Trends in Cannabis-Related Attitudes and Behaviors among Cannabis-Using Adolescent and Young Adult Outpatients following Medical Cannabis Legalization in Massachusetts

Abstract

Background: Among youth already using cannabis, legalization of medical cannabis may influence cannabis-related attitudes and behaviors, including increasing access through use of someone else's medical cannabis (diversion). Objective: To examine cannabis-related attitudes and behaviors (including diverted cannabis use) in cannabis-using youth in the four years following medical cannabis legalization. Additionally, we investigated characteristics of youth who used vs. did not use diverted medical cannabis. Methods: Data were collected in Boston from 2013 (when medical cannabis legislation took effect in Massachusetts) through 2016 (when recreational cannabis use became legal in Massachusetts). Cannabis-using youth (age 13–24) presenting to an outpatient adolescent substance use treatment program (ASUTP) or recruited for an adolescent medicine clinic study (AMCS) completed a confidential survey on demographic characteristics and cannabis use behaviors and attitudes. We used multiple logistic regression to analyze changes in attitudes and behaviors over three years versus the reference year (2013), controlling for demographics. We used chi-square to compare characteristics of youth reporting use of diverted medical cannabis versus those not. Results: The sample included 273 cannabis-using youth (ASUTP n = 203, AMCS n = 70; 2013 n = 67, 2014 n = 67, 2015 n = 77, 2016 n = 62). Mean ± SD age was 18.2 ± 2.5 years, 32% were female, 58% were White non-Hispanic, and 70% had college-graduate parents. In 2013, most youth reported that cannabis was easy to obtain (97.9%), and that occasional cannabis use had “no” or “slight” risk of harm (89.4%), with little change across years. In 2016, 44% of youth reported using someone else's medical cannabis, versus 15% in 2013 (aOR 4.66, 95% CI 1.81, 11.95). Youth using diverted medical cannabis had higher likelihood of reporting riding with a driver, or driving themselves, after cannabis use (both p < .01). Conclusion: Among at-risk youth in Massachusetts, use of diverted medical cannabis increased after medical cannabis legalization, and those using diverted medical cannabis reported higher risk for cannabis-related traffic injury.

Keywords

Marijuana cannabis adolescent health behavioral health substance use health policy

Cannabis is the second most commonly used substance among adolescents 1 and is the leading drug reported by adolescents entering substance use treatment. 2 ^, 3 Growing evidence indicates that exposure to cannabis during adolescence poses serious risks, including harm to brain development and long-term cognitive function. 4 Cannabis use at any age has been linked to an increased risk for a variety of psychiatric disorders, including depression, anxiety, schizophrenia, and substance use disorder. 5 Cannabis use also poses public safety risks, like motor vehicle injury or mortality due to driving impairment. 6 ^, 7

Perceived risk and attitudes are important predictors of cannabis use. 8 – 10 Adolescent perceptions of cannabis riskiness have declined over time, 11 despite cannabis being much more potent today than it was several decades ago. 12 With growing legalization of medical cannabis use 13 ^, 14 (as of November 2020, there are 40 U.S. states and territories with legalized medical cannabis 14 – 16 ), youth may have easier access to cannabis products, for example, through a registered cardholder (diversion), or may relax their perception of risk toward the substance, despite its adverse effects on adolescent health. 17

After decriminalizing the possession of cannabis in small amounts in 2008, Massachusetts legalized cannabis for medical use in 2012, becoming the eighteenth state to do so. 18 The medical marijuana law (MML) went into effect on January 1, 2013. It permits those with qualifying “debilitating” health conditions—such as cancer, glaucoma, Crohn's disease, Parkinson's disease, or multiple sclerosis—to use and possess cannabis under a physician's supervision and the approval of the Massachusetts Department of Public Health. 18 – 20 The policy eliminates civil and criminal penalties associated with cannabis use for registered patients, allows for the creation of statewide retail dispensaries, and permits home cultivation in limited amounts. 19 No minimum age for obtaining a medical card was specified, although patients under 18 years old require a valid diagnosis from at least two licensed Massachusetts physicians before qualifying. By 2013, when Massachusetts’ policy went into effect, several surrounding states already had MMLs in place—Maine (1999), Vermont (2004), Rhode Island (2007), and Connecticut (2012)—with New Hampshire following shortly after (2013). 14

The passage of the MML presented a unique “natural experiment” to examine changes associated with this policy in cannabis-related attitudes and behaviors among youth, including diverted cannabis use. Youth already using cannabis are a particularly important population to study as they may be more sensitive to the effects of policies that liberalize access to cannabis, compared to non-using youth. The primary objectives of this study were to (1) to determine how cannabis-related behaviors, attitudes, or perceptions among cannabis-using youth in Massachusetts changed over the initial years following MML passage, (2) to examine the trend in youth use of diverted medical cannabis during this time period, and (3) to compare the characteristics of youth using versus not using diverted medical cannabis.

Methods

Setting and participants

Data for this analysis were collected from 2013 (when Massachusetts’ MML took effect) through November 2016 (when recreational cannabis use became legal for adults ≥21 years old in Massachusetts 20 ). This period captures the first four years of MML rollout and stops at the point of recreational legalization, so as to minimize “contamination” of the recreational policy's passage on the medical law's effects in our study sample.

