Longitudinal examination of coping-motivated marijuana use and problematic outcomes among emerging adults

Abstract

Objective

Cross-sectional research shows that coping-motivated marijuana use is associated with marijuana use and problems. However, limited research has examined how coping-motivated use might longitudinally relate to these outcomes. We examined the temporal relationship of coping-motivated marijuana use with severity of use and marijuana-related problems.

Method

Participants were 226 emerging adults, aged 18–25 years old, who currently used marijuana. Multilevel generalized linear models were used to evaluate the association between change in coping motives with change in frequency of marijuana use and marijuana problem severity from baseline to 6- and 12-month follow-ups.

Results

In the adjusted models, frequency of marijuana use was positively associated with between subject differences (IRR = 1.49; 95%CI: 1.30, 1.71; p < .001) but not within subject change over time (IRR = 1.09; 95%CI: 0.97, 1.22; p = .139) in use of marijuana to cope. Additionally, marijuana problem severity scores were associated positively with between subject differences (IRR = 1.45; 95%CI: 1.21, 1.75; p < .001) and within subject changes over time (IRR = 1.30; 95%CI: 1.07, 1.57; p < .01) in use of marijuana to cope.

Conclusions

Changes in coping-motivated use of marijuana in emerging adults were directionally associated with changes in marijuana use and marijuana problems up to 12 months post-baseline. Results highlight the possible bi-directional relationship between coping motives and marijuana use and problems. Findings could be valuable in helping practitioners go beyond quantity and frequency as sufficient metrics of marijuana use problems. Motives for use may reflect additional problems and the clinical need to explore these possibilities.

1. INTRODUCTION

1.1. Marijuana use is prevalent and increasing

Other than alcohol or tobacco, marijuana is the most commonly used substance among people aged 12 or older in the U.S., with past month use exceeding 27.7 million individuals (SAMHSA, 2019). Moreover, as marijuana legalization for medical and/or recreational use expands across states, the number of past year marijuana initiates continues to rise, with 8,400 new users each day in 2018 (SAMHSA, 2019). Although many individuals self-report using marijuana without any negative consequences (C. M. Lee, Neighbors, & Woods, 2007; Macleod et al., 2004; Stein, Hagerty, Herman, Phipps, & Anderson, 2011), there is a subset of individuals who appear to be at risk for negative outcomes related to their use. An important factor that might distinguish those at risk for marijuana-related problems from others is their motivation or reasons for using marijuana.

1.2. Marijuana use motives explain individual differences in marijuana use and related factors

Common reasons for using marijuana include use to cope with psychological distress (e.g., “To forget my worries”), to enhance experiences (e.g., “Because it’s fun”), for social reasons (e.g., “Because it makes a social gathering more enjoyable”), or for conformity purposes (e.g. “To fit in with the group I like”) (Cooper, Kuntsche, Levitt, Barber, & Wolf, 2016; Simons, Correia, Carey, & Borsari, 1998; Simons, Gaher, Correia, Hansen, & Christopher, 2005). While an individual can use marijuana for all of these reasons, enhancement or social reasons are the most commonly reported motives (e.g., Lee et al., 2007). Enhancement-motivated use is driven by a desire for excitement or joy, whereas using marijuana for social reasons facilitates social cohesion (Simons et al., 1998; Simons et al., 2005). One study found that enhancement-motivated use accounted for 23.1% of the variance in past 30-day use and that social motives accounted for 16.1% of the variance, whereas coping-motivated use only accounted for 9% of the variance (Bonn-Miller, Zvolensky, & Bernstein, 2007). Although linked to frequency of use (Bonar et al., 2017), enhancement and social motives have limited associations with negative factors like marijuana-related problems or psychological distress (Bonn-Miller, Zvolensky, et al., 2007; Brodbeck, Matter, Page, & Moggi, 2007; Terry-McElrath, O’Malley, & Johnston, 2009). Finally, although using marijuana to conform is associated with social anxiety symptoms (Buckner, Bonn-Miller, Zvolensky, & Schmidt, 2007), it has been found to negatively correlate with recent marijuana use (Bonn-Miller, Zvolensky, et al., 2007). In sum, most reasons for marijuana use are not associated with negative sequelae. Conversely, coping-motivated marijuana use is consistently associated with a number of negative mental health- and substance-related factors, suggesting it warrants specific attention in clinical treatment and research.

