cannabis use and social functioning as manifested by negative relationships with others, but there are too few good-quality studies to provide conclusive evidence of causation. Palamar et al. (2014) examined various psychosocial outcomes in a nationally representative sample of high school seniors (n = 7,437) from the Monitoring the Future study. They found that participants who had used cannabis 40 or more times had compromised relationships with teachers, supervisors, and parents. Cannabis users reported less interest in activities and more trouble with police. Interestingly, the adverse psychosocial outcomes for cannabis were less than those for alcohol. In a sample of African American and Puerto Rican young adults, cannabis use was associated with rebelliousness and engagement with fewer productive activities (Brook et al., 2002).

Chassin et al. (2010) reported that in a sample of juvenile offenders, cannabis use in adolescence was inversely related to “psychosocial maturity” (i.e., a measure of responsibility, temperance, and perspective taking) in young adulthood (χ² (5) = 13.49, p = 0.02; comparative fit index [CFI] = 0.991, RMSEA = 0.038). Such maturity is integral to being able to successfully engage in social relationships and to transition into adult social roles. Interestingly, in some cases the temporal sequencing of cannabis use and maturity fluctuated over time, suggesting that these relationships were not static; increases in cannabis use were associated with reduced maturity, and reductions in cannabis use were associated with increases in maturity.

There is some evidence to suggest that a higher frequency and persistence of cannabis use or, in particular, cannabis use during adolescence is associated with some negative social outcomes. Among a low-income sample of 274 African Americans, Green and Ensminger (2006) found that “heavy” (>20 times) cannabis use during adolescence (i.e., before age 17 years) was associated with poorer functioning in some social roles at ages 32 to 33 years. Compared to never using or experimenting with cannabis, heavy cannabis use was associated with unemployment (ES, −0.159; 95%

CI = −0.303 to −0.155; p = 0.030) and to parenting outside of marriage (ES, 0.109; 95% CI = −0.042–0.261).

Discussion of Findings

In the systematic review and primary literature, the findings indicate inconsistent relationships between cannabis use and social functioning. The primary literature included studies in which there was a relationship between cannabis use and adverse outcomes such as compromised relationships with authority figures and poorer functioning in social roles such as employment and parenting. Various limitations faced by the primary literature are described on page 282.

Researchers have hypothesized—and some studies have reported—that cannabis use is linked to negative social functioning and the ability to appropriately handle social roles. The relationships among these variables are complex, as are the ways in which the specific variables of interest are measured. In addition, all such research requires the careful consideration of a wide range of control variables that include sociodemographic confounds (e.g., gender/sex, family SES), the use of other substances (alcohol, other illicit drugs), and psychological problems such as depression or a personality disorder (Macleod et al., 2004). This complexity requires that researchers use sophisticated data-analytic techniques (e.g., propensity scoring to reduce selection bias; see Chassin et al., 2010). The use of less sophisticated approaches (e.g., correlations, logistic regression) can lead to an overestimation of the association between cannabis use and negative social outcomes (Macleod et al., 2004).

RESEARCH GAPS

To address the research gaps relevant to cognitive health and psychosocial functioning, the committee suggests the following:

• The systematic reviews that were reviewed by the committee did not necessarily parallel those in other fields of research that are covered in this report. As such, more studies that report quantitative data on the psychosocial effects of cannabis use are required to allow for a greater degree of comparison with the effects of cannabis use on the other health endpoints discussed in this report.

• It will be necessary to conduct further research on the developmental implications of cannabis use across age groups, particularly among adolescents, children, and the older populations. While the National Institute on Drug Abuse’s Adolescent Brain Cognitive Development study is in progress (see Box 11-2), at the time that this report was released, the findings of that study had not been published.

SUMMARY

This chapter summarizes the good- and fair-quality psychosocial literature published since 1999. The committee found that there is moderate evidence of an association between cannabis use and the impairment of the cognitive domains of verbal learning and attention but insufficient evidence for an association between cannabis use and the impairment of working memory. There is mixed evidence for the persistence of impairments or the recovery of function following an abstinence period of 24 hours or several weeks (25–32 days) without cannabis use in the domains of working memory, attention, and verbal learning (Broyd et al., 2016).