The sample included cannabis-using adolescents aged 13–24 years old who were recruited from two Boston, Massachusetts, clinic populations: an outpatient adolescent substance use treatment program and five clinics participating in a research study on a cannabis intervention for adolescents.

Adolescent substance use treatment program (ASUTP)

Eligible participants were adolescents presenting to ASUTP who reported any past-year cannabis use during their initial evaluation. Patients deemed medically or emotionally unstable by their provider at the time of their visit were excluded. For patients who met the eligibility criteria, their provider described the study purpose, risks and benefits, voluntary nature of participation, and confidentiality protections and presented them and their parents with an informational flyer reviewing study purposes and procedures. Patients who agreed to participate were provided with the questionnaire, which was completed privately in the exam room or waiting room and returned to an anonymous survey collection box. Completion of the questionnaire served as evidence of assent (if younger than 18 years old) or consent (if 18 years or older). No compensation for participation was provided. Of 363 eligible patients, 214 completed a questionnaire (58.9%).

Adolescent medicine clinic study (AMCS)

AMCS participants were recruited from five clinics affiliated with an urban children's hospital, including three pediatric/adolescent clinics on the hospital campus and two hospital-affiliated community health clinics. Detailed information about the study methods has been described previously. 21 Eligible participants were patients aged 15–24 arriving for routine primary care visits who used cannabis ≥3 times per week, able to read and understand English, deemed medically and emotionally stable by their provider at their time of visit, and able to enroll in a 4-month cannabis intervention study. Those reporting heavy or dangerous substance use other than cannabis in the past 30 days were excluded, as were parenting youth. Three hundred and nineteen patients were interested and referred to be screened. Of 139 deemed eligible, 70 enrolled and completed the baseline questionnaire (50.4%).

The hospital institutional review board approved this study and waived parental permission for participants under 18 years old.

Data collection

Each ASUTP patient completed a 5-minute, self-administered, paper-pencil survey containing 28 total items. Each AMCS patient in the cannabis intervention study completed a baseline computerized survey, which included 24 questions used for this analysis. Both surveys collected information regarding cannabis use history and frequency (13 questions, e.g., “How old were you when you first started using marijuana at least once a week?”), ever in lifetime driving while high (ASUTP: “During your LIFE, how many times did you DRIVE when you had been using marijuana?”; AMCS: “Have you ever driven a car or motorcycle while high on marijuana?”; see Analysis for reconciliation of response categories), diverted medical cannabis use (“Have you ever gotten marijuana from someone who had obtained it with his/her medical marijuana card?”), and demographic characteristics (9 questions). Additionally, the ASUTP survey included 6 items on perceived risk of harm from using cannabis, and about the effects of the MML policy change (including on ease of access, potency, price, and legal repercussions).

Analysis

Of 284 participants (214 from ASUTP, 70 from AMP) 7 were excluded due to incomplete data for analyses, and 4 were excluded for being medical cannabis cardholders, leaving a total analytic sample of 273.

Statistical analyses were performed using SAS^® Studio (SAS Institute Inc., Cary, NC, USA). In determining variables for analyses, we reconciled response categories for similar but not identical items between the two surveys. We computed descriptive statistics on the demographics overall and for each individual year following MML passage. Age-related variables were dichotomized to preserve cell sizes for analysis. Participant age was dichotomized into <18 years and 18 or older, reflecting the critical age for eligibility to obtain a medical cannabis card as an adult. We dichotomized age of initiation of any cannabis use into <16 years vs. 16 or older, as initiation before age 16 has been shown to carry enhanced neurobiological risks. 22 ^, 23 Age of initiating weekly use was dichotomized into <17 years vs. 17 or older, as our previous research suggested that weekly cannabis use develops, on average, one year after initiating any use. 24 We dichotomized past-90-day cannabis use into >60 days, reflecting daily or near daily use 25 , versus ≤60 days. Ever driving after cannabis use, measured as a frequency in the ASUTP survey, was dichotomized to “never” versus “at least once” for consistency with AMCS response categories. The driving variable was not age-restricted as participants as young as 13 years old in our sample reported having driven after cannabis use. We used the chi-square test to compare participants’ demographic characteristics across years. We controlled for any significant demographic differences across years in our analysis of changes in behaviors and perceptions across years. Statistical significance for all tests was set at p < .05.

We used multiple logistic regression to compute adjusted odds ratios comparing cannabis-related behaviors and perceptions in the years following MML passage versus the reference year (2013), controlling for demographic covariates (age, gender, recruitment site, race/Hispanic ethnicity, number of parents living at home, highest parent education level). Because of multicollinearity with age, highest completed grade in school, while significantly different across years, was not entered into the model.

Finally, we used multiple logistic regression to compare the demographic characteristics and other cannabis-related behaviors of youth using diverted medical cannabis versus those not using it. When data from both clinic samples were available for a cannabis variable, we controlled for recruitment site. Because of small cell sizes for some categories, we dichotomized race/ethnicity into White non-Hispanic vs. other for this analysis.

Results

Sample characteristics

Overall sample

The sample consisted of 273 youth, 203 (74.4%) recruited from the ASUTP and 70 (25.6%) from the AMCS (Table 1). Mean participant age was 18.2 ± 2.5 years old, with 48.0% under age 18; 32.4% were female, 57.7% were White, non-Hispanic, 59.7% lived with two or more parents at home, 69.9% had college-graduate parent(s), and 74.6% were still in school. Nearly the entire sample (96.7%) were Massachusetts residents.