1.3. Using marijuana to cope is associated with myriad clinical concerns

Consistent with the stress-coping model (Wills & Shiffman, 1985) and the self-medication hypothesis initially purported in the alcohol use literature (Levenson, Sher, Grossman, Newman, & Newlin, 1980), people may consume marijuana as a coping response to stress, with the substance being used to engender positive affect and/or decrease an aversive mood. This dovetails with Cox and Klinger’s Incentive Motivation Model (Cox & Klinger, 2004),which suggests that individuals use substances to elicit an affective change. Thus, coping-motivated use might be negatively reinforced, suggesting underlying emotional disturbance. For instance, in a cross-sectional study of emerging adults with a history of trauma exposure, coping-motivated use was associated with posttraumatic stress symptoms, but other motives were not (Bonn-Miller, Vujanovic, Feldner, Bernstein, & Zvolensky, 2007). A recent study showed that coping motives are an important marker of marijuana use and related problems among those with a history of childhood physical abuse (Meshesha et al., 2019). In another recent cross-sectional study, results showed that depressive symptoms were associated with marijuana use and problems, and that this relationship might be explained by individuals’ problem-focused, ruminative thoughts (Bravo et al., 2019). Indeed, coping–motivated marijuana use is correlated with negative affect, anxious arousal, and depressive symptoms (Mitchell, Zvolensky, Marshall, Bonn-Miller, & Vujanovic, 2007). Coping-motivated marijuana use is also uniquely associated with emotional dysregulation (Bonn-Miller, Vujanovic, & Zvolensky, 2008) and social anxiety symptoms (Buckner et al., 2007) relative to other motives.

Unfortunately, coping-motivated use might also lead to problems. Individuals who report using marijuana to cope with distress are more likely to report problems related to their use, such as interpersonal strife, decreased productivity, or negative feelings about themselves (Buckner, 2013; C. M. Lee et al., 2007). Moreover, individuals who predominantly use marijuana to cope are more likely to endorse DSM-5 marijuana use disorder symptoms (Moitra, Christopher, Anderson, & Stein, 2015; Schultz, Bassett, Messina, & Correia, 2019). Taken together, these data paint a picture of a person who uses marijuana to cope as someone likely to have psychological distress and experience problems related to their use of marijuana.

In addition to being associated with marijuana-related problems, psychological distress, and trauma history, coping-motivated marijuana use could be associated with more persistent use. Coping-motivated use directly affects individuals’ perceived barriers to marijuana cessation, thus undermining abstinence (Peraza et al., 2019) and longitudinal data show coping-motivated use relates to future use (Sofis, Budney, Stanger, Knapp, & Borodovsky, 2020). The cross-sectional data seem to suggest that increases in coping motives might relate to increases in marijuana use or related problems. Two studies in the cognitive behavioral therapy field show that reductions of coping-motivated use can lead to reduced marijuana use (Banes, Stephens, Blevins, Walker, & Roffman, 2014; Blevins, Banes, Stephens, Walker, & Roffman, 2016), yet coping motives were not specifically targeted in treatment. Instead, they reduced as a by-product of the interventions. In the early stages of a longitudinal study, adolescents’ use of marijuana to cope with negative emotions and experiences was shown to be a risk factor for continued use in adulthood (Patrick, Schulenberg, O’Malley, Johnston, & Bachman, 2011). These results were further supported, as coping-motivated use in early adulthood predicted persistent use over 15 additional years of follow-up (Patrick, Bray, & Berglund, 2016). However, the field has had limited opportunities to further test the hypothesis that changes in coping-motivated use will be directionally consistent with changes in marijuana use and related problems. Thus, it remains unclear to what degree coping motives might fluctuate naturally or in response to intervention, and how those fluctuations might temporally relate to changes in marijuana use and marijuana-related problems. This is important as the field continues to consider ways to enhance treatment and gain more specificity about treatment targets.