The committee found that it is difficult to document a direct link between cannabis use and negative educational outcomes because other variables play a role. There is some evidence to suggest that a higher frequency and persistence of cannabis use is associated with some negative educational outcomes. The age at which cannabis use is initiated may be important in determining negative educational outcomes. Educational outcomes related to cannabis use tend to be confounded with the use of other substances, particularly tobacco/smoking cigarettes. The primary literature has shown that there is an association between cannabis use and social functioning as manifested by negative relationships with others, but there are too few good-quality studies to provide conclusive evidence. There is some evidence to suggest that a higher frequency and persistence of cannabis use or cannabis use during adolescence is associated with some negative social outcomes. The literature provides limited evidence to support the hypothesis that cannabis use contributes to negative social functioning (e.g., conduct disorder, immature behavior) or to a failure to engage in developmentally appropriate social roles (e.g., marriage, parenting). The committee has formed a number of research conclusions related to these health endpoints (see Box 11-3); however, it is important that each of these conclusions be interpreted within the context of the limitations discussed in the Discussion of Findings sections.

REFERENCES

Brumback, T., N. Castro, J. Jacobus, and S. Tapert. 2016. Effects of marijuana use on brain structure and function: neuroimaging findings from a neurodevelopmental perspective. International Review of Neurobiology 129:33–65.

Chassin, L., J. Dmitrieva, K. Modecki, L. Steinberg, E. Cauffman, A. R. Piquero, G. P. Knight, and S. H. Losoya. 2010. Does adolescent alcohol and marijuana use predict suppressed growth in psychosocial maturity among male juvenile offenders? Psychology of Addictive Behaviors 24(1):48–60.

Conrod, P. K., and K. Nikolaou. 2016. Annual research review: On the developmental neuropsychology of substance use disorders. Journal of Child Psychology and Psychiatry 57(3):371–394.

Crane, N.A., R. M. Schuster, and R. Gonzalez. 2013. Preliminary evidence for a sex-specific relationship between amount of cannabis use and neurocognitive performance in young adult cannabis users. Journal of the International Neuropsychological Society 19:1009–1015.

Degenhardt, L., C. Coffey, J. B. Carlin, W. Swift, E. Moore, and G. C. Patton. 2010. Outcomes of occasional cannabis use in adolescence: 10-year follow-up study in Victoria, Australia. The British Journal of Psychiatry 196(4):290–295.

Feldstein Ewing, S.W., S. J. Blakemore, and A. Sakhardande. 2014. The effect of alcohol consumption on the adolescent brain: A systematic review of MRI and fMRI studies of alcohol-using youth. NeuroImage: Clinical 5:420–437.

Feldstein Ewing, S.W., T. I. Lovejoy, and E. Choo. 2017. How has legal recreational cannabis impacted adolescents in your state? A window of opportunity. American Journal of Public Health 107(2):246–247.

Fergusson, D. M., and J. M. Boden. 2008. Cannabis use and later life outcomes. Addiction 103(6):969–976.

Fergusson, D. M., M. T. Lynskey, and L. J. Horwood. 1996. The short-term consequences of early onset cannabis use. Journal of Abnormal Child Psychology 24:499–512.

Filbey, F. M., T. McQueeny, S. Kadamangudi, C. Bice, and A. Ketcherside. 2015. Combined effects of marijuana and nicotine on memory performance and hippocampal volume. Behavioural Brain Research 293:46–53.

Giedd, J. N. 2015. The amazing teen brain. Scientific American 312(6):32–37.

Grant, I., R. Gonzalez, C. L. Carey, L. Natarajan, and T. Wolfson. 2003. Non-acute (residual) neurocognitive effects of cannabis use: A meta-analytic study. Journal of the International Neuropsychological Society 9:679–689.

Green, K. M., and M. E. Ensminger. 2006. Adult social behavioral effects of heavy adolescent marijuana use among African Americans. Developmental Psychology 42(6):1168–1178.

Hanson, K. L., J. L. Winward, A. D. Schweinsburg, K. L. Medina, S. A. Brown, and S. F. Tapert. 2010. Longitudinal study of cognition among adolescent marijuana users over three weeks of abstinence. Addictive Behaviors 35(11):970–976.

Horwood, L. J., D. M. Fergusson, M. R. Hayatbakhsh, J. M. Najman, C. Coffey, G. C. Patton, E. Silins, and D. M. Hutchinson. 2010. Cannabis use and educational achievement: Findings from three Australasian cohort studies. Drug and Alcohol Dependence 110(3):247–253.

Hyggen, C. 2012. Does smoking cannabis affect work commitment? Addiction 107(7): 1309–1315.