Table 1

Sample characteristics overall and across data collection years, 2013–2016.

	Total SampleN (%)	2013n (%)	2014n (%)	2015n (%)	2016n (%)	Year comparisonp-value
Total respondents	273	67 (24.5)	67 (24.5)	77 (28.2)	62 (22.7)	–
Clinic^a						<.001
AMCS	70 (25.6)	20 (29.8)	25 (37.3)	24 (31.2)	1 (1.6)
ASUTP	203 (74.4)	47 (70.2)	42 (62.7)	53 (68.8)	61 (98.4)
Age, under 18 years^b	131 (48.0)	22 (32.8)	30 (44.8)	35 (45.5)	44 (71.0)	<.001
Female	88 (32.4)	23 (34.3)	24 (35.8)	27 (35.5)	14 (22.6)	.32
Race						<.001
White non-Hispanic	157 (57.7)	43 (64.2)	26 (38.8)	40 (52.6)	48 (77.4)
Black non-Hispanic	46 (16.9)	12 (17.9)	16 (23.9)	15 (19.7)	3 (4.8)
Hispanic	41 (15.1)	7 (10.5)	19 (28.4)	10 (13.2)	5 (8.1)
Other	28 (10.3)	5 (7.5)	6 (9.0)	11 (14.5)	6 (9.7)
Currently in school	200 (74.6)	52 (77.6)	44 (65.7)	54 (73.0)	50 (83.3)	.13
Completed high school or higher	127 (46.9)	36 (54.5)	34 (50.8)	39 (51.3)	18 (29.0)	.02
Two or more parents at home	160 (59.7)	35 (52.2)	38 (56.7)	41 (55.4)	46 (76.7)	.02
Parent(s) graduated college	172 (69.9)	39 (63.9)	32 (59.3)	50 (72.5)	51 (82.3)	.03

Notes.

AMCS = adolescent medicine clinic study; ASUTP = adolescent substance use treatment program.

Range: 13–24 years.

The mean age of cannabis use initiation was 14.4 ± 1.8 years old, with 79.5% initiating under the age of 16 (data not shown). Roughly one in four participants (23.7%, n = 45) reported more than 60 days of cannabis use in the past 90 days. The mean age at which cannabis use frequency became weekly or more often was 15.4 ± 2.0 years old, with 71.7% beginning weekly cannabis use under the age of 17. About half (47.8%, n = 129) reported ever driving after cannabis use, and a majority (80.1%, n = 161) reported ever riding with a driver who had used cannabis.

Comparison of clinic samples

AMCS participants were more likely to be female (60% vs. 22.8%, p < .001) and less likely to have college-graduate parents (29.8% vs. 79.4%, p < .001) compared to ASUTP participants (Supplemental Table). Youth from the two clinics significantly differed by race/ethnicity (p < .001), with the proportion of White, non-Hispanic participants higher for ASUTP than AMCS (73.8% vs. 11.4%, respectively). AMCS participants were older than ASUTP participants (20.7 ± 1.9 years vs. 17.3 ± 2.1 years, p < .001), and initiated lifetime (p < .001) and weekly (p = .001) cannabis use at older ages than ASUTP participants.

Comparison across years

Observations were distributed roughly evenly across data collection years (2013 n = 67, 2014 n = 67, 2015 n = 77, 2016 n = 62). Across the year-specific samples, more participants in 2016 were under age 18, were White non-Hispanic, had not completed high school, had two parents at home, and had college-graduate parents, compared to earlier years (Table 1). These differences are largely attributable to the varying proportions of participants in each year accounted for by ASUTP patients, with the 2016 sample consisting nearly entirely of ASUTP patients. There were no differences across years in the proportion of participants residing in Massachusetts (data not shown).

Trends in cannabis use-related behaviors

Table 2 presents the comparison of cannabis-related behavioral profiles of participants across years, adjusting for clinic, gender, race, age, number of parents at home, and parent education. Frequency of cannabis use did not differ significantly across years, with roughly one in four participants reporting more than 60 days of use in the past 90 days across all years. The proportion of youth reporting younger initiation of cannabis use tended to be higher in 2015 and 2016 compared to 2013 and 2014 (>80% under age 16 vs. ∼75%), but the differences across years were not statistically significant. The proportion of participants reporting riding with a driver who had been using cannabis was significantly lower in 2015 compared to 2013 (70.6% vs. 93.6%), with an adjusted odds ratio of 0.20 (95% CI: 0.05, 0.80); differences from 2013 were not significant for 2014 or 2016. We found no significant differences across years in rates of driving after cannabis use, nor in rates of use before/during school or work.

Table 2

Change across years (2013–2016) in youth cannabis use-related behaviors: adjusted odds ratios comparing rates between 2013 and subsequent years.