1.4. The present study

In this study, we examined the 12-month longitudinal association among marijuana use frequency, marijuana-related problems, and coping-motivated use in 226 emerging adults who were part of a larger clinical trial aimed at reducing risky behaviors among emerging adults who use alcohol and marijuana (Stein et al., 2018). The active intervention did not produce any differential reductions in substance use, compared to a control condition, but important elements of the sample (e.g., equal representation of college and non-college students, emerging adult age group) provided valuable data for secondary analyses. Emerging adults (Arnett, 2001), those aged 18–25 years old, have the highest rates of past year marijuana use (34.8%) compared to youths aged 12–17 (12.5%) and older adults (13.3%) (SAMHSA, 2019). In emerging adults, this higher use is likely to be attributed to coping purposes (Titus, Godley, & White, 2006) and coping-motivated use is the most robust predictor of marijuana problems. (Buckner, 2013). Moreover, marijuana use is positively associated with functional impairments, such as in ability to complete work, in emerging adults (Osuch, Vingilis, Ross, Forster, & Summerhurst, 2013). Overall, emerging adults are a key demographic subset in which these unanswered temporal questions of the relationship between coping-motivated use and marijuana outcomes should be explored. We predicted that changes in coping-motivated use (e.g., increases, decreases) would be significantly associated with directional, longitudinal changes (e.g., increased, decreased) in: (a) marijuana use; and, (b) marijuana-related problems.

2. MATERIAL AND METHODS

The study was approved by the Butler Hospital Institutional Review Board. Study participants were recruited for a larger study of health behaviors and sexual risk among emerging adults. This parent study compared the effects of a motivational intervention to a health education control condition on substance and sexual health outcomes. The active intervention focused primarily on themes of emerging adulthood (e.g., identity exploration, instability, self-focus, feeling in-between, a sense of possibilities) and the individuals’ relationship to substance use. Motives for use, such as coping, were not a target of the intervention. Recruitment targeted the Rhode Island/southeastern Massachusetts geographical area through advertisements placed online, in local college newspapers, on public transportation, and on commercial radio. Interested individuals contacted study staff and completed a 10-minute anonymous phone screen. The screen included questions about demographic characteristics, substance use, sexual activity, mental health, and general health. Eligible individuals were invited for an in-person interview at the research site and offered compensation ($40) and free sexually transmitted infection testing. Additionally, they completed follow-up assessments at 6- and 12-months post-baseline.

Eligibility criteria for the parent study included being 18–25 years old, drinking alcohol and/or using marijuana in the last month, being heterosexually active in the last six months, not married or living with a romantic partner for the last 12 months, not having suicidal ideation in the past two weeks, and living within 30 minutes of the research site. All participants examined here were active marijuana users at baseline. Of the 2,645 individuals screened by phone, 2,002 were ineligible. The most common reasons for ineligibility were not using both substances at the study threshold levels (n=976), having suicidal ideation (n=234), being outside the age range (n=148), cocaine or opioid use (n=81), and living at a distance (n=53). The remaining individuals were ineligible for other or multiple reasons. The 643 eligible persons were invited for an interview and 402 were either not interested (n=47 actively refused; n=54 passively refused, i.e. said they would call back to schedule an appointment, but never did), or did not keep a scheduled appointment (n=301) at the study site. Two hundred and forty-one persons provided written informed consent and completed a baseline interview after which 13 persons were found to be ineligible and 2 withdrew their participation. The final sample here of active marijuana users included 226 persons.

2.1. Measures

In addition to demographics assessed at baseline, the following data were collected at baseline, 6-, and 12-month assessments.

Marijuana Problem Scale (MPS; Stephens, Roffman, & Curin, 2000; Stephens et al., 2004)

The MPS is a reliable and valid measure of 19 problems related to marijuana in the past 90 days. These ranged from losing a job to having withdrawal symptoms to having problems in one’s family as a direct consequence of marijuana use. Possible scores range from 0–48. In this sample, internal consistency was .81.

Reasons for Marijuana Use (RMU; Cooper, Russell, Skinner, & Windle, 1992)

Consistent with prior studies (Moitra et al., 2015), we used the adapted Reasons for Drinking measure developed by Cooper and colleagues to examine marijuana use motives. The RMU was chosen because it is brief (15 items), compared to other motives measures (e.g., 25-item Marijuana Motives Measure; (Simons et al., 1998). Additionally, it focuses on the three most common reasons for use: (1) coping; (2) enhancement; and, (3) social. Because the parent trial did not target use motives, this measure was deemed sufficient for secondary data collection. The RMU subscales had a possible range of 1–4, corresponding to, “Never/Almost Never,” “Sometimes,” “Often,” and, “Almost Always.” In this sample, internal consistency reliabilities were .84, .84, and .73 for the coping, enhancement, and social scales, respectively.

Timeline Follow-back measure (TLFB; Sobell & Sobell, 1996)

a semi-structured interview that uses a calendar-guided approach (Fals-Stewart, O’Farrell, Freitas, McFarlin, & Rutigliano, 2000), assessed marijuana use in the past 30 days at baseline and frequency of use since the previous assessment at follow-up.