IOM (Institute of Medicine). 1999. Marijuana and medicine: Assessing the science base. Washington, DC: National Academy Press.

Jacobus, J., L. K. Squeglia, A. D. Meruelo, N. Castro, T. Brumback, J. N. Giedd, and S. F. Tapert. 2015. Cortical thickness in adolescent marijuana and alcohol users: A three-year prospective study from adolescence to young adulthood. Developmental Cognitive Neuroscience 16:101–109.

Johnston, L. D., P. M. O’Malley, R. A. Miech, J. G. Bachman, and J. E. Schulenberg. 2015. Monitoring the Future national survey results on drug use, 1975–2014. 2014 Overview: Key findings on adolescent drug use. Ann Arbor: Institute for Social Research, the University of Michigan.

Lee, J. O., K. G. Hill, L. A. Hartigan, J. M. Boden, K. Guttmannova, R. Kosterman, J. A. Bailey, and R. F. Catalano. 2015a. Unemployment and substance use problems among young adults: Does childhood low socioeconomic status exacerbate the effect? Social Science and Medicine 143:36–44.

Lee, J. Y., J. S. Brook, S. J. Finch, and D. W. Brook. 2015b. Trajectories of marijuana use from adolescence to adulthood predicting unemployment in the mid 30s. American Journal on Addictions 24(5):452–459.

Lisdahl, K. M., E. R. Gilbart, N. E. Wright, and S. Shollenbarger. 2013. Dare to delay? The impacts of adolescent alcohol and marijuana use onset on cognition, brain structure, and function. Frontiers in Psychiatry 4(53):1–19.

Macleod, J., R. Oakes, A. Copello, I. Crome, M. Egger, M. Hickman, T. Oppenkowski, H. Stokes-Lampard, and G. Davey Smith. 2004. Psychological and social sequelae of cannabis and other illicit drug use by young people: A systematic review of longitudinal, general population studies. Lancet 363(9421):1579–1588.

Martin-Santos, R., A. B. Fagundo, J. A. Crippa, Z. Atakan, S. Bhattacharyya, P. Allen, P. Fusar-Poli, S. Borgwardt, M. Seal, G. F. Busatto, and P. McGuire. 2010. Neuroimaging in cannabis use: A systematic review of the literature. Psychological Medicine 40(3):383–398.

McCaffrey, D. F., R. L. Pacula, B. Han, and P. Ellickson. 2010. Marijuana use and high school dropout: The influence of unobservables. Health Economics 19(11):1281–1299.

Mokrysz, C., R. Landy, S. H. Gage, M. R. Munafo, J. P. Roiser, and H. V. Curran. 2016. Are IQ and educational outcomes in teenagers related to their cannabis use? A prospective cohort study. Journal of Psychopharmacology. doi: 0269881115622241.

Palamar, J. J., M. Fenstermaker, D. Kamboukos, D. C. Ompad, C. M. Cleland, and M. Weitzman. 2014. Adverse psychosocial outcomes associated with drug use among U.S. high school seniors: A comparison of alcohol and marijuana. American Journal of Drug and Alcohol Abuse 40(6):438–446.

Popovici, I., and M. T. French. 2014. Cannabis use, employment, and income: Fixed-effects analysis of panel data. Journal of Behavioral Health Services & Research 41(2):185–202.

Roten, A., N. L. Baker, and K. M. Gray. 2015. Cognitive performance in a placebo-controlled pharmacotherapy trial for youth with marijuana dependence. Addictive Behaviors 45:119–123.

Schmidt, L. A., L. M. Jacobs, and J. Spetz. 2016. Young people’s more permissive views about marijuana: Local impact of state laws or national trend? American Journal of Public Health 106(8):1498–1503.

Schreiner, A. M., and M. E. Dunn. 2012. Residual effects of cannabis use on neurocognitive performance after prolonged abstinence: A meta-analysis. Experimental and Clinical Psychopharmacology 20(5):420–429.

Wells, G. A., B. Shea, D. O’Connel, J. Peterson, V. Welch, M. Losos, and P. Tugwell. 2014. The Newcastle–Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp (accessed November 24, 2016).

Zhang, C., J. S. Brook, C. G. Leukefeld, and D. W. Brook. 2016. Trajectories of marijuana use from adolescence to adulthood as predictors of unemployment status in the early forties. American Journal on Addictions 25(3):203–209.