	2013	2014		2015		2016
	n (%)	n (%)	aOR (95% CI)	n (%)	aOR (95% CI)	n (%)	aOR (95% CI)
Near-daily use in the past 90 days^a,b	11 (23.4)	9 (22.5)	0.78 (0.26, 2.33)	12 (25.5)	0.99 (0.35, 2.81)	13 (23.2)	0.98 (0.38, 2.58)
Age of initiation
Initiated any cannabis use at age <16 years	50 (75.8)	48 (72.7)	0.61 (0.22, 1.74)	62 (82.7)	1.15 (0.39, 3.41)	53 (86.9)	0.77 (0.23, 2.50)
Initiated weekly cannabis use at age <17 years	38 (62.3)	43 (72.9)	1.47 (0.49, 4.47)	55 (79.7)	2.65 (0.89, 7.89)	39 (70.9)	0.57 (0.20, 1.65)
Used before/during school or work in the past 90 days	46 (68.7)	46 (73.0)	2.07 (0.69, 6.21)	51 (72.9)	1.77 (0.63, 4.93)	51 (89.5)	2.00 (0.63, 6.39)
Drove after cannabis use	39 (58.2)	28 (42.4)	0.71 (0.31, 1.59)	35 (46.7)	0.64 (0.30, 1.38)	27 (43.5)	0.66 (0.30, 1.44)
Rode with driver who used cannabis^a	44 (93.6)	33 (78.6)	0.37 (0.08, 1.63)	36 (70.6)	0.20 (0.05, 0.80)*	48 (78.7)	0.32 (0.08, 1.26)

Notes.

p < .05.

Assessed in ASUTP sample only.

At least 61 days in the past 90 days.

Adjusted odds ratio (aOR) controls for sample characteristics that changed over time: clinic, race, age, number of parents, and parent education, plus gender, which did not change over time.

Trends in cannabis-related perceptions

Table 3 presents the comparison of cannabis-related attitudes across years, adjusting for demographic characteristics. Questions regarding attitudes and perceptions were only asked of ASUTP participants.

Table 3

Change across years (2013–2016) in youth cannabis-related attitudes: adjusted odds ratios comparing rates between 2013 and subsequent years.

	2013	2014		2015		2016
	n (%)	n (%)	aOR (95% CI)	n (%)	aOR (95% CI)	n (%)	aOR (95% CI)
Price
Increase	3 (6.7)	5 (12.5)	0.74 (0.11, 4.76)	2 (2.9)	0.25 (0.03, 2.04)	3 (5.3)	0.44 (0.07, 2.61)
No change	33 (73.3)	31 (77.5)	1.62 (0.51, 5.21)	43 (84.3)	2.36 (0.77, 7.20)	47 (82.5)	2.48 (0.86, 7.17)
Decrease	9 (20.0)	4 (10.0)	0.59 (0.15, 2.34)	6 (11.7)	0.62 (0.18, 2.12)	7 (12.3)	0.46 (0.14, 1.53)
Ease of obtaining
Increase	7 (15.6)	5 (12.2)	0.98 (0.26, 3.67)	15 (29.4)	1.82 (0.61, 5.45)	14 (24.1)	1.52 (0.52, 4.46)
No change	37 (82.2)	34 (82.9)	0.87 (0.25, 2.97)	36 (70.6)	0.65 (0.23, 1.90)	42 (72.4)	0.64 (0.23, 1.77)
Decrease	1 (2.2)	2 (4.9)	2.54 (0.12, 59.41)	0 (0.0)	n/a	2 (2.5)	1.67 (0.11, 25.04)
Potency
Increase	15 (34.1)	12 (31.6)	0.94 (0.34, 2.59)	16 (31.4)	1.02 (0.41, 2.56)	22 (38.6)	1.24 (9.52, 2.95)
No change	28 (63.6)	25 (65.8)	0.98 (0.36, 2.66)	34 (66.7)	0.99 (0.40, 2.46)	33 (57.9)	0.77 (0.33, 1.81)
Decrease	1 (2.3)	1 (2.6)	2.33 (0.11, 49.41)	1 (2.0)	0.83 (0.04, 17.44)	3 (3.5)	1.76 (0.14, 22.61)
Likelihood of getting caught
Increase	4 (9.1)	4 (9.8)	1.61 (0.32, 7.97)	6 (11.8)	1.35 (0.31, 5.85)	4 (6.9)	0.70 (0.15, 3.25)
No change	25 (56.8)	27 (65.9)	1.34 (0.50, 3.55)	29 (56.9)	1.21 (0.50, 2.93)	31 (53.4)	0.86 (0.37, 2.01)
Decrease	15 (34.1)	10 (24.4)	0.58 (0.20, 1.70)	16 (31.4)	0.69 (0.26, 1.78)	23 (39.7)	1.27 (0.53, 3.05)
Availability
Fairly or very easy	46 (97.9)	37 (90.2)	0.18 (0.02, 1.89)	51 (100.0)	n/a	58 (96.7)	0.77 (0.06, 9.41)
Risk of harm from occasional use
No or slight risk	42 (89.4)	38 (92.7)	1.68 (0.33, 8.69)	46 (90.2)	1.20 (0.30, 4.79)	57 (93.4)	1.86 (0.45, 7.79)
Risk of harm from regular use
No or slight risk	32 (68.1)	32 (78.1)	1.37 (0.47, 4.00)	40 (78.4)	1.46 (0.56, 3.96)	48 (80.0)	1.62 (0.63, 4.16)

Notes. Attitudes assessed in ASUTP sample only.