2.2. Data Analysis

We present descriptive statistics to summarize the baseline characteristics of the sample. Frequency of marijuana use and number of marijuana problems endorsed, based on severity, can be conceptualized as count variables that are zero-limited and positively skewed. We considered specificity testing as discussed by Garber and Hollon (1991) but determined that inclusion of covariates in our model implicitly tests for spuriousness. Instead, we used a multilevel generalized linear model with log link, Poisson family error distribution, and robust standard errors to evaluate the association between change in coping with change in frequency of marijuana use and marijuana problem severity. We generated orthogonal variables to represent the between subject differences in use of marijuana to cope, and the effect of within subject change in use of marijuana to cope. The between subject is calculated as $x\overline{i}j$, where i is the ith subject and j is the jth observation (baseline, 6-, and 12-months). The effect representing within subject change is calculated as the within subject deviation (i.e., xij -$x\overline{i}j$; (Rabe-Hesketh & Scrondal, 2012). The between effect estimates between subject differences regardless of assessment month; the within subject effect estimates the effects of within subject changes between assessment periods. This method evaluates the within subject effect of increases or decreases in motivations across time periods without assuming such changes are linear. Because these data were derived from a randomized controlled trial, we included intervention arm as a covariate to account for any cohort effect, although this was unlikely given that trial outcomes were not different between the two study conditions. Other covariates included age, gender (1 if male), race (1 if non-Latinx White), ethnicity (1 if Latinx), marijuana enhancement motives, marijuana social motives, and indicator variable for month of assessment.

3. RESULTS

Participants averaged 21.2 (± 2.0) years of age, 55.3% were male, 12.4% were Latinx, 66.8% were non-Latinx White, 11.1% were Black, and 22.1% identified other or multiple racial origins (Table 1). Race was dichotomized to contrast non-Latinx Whites to people of color in subsequent analyses. The mean 30-day rate of marijuana use was 19.4 (± 21.7; Median = 21.7) days. On average, participants endorsed 5.2 (± 3.6; Median = 5.0) statements as minor or severe marijuana problems. Mean scores were 2.4 (± 0.7), 2.5 (± 0.7), and 3.1 (± 0.6) on use of marijuana to cope, use of marijuana to socialize, and use of marijuana for enhancement, respectively; index scores ranged from 1 – 4. About 75.7% of participants (n=171) were observed at 6-months, 66.4% at 12-months (n=150), and 81.0% at either the 6- or 12-month assessments (n=183).

Table 1.

Sample Description (n = 226).

	n (%)	Mean (± SD)	Median	Range
Age		21.2 (± 2.0)	21	18 – 25
Gender (Male)	125 (55.3%)
Ethnicity (Latinx)	28 (12.4%)
Race
White	151 (66.8%)
Black	25 (11.1%)
Other	50 (22.1%)
30-Day Marijuana Use Ratea		19.4 (± 21.7)	21.7	0 – 30
Marijuana Problem Severityb		5.2 (± 3.6)	5.0	0 – 16
Use Marijuana to Copec		2.4 (± 0.7)	2.2	1 – 4
Use Marijuana to Socializec		2.5 (± 0.7)	2.4	1 – 4
Use Marijuana for Enhancementc		3.1 (± 0.6)	3.2	1 – 4
Observed @ 6-Months	169 (75.8%)
Observed @ 12 - Months	150 (66.4%)
Observed @ 6- or 12-Months	183 (81.0%)

^aFrequency of marijuana use was expressed as a 30-day rate ((30-day rate = # of days used marijuana / TLFB days observed) * 30)).

^bNineteen item index coded 0 if no problem, 1 if minor or severe problem (Cronbach’s α = .81).

^cThe motivation to use marijuana measures are 5-item indexes with items coded 1 = “Never/Almost Never”, 2 = “Sometimes”, 3 = “Often”, and 4 = “Almost Always.” Index scores are the mean of 5-items comprising the scale. Cronbach’s α = .84, .77, and .81 for coping, to socialize, and for enhancement, respectively.