Adjusted odds ratio (aOR) controls for sample characteristics that changed over time: clinic, race, age, number of parents, and parent education, plus gender, which did not change over time.

Following the MML, the majority of respondents perceived no change in price, ease of obtaining cannabis, cannabis potency, and likelihood of getting caught, and we observed no significant differences in these attitudes across years. An overwhelming majority (97.9%) reported that cannabis was “fairly easy” or “very easy” to obtain, and that occasional cannabis use had “no” or “slight” risk of harm (89.4%) in 2013, with little change across years. Although the percentage of respondents viewing regular cannabis use as having “no” or “slight” risk of harm increased from 68.1% in 2013 to 80.0% in 2016, the difference was not statistically significant.

Trends in use of diverted medical cannabis

As shown in Figure 1, we observed a significant increase in the percentage of youth in this sample reporting using diverted medical cannabis, from 14.9% (n = 10) in 2013 to 44.3% (n = 27) in 2016, with an adjusted odds ratio (aOR) of 4.66 (95% CI: 1.81, 11.95). Nearly one-quarter reported ever obtaining diverted medical cannabis from sources registered in other states, which included nearby states such as RI (n = 8), ME (n = 5), NH (n = 4), and more distant states like CA (n = 6) and CO (n = 2). A modest proportion of youth using diverted medical cannabis identified multiple states (17.6%), suggesting access to multiple sources of medical cannabis. Sources of diverted medical cannabis tended to be peers, with 51.5% of respondents reporting their source to be age 21 years or younger. Nearly two-thirds (64.7%) of youth under the age of 18 acquired medical cannabis from a source age 21 or younger, while youth age 18 or older were more likely to have a medical cannabis source older than 21 (64.5%, p = .02).

Figure 1

Change across years (2013–2016) in rates of self-reported use of cannabis obtained from a medical cannabis cardholder.

Characteristics associated with diverted medical cannabis use

Youth reporting use of diverted medical cannabis were significantly more likely to be recruited from ASUTP, compared to those not reporting such use (83.8% vs. 70.4%; OR = 2.18, 95% CI: 1.07, 4.46); no other significant demographic differences were found (Table 4). There were no differences between groups in cannabis use frequency, age of initiation, and report of using before school or work. However, youth using diverted cannabis reported significantly higher rates of driving after cannabis use (67.7% vs. 40.7%; aOR: 2.97, 95% CI: 1.65, 5.33) and riding with a driver who had been using cannabis (93.0% vs. 74.6%; OR: 4.50, 95% CI: 1.52, 13.34), compared to youth not using diverted medical cannabis.

Table 4

Comparison of demographic and cannabis-related characteristics of adolescents reporting use of cannabis obtained from a medical cannabis cardholder versus those who did not.

	Use of cannabis obtained from a medical cannabis cardholder
	Non (%)	Yesn (%)	OR (95% CI)
Demographics
Total	199 (74.5)	68 (25.4)	–
Clinic
AMCS	59 (29.6)	11 (16.2)	Ref.
ASUTP	140 (70.4)	57 (83.8)	2.18 (1.07, 4.46)*
Age, under 18 years	90 (45.2)	37 (54.4)	1.45 (0.83, 2.51)
Female	68 (34.2)	18 (26.9)	0.71 (0.38, 1.31)
Race^a
White non-Hispanic	112 (56.3)	41 (61.2)	Ref.
Black, Hispanic, or Other	87 (43.7)	26 (38.8)	0.82 (0.46, 1.44)
Currently in school	149 (75.3)	47 (73.4)	0.91 (0.48, 1.73)
Completed high school or higher	95 (47.7)	49 (43.9)	0.86 (0.49, 1.50)
Two or more parents at home	115 (53.7)	39 (59.1)	1.02 (0.58, 1.80)
Parent(s) graduated college	122 (67.4)	44 (74.6)	1.42 (0.73, 2.75)
Cannabis-related characteristics
Near-daily use in the past 90 days^b,c	30 (22.7)	13 (25.0)	1.13 (0.54, 2.40)
Age of initiation^d
Initiated any cannabis use at age <16 years	150 (76.9)	58 (86.6)	1.59 (0.71, 3.57)
Initiated weekly cannabis use at age <17 years	126 (70.4)	45 (76.3)	1.17 (0.58, 2.36)
Used cannabis before/during school or work in the past 90 days^d	136 (72.0)	53 (85.5)	1.73 (0.72, 4.19)
Drove after cannabis use^d	81 (40.7)	46 (67.7)	2.97 (1.65, 5.33)***
Rode with driver who used cannabis^d	103 (74.6)	53 (93.0)	4.50 (1.52, 13.34)**

Notes. OR = odds ratio.

p < .05.

p < .01.

***

p < .001.

Categories collapsed for analysis.

At least 61 days in the past 90 days.

Assessed in the ASUTP sample only, and odds ratios are adjusted for clinic.

Assessed in both clinic samples, and odds ratios are adjusted for clinic.

Discussion

Given the growing trend of medical cannabis legalization in the U.S., it is critical to understand how cannabis policy affects youth, particularly cannabis-using youth, who may be more vulnerable to the effects of liberalized cannabis legislation. This study adds to emerging evidence that state MML enactment may be associated with subsequent changes in youth cannabis-using behavior. 26 Three years following the 2013 passage of Massachusetts’ MML, we observed a significantly increased rate of diverted medical cannabis use in this sample of cannabis-using youth. These results are consistent with two studies with similar populations in Colorado, which found substantial rates of medical cannabis diversion reported by adolescent outpatients of a substance use treatment center (49% in one study, and 74% in the other). 27 ^, 28

The time lag between MML passage and increased youth use of diverted medical cannabis in this study is in keeping with findings from a 2019 study examining the effects of cannabis policy changes on youth in 38 countries. This study found that the association between cannabis policy liberalization and greater regular cannabis use was only significant after the policy had been in effect for more than 5 years. 29 Although some components of the MML were initiated immediately upon its passage, such as permitting home growth and cultivation in limited quantities, the delayed effect we observed may be due to the time it takes for a law to become fully implemented, including the opening of dispensaries. Our observation of a significant increase in the rate of diverted medical cannabis use only in 2016, compared to 2013, is understandable in light of statewide data for the number of medical cannabis cardholders in Massachusetts during the same period. Cardholder numbers were close to zero in 2013 and 2014, increasing to 1,423 in 2015, and leaping to 18,476 in 2016. 30 No dispensaries were open until mid-2015, and by 2016, four were open for business. As seen in Massachusetts, MML implementation can take some time and thus the effects on youth may not be evident for several years.

Given the expansion of Massachusetts’ medical cannabis landscape by 2016, and our majority-Massachusetts resident sample, we expected that youth would report obtaining diverted medical cannabis from a Massachusetts source. However, it is notable that nearly one in four obtained medical cannabis through non-Massachusetts sources, including those in non-bordering states, reflecting diffusion or “contamination” from other states with more established cannabis policies. 31 ^, 32 Moreover, given the lack of a minimum legal age for becoming a registered cardholder in Massachusetts, it is not surprising that participants, particularly those under age 18, were most likely to obtain diverted medical cannabis from another young person. Recruitment site being the sole demographic predictor of youth reporting diverted medical cannabis use may be explained by underlying differences between clinic samples; for example, the relatively higher-socioeconomic ASUTP participants may be likelier to afford medical cannabis, and drive cars as their method of transportation.

Of particular concern is our finding that youth using diverted medical cannabis showed significantly heightened risks for driving after using cannabis, or riding with a driver who had been using. There is evidence that cannabis-using youth may not perceive cannabis as impairing; some youth believe cannabis makes them drive better or intentionally use while driving (e.g., for “blunt rides”). 33 – 35 The medicalization of cannabis may also contribute to lower perceived driving risks if youth regard cannabis as a “medication” rather than a “drug.” 34 – 36 The risks of injury and mortality from driving under the influence cannot be overstated, especially among young people, for whom motor vehicle crashes are the leading cause of death. 37 Cannabis use has been found to be associated with a doubling of the risk of motor vehicle crashes. 38

Finally, in our sample of youth identified as having problematic substance use, we found little change over time in perceived risk of harm or perceived MML effects on ease of access, potency, price, or legal repercussions following MML passage. Although previous studies have found changes in cannabis-related attitudes among general youth populations after MML passage, 39 the lack of changes found in our study may be due to our sample consisting of cannabis-using youth who already had access to cannabis and a low perceived risk of harm from use. Participants reported very easy access and very low perceived risk of harm across all years, thus there was little room for legalization to ease these perceptions. The small sample may also have limited our ability to detect differences.

Strengths of this study include its inclusion of four years of data post-MML, allowing the detection of more delayed policy effects on youth, and its focus on the higher-risk population of cannabis-using youth, who may be more vulnerable to the effects of liberalized access to cannabis, compared to non-using youth. As such, this paper is a valuable addition to the existing body of literature on MML effects on youth, which consists primarily of studies on general youth populations and includes shorter study time windows after MML passage. 8 ^, 40 – 42

This study has several potential limitations. The sample was small and limited to youth recruited from Massachusetts clinical settings; thus, findings may not be generalizable to youth in other regions or settings. Youth presenting to a substance use treatment clinic (voluntarily and/or at the behest of parents/guardians) and those agreeing to participate in a cannabis intervention study may differ from other clinical adolescent populations. Survey response rates were 50% and 59% between the two recruitment sites, which may have resulted in non-response bias. Our data relied on self-report, which may be subject to recall error and social desirability bias contributing to underreporting of sensitive behaviors such as driving under the influence. As previously noted, youth in our sample already had access to cannabis and few concerns about using it, so there may have been a ceiling effect limiting our ability to detect the MML's impact on change in attitudes and perceptions. Because of multiple comparisons, our data may be subject to Type II error. Given its repeated cross-sectional design, causality cannot be inferred. ASUTP participants were asked about “smoking” cannabis, which may have missed the growing use of edibles and THC vaping. 43 Finally, our study did not assess participants’ residence proximity to dispensaries nor to surrounding states. Their exposures to medical cannabis policies of other states that had MMLs much longer than Massachusetts’ (e.g., Maine enacted its MML in 1999) may have already influenced the views and behavior of youth living out of state but nearby. 31 ^, 32

Although Massachusetts now permits both medical and recreational cannabis for adult use, there are still many states that have chosen to enact, or are considering enacting, laws permitting medical cannabis use only. Even in states where neither is the case, our findings make evident that medical cannabis is diverted across state lines. With the increasing number of states decriminalizing or legalizing medical (and recreational) cannabis, there is a critical need for further research to elucidate the effects of cannabis policy on youth. Our findings point to a need for healthcare providers, inside and outside of “MML states,” to assess use of diverted medical cannabis when youth report cannabis use. Providers should counsel youth about the well-known risks of cannabis —including neurocognitive harm and public safety risks 26 —regardless of whether it is labeled “medical.” Finally, providers should be especially cognizant of screening youth who use diverted medical cannabis for riding and driving under the influence of cannabis. As liberalized cannabis policies become more prevalent, we must continue to monitor the effects of cannabis policy on adolescents.