Males used marijuana significantly more frequently than females (IRR = 1.42; 95%CI: 1.22, 1.66; p < .001), but frequency of marijuana use was not associated significantly with intervention arm, age, race, or ethnicity (Table 2). Frequency of marijuana use did not differ significantly by month of assessment (Wald χ² = 2.99, df = 2, p = .224). Frequency of marijuana use was positively and significantly associated with between subject differences (IRR = 1.49; 95%CI: 1.30, 1.71; p < .001) but was not associated significantly with within subject differences in use of marijuana to cope over time or with the between or within measures of using marijuana to socialize or for enhancement.

Table 2.

Mixed Effects Poisson Regression Estimating the Between and Within Subject Effects of Motivations for Marijuana Use on Rate of Marijuana Use (n = 226 Persons Observed on 544 Occasions).

		95% CIa
	IRR	LCL	UCL	za	p = a
Active Intervention (Yes)	1.12	0.98	1.30	1.62	0.104
Years Age	1.03	0.99	1.06	1.43	0.154
Sex (Male)	1.42	1.22	1.66	4.48	0.000
Race (White)	1.05	0.88	1.24	0.55	0.580
Latinx (Yes)	0.93	0.72	1.18	−0.62	0.538
Assessment Month
6-month	0.94	0.87	1.02	−1.55	0.121
12-month	0.94	0.87	1.02	−1.52	0.130
Baseline [REF]	[1.00]
Coping (Between Subject)c	1.49	1.30	1.71	5.69	0.000
Coping (Within Subject Change)d	1.09	0.97	1.23	1.48	0.139
Socialize (Between Subject)c	0.91	0.78	1.05	−1.33	0.183
Socialize (Within Subject Change)d	0.98	0.89	1.08	−0.33	0.745
Enhancement (Between Subject)c	1.08	0.92	1.26	0.89	0.372
Enhancement (Within Subject Change)d	1.10	0.99	1.24	1.72	0.085
Constant	3.29	1.26	8.62	2.43	0.015

^aConfidence interval estimates and tests of significance are based on the Huber-White robust Variance estimator.

^bRate of marijuana use did not vary significantly by month of assessment (Wald χ² = 2.99, df = 2, p = .224).

^cFor each subject this is calculated as the mean score on the corresponding motivation to use measure across the baseline, 6-, and 12-month assessments.

^dWithin subject deviation from the mean at each time point.

Marijuana problem severity was associated inversely with age (IRR = 0.96; 95%CI: 0.92, 1.00; p = .041; Table 3). Marijuana problem severity was not associated significantly with intervention arm, race, or ethnicity. Marijuana problem severity was associated significantly with time of assessment (Wald χ² = 37.72, df = 2, p < .001); relative to baseline, marijuana problem severity index scores were lower at 6-months (IRR = 0.81; 95%CI: 0.73, 0.89; p < .001) and 12-months (IRR = 0.66; 95%CI: 0.57, 0.76; p < .001). Marijuana problem severity scores were positively and significantly associated with between subject differences (IRR = 1.45; 95%CI: 1.21, 1.75; p < .001) and within subject changes (IRR = 1.30; 95%CI: 1.07, 1.57; p = .007) in use of marijuana to cope. Marijuana problem severity was not associated significantly with between or within subject differences in either use of marijuana to socialize or for enhancement.

Table 3.

Mixed Effects Poisson Regression Estimating the Between and Within Subject Effects of Motivations for Marijuana Use on Marijuana Problem Severity (n = 226 Persons Observed on 544 Occasions).

		95% CIa
	IRR	LCL	UCL	za	p = a
Active Intervention (Yes)	0.98	0.83	1.16	−0.24	0.813
Years Age	0.96	0.92	1.00	−2.04	0.041
Sex (Male)	1.17	0.99	1.38	1.81	0.070
Race (White)	0.96	0.80	1.16	−0.42	0.674
Latinx (Yes)	0.91	0.70	1.18	−0.69	0.487
Assessment Month
6-month	0.81	0.73	0.89	−4.19	0.000
12-month	0.66	0.57	0.76	−5.95	0.000
Baseline [REF]	[1.00]
Coping (Between Subject)c	1.45	1.21	1.75	3.94	0.000
Coping (Within Subject Change)d	1.30	1.07	1.57	2.69	0.007
Socialize (Between Subject)c	1.16	0.97	1.39	1.67	0.095
Socialize (Within Subject Change)d	1.11	0.97	1.29	1.48	0.139
Enhancement (Between Subject)c	0.99	0.84	1.18	−0.08	0.940
Enhancement (Within Subject Change)d	1.02	0.85	1.21	0.19	0.853
Constant	3.29	1.24	8.78	2.38	0.000

^aConfidence interval estimates and tests of significance are based on the Huber-White robust Variance estimator.