Footnotes

Acknowledgements

The authors would like to acknowledge the patients who provided data and made this study possible. We extend our thanks to Julie Lunstead and the rest of the adolescent substance use treatment program team, as well as Bea Duvert and the rest of the adolescent medicine clinic study team.

Author contributions

MO conducted data analysis, and with SKH, led the writing of this manuscript. SL and LS provided critical feedback and contributions to the manuscript. SL and LS directed participant enrollment and data collection.

Disclosure statement

None of the authors report a conflict of interest. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of Boston Children's Hospital, Harvard Medical School, or any of its affiliates.

References

National Institute on Drug Abuse. Monitoring the Future 2017 Survey Results. 2020. https://www.drugabuse.gov/related-topics/trends-statistics/infographics/monitoring-future-2017-survey-results. Published December 12, 2017. Accessed April 21, 2020.

Substance Abuse and Mental Health Services Administration, Office of Applied Studies. Treatment Episode Data Set (TEDS) Highlights – 2007; 2009. https://wwwdasis.samhsa.gov/dasis2/teds_pubs/2007_teds_highlights_rpt.pdf. Accessed March 3, 2020.

Marzell

, Sahker

, Arndt

Trends of Youth Marijuana Treatment Admissions: increasing admissions contrasted with decreasing drug involvement. Subst Use Misuse. 2017;52(13):1778–1783.

National Institute on Drug Abuse. What are marijuana's long-term effects on the brain? 2020. https://www.drugabuse.gov/publications/research-reports/marijuana/what-are-marijuanas-long-term-effects-brain. Accessed March 2, 2020.

National Institute on Drug Abuse. Is there a link between marijuana use and psychiatric disorders? 2020. https://www.drugabuse.gov/publications/research-reports/marijuana/there-link-between-marijuana-use-psychiatric-disorders. Accessed March 2, 2020.

Hartman

, Huestis

MA.

Cannabis effects on driving skills. Clin Chem. 2013;59(3):478–492.

Rogeberg

, Elvik

The effects of cannabis intoxication on motor vehicle collision revisited and revised. Addiction. 2016;111(8):1348–1359.

Blevins

, Marsh

, Banes

, Stephens

, Walker

, Roffman

RA.

The implications of cannabis policy changes in Washington on adolescent perception of risk, norms, attitudes, and substance use. Subst Abuse Res Treat. 2018;12:1178221818815491.

Frohe

, Leeman

, Patock-Peckham

, Ecker

, Kraus

, Foster

DW.

Correlates of cannabis vape-pen use and knowledge among U.S. college students. Addict Behav Rep. 2018;7:32–39.

10.

Parker

, Anthony

JC.

Population-level predictions from cannabis risk perceptions to active cannabis use prevalence in the United States, 1991–2014. Addict Behav. 2018;82:101–104.

11.

National Institute on Drug Abuse. What is the scope of marijuana use in the United States? 2020. https://www.drugabuse.gov/publications/research-reports/marijuana/what-scope-marijuana-use-in-united-states. Accessed March 2, 2020.

12.

Chadi

, Hadland

SE.

Adolescents and perceived riskiness of marijuana: Why care?

J Adolesc Health. 2018;63(4):377–378.

13.

Capriotti

Medical Marijuana. Home Healthc Now. 2016;34(1):10–15.

14.

National Conference of State Legislatures. State Medical Marijuana Laws. https://www.ncsl.org/research/health/state-medical-marijuana-laws.aspx. Published 2019. Accessed March 3, 2020.

15.

Sacramento County Public Law Library. Medical Marijuana Laws in California. https://saclaw.org/articles/marijuana-laws-in-california-edl/. Published 2019. Accessed March 3, 2020.

16.

Grigsby

, Hoffmann

, Moss

MJ.

Marijuana use and potential implications of Marijuana Legalization. Pediatr Rev. 2020;41(2):61–72.

17.

American Academy of Child and Adolescent Psychiatry. Medical Marijuana. https://www.aacap.org/aacap/Policy_Statements/2012/AACAP_Medical_Marijuana_Policy_Statement.aspx. Accessed March 2, 2020.

18.

Altieri

Massachusetts Becomes 18th State to Legalize Medical Marijuana. NORML Blog, Marijuana Law Reform. https://blog.norml.org/2012/11/06/massachusetts-becomes-18th-state-to-legalize-medical-marijuana/. Published November 7, 2012. Accessed March 2, 2020.

19.

Cannabis Control Commission: Commonwealth of Massachusetts. Medical use of Marijuana: an act to ensure safe access to Marijuana. 2017.