^bMarijuana problem severity varied significantly by month of assessment (Wald χ² = 37.72, df = 2, p < .001).

^cFor each subject this is calculated as the mean score on the corresponding motivation to use measure across the baseline, 6-, and 12-month assessments.

^dWithin subject deviation from the mean at each time point.

4. DISCUSSION

After controlling for demographic variables, time, intervention arm, and other marijuana use motives, coping-motivated use was significantly associated with marijuana use and marijuana-related problems at follow-up. Interestingly, results showed significant within and between subject effects for problem severity and between subject effects for rate of use. The within subject effect means that as use to cope increases (decreases) over time, scores on marijuana use problem severity increase (decrease) over time. The between subject effect means that persons with a higher average level of use to cope have higher average use frequency and problem severity. These results align with a recent investigation by Colder and colleagues (2019), which showed that among adolescents who aged into emerging adulthood over three years, between and within person marijuana coping motives were associated with greater marijuana problems. Previous marijuana motives research showed that coping motives were associated with psychiatric symptoms, such as anxiety and depression (Mitchell et al., 2007), and that coping-motivated use was more common among individuals with marijuana-related problems (Buckner, 2013). As observed by Bonar and colleagues (2017), motives for use can predict consumption, with coping motives being associated with higher use, although this result is not consistently supported (Pearson, Bravo, Conner, & Parnes, 2020).

In this sample, males used significantly more marijuana and had greater marijuana problems. As such, taken across all time periods, the mean frequency of marijuana use is higher for males than females. These findings align with extant data showing that males are more likely to use marijuana, particularly at problematic rates (Chen, Yu, Lasopa, & Cottler, 2017; LaBrie, Grossbard, & Hummer, 2009). Results showed that the expected marijuana use decreased if/when coping motives decreased; conversely, use and problems increased as coping increased. These results were consistent with our hypotheses and align with prior research. For example, 3-year longitudinal data from college students showed that coping motives and marijuana use had cross-lagged associations, suggesting that marijuana use reinforces coping motives and vice versa (C. A. Lee, Derefinko, Davis, Milich, & Lynam, 2017). Collectively, these data underscore the likely bi-directional relationship between coping motives and marijuana use and problems, which could be rooted in negative reinforcement. In sum, this study adds to the burgeoning longitudinal marijuana research, and found that using to cope is associated with higher marijuana use and with higher levels of problems associated with use.

4.1. Limitations

The study was limited by our use of a marijuana use motives measure that consisted of select subscales from a larger measure. However, we have used the measure in other research and it appears to be a valid and reliable assessment (Moitra et al., 2015). Second, because these data are from a clinical trial, most participants’ marijuana use decreased over time, which might be different from a naturalistic cohort. The strength of the sample was that it was not a treatment-seeking sample, it was ethnically/racially diverse, half female, and unlike most emerging adult cohorts, included a substantial number of individuals not currently in school (40.3%). Lastly, we did not use a cross-lagged panel model in analyses. This means that we cannot assert that change in use to cope causes change in frequency of use or problem severity. Although this is conceptually possible, no data are available to inform whether or not it is likely that coping-motivated use precedes these other variables or vice versa. Instead, results demonstrate that these variables fluctuate concurrently over time.

4.2. Conclusions

Understanding of marijuana use motives is evolving and in light of recent legalization, some argue that a health-related motive (e.g., to have a better appetite, to feel more energetic) might be an important driver of marijuana use (Chabrol, Beck, & Laconi, 2017). This could be a slight variation of coping-motivated use, in which marijuana is expected to facilitate a positive mood or physical change. Finally, these findings could be valuable in helping practitioners go beyond quantity and frequency as sufficient metrics of marijuana use problems, especially as legalization spreads. The way marijuana is used (e.g., to cope) may reflect additional problems and the clinical need to explore these possibilities. Indeed, intervention on coping motives might be a key pathway to beneficial outcomes. This could start with helping patients recognize the costs associated with coping-motivated use. It will also depend on patients learning (or more frequently using) healthier coping strategies. By practicing alternative coping approaches, patients might gain insight into those approaches that are most help to them with the least consequences.

Highlights.

• Longitudinal analysis of coping-motivated marijuana use and related outcomes.

• Changes in coping-motivated use were related to changes in marijuana use.

• Changes in coping-motivated use were related to changes in marijuana problems.