20.

Cannabis Control Commission: Commonwealth of Massachusetts. Adult Use of Marijuana; 2019.

21.

Shrier

, Burke

, Kells

Pilot randomized trial of MOMENT, a motivational counseling-plus-ecological momentary intervention to reduce marijuana use in youth. Mhealth. 2018;4(7):29–29.

22.

Richmond-Rakerd

, Slutske

, Wood

PK.

Age of initiation and substance use progression: a multivariate latent growth analysis. Psychol Addict Behav J Soc Psychol Addict Behav. 2017;31(6):664–675.

23.

Sagar

, Dahlgren

, Gönenç

, Racine

, Dreman

, Gruber

SA.

The impact of initiation: early onset marijuana smokers demonstrate altered Stroop performance and brain activation. Dev Cogn Neurosci. 2015;16:84–92.

24.

Shrier

, Rhoads

, Burke

, Walls

, Blood

EA.

Real-time, contextual intervention using mobile technology to reduce marijuana use among youth: a pilot study. Addict Behav. 2014;39(1):173–180.

25.

National Institute on Drug Abuse. Monitoring the Future Survey: High School and Youth Trends DrugFacts. National Institute on Drug Abuse. https://www.drugabuse.gov/publications/drugfacts/monitoring-future-survey-high-school-youth-trends. Published December 18, 2019. Accessed September 16, 2020.

26.

Chadi

, Weitzman

, Levy

Understanding the impact of National and State Medical Marijuana policies on adolescents. Curr Addict Rep. 2018;5(2):93–101.

27.

Salomonsen-Sautel

, Sakai

, Thurstone

, Corley

, Hopfer

Medical marijuana use among adolescents in substance abuse treatment. J Am Acad Child Adolesc Psychiatry. 2012;51(7):694–702.

28.

Thurstone

, Lieberman

, Schmiege

SJ.

Medical marijuana diversion and associated problems in adolescent substance treatment. Drug Alcohol Depend. 2011;118(2–3):489–492.

29.

Shi

, Lenzi

, An

Cannabis Liberalization and Adolescent Cannabis use: a cross-national study in 38 Countries. Doran CM, ed. PLoS One. 2015;10(11):e0143562.

30.

Massachusetts Department of Public Health. Medical Use of Marijuana Program monthly dashboards. Mass.gov. https://www.mass.gov/lists/medical-use-of-marijuana-program-monthly-dashboards. Accessed August 20, 2020.

31.

Chen

. Information diffusion in the evaluation of Medical Marijuana Laws’ impact on risk perception and use. Am J Public Health. 2016;106(12):e8.

32.

Chen

, Yu

, Stanton

, Cook

, Chen

D-G

, Okafor

Medical Marijuana Laws and Marijuana use among U.S. adolescents: evidence from Michigan Youth Risk Behavior Surveillance Data. J Drug Educ. 2018;48(1–2):18–35.

33.

Mehta

, Burke

, Shrier

LA.

Contexts of Marijuana use and cessation among frequently-using youth in primary care. Youth Soc. 2020;2020:0044118X20951060.

34.

Arkell

, Lintzeris

, Mills

, Suraev

, Arnold

, McGregor

IS.

Driving-related behaviours, attitudes and perceptions among Australian medical cannabis users: results from the CAMS 18–19 survey. Accid Anal Prev. 2020;148:105784

35.

Wadsworth

, Hammond

International differences in patterns of cannabis use among youth: prevalence, perceptions of harm, and driving under the influence in Canada, England & United States. Addict Behav. 2019;90:171–175.

36.

Keyes

, Wall

, Cerdá

How does state marijuana policy affect U.S. youth? Medical marijuana laws, marijuana use and perceived harmfulness: 1991–2014. Addict Abingdon Engl. 2016;111(12):2187–2195.

37.

Centers for Disease Control and Prevention. Teen Drivers: Get the Facts. https://www.cdc.gov/motorvehiclesafety/teen_drivers/teendrivers_factsheet.html. Published October 31, 2019. Accessed September 21, 2020.

38.

M-C

, Brady

, DiMaggio

, Lusardi

, Tzong

, Li

Marijuana use and motor vehicle crashes. Epidemiol Rev. 2012;34:65–72.

39.

Wen

, Hockenberry

, Druss

BG.

The effect of Medical Marijuana Laws on Marijuana-related attitude and perception among US adolescents and young adults. Prev Sci. 2019;20(2):215–223.

40.

Anderson

, Hansen

, Rees

DI.

Medical Marijuana laws and Teen Marijuana use. Am Law Econ Rev, 2015;17(2):495–528.

41.

Cerdá

, Mauro

, Hamilton

Association between recreational Marijuana Legalization in the United States and changes in Marijuana use and Cannabis use disorder from 2008 to 2016. JAMA Psychiatry. 2020;77(2):165–171.

42.

Plunk

, Agrawal

, Harrell

The impact of adolescent exposure to medical marijuana laws on high school completion, college enrollment and college degree completion. Drug Alcohol Depend. 2016;168:320–327.

43.

Patrick

, Miech

, Kloska

, Wagner

, Johnston

LD.

Trends in Marijuana vaping and edible consumption from 2015 to 2018 among adolescents in the US. JAMA Pediatr. 2020;174(9):900–902.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB