Associations of Healthcare Service Utilization With Cannabis Use Status,Use Reasons,and Use Characteristics Among Those Age 50 and Older

Abstract

Cannabis use has increased steadily among older adults, and they are a significant proportion of medical cannabis users. Using 2015–2019 National Survey on Drug Use and Health data (n = 44,007 age 50+), we examined whether the numbers of emergency department (ED) visits and nights hospitalized are associated with cannabis use status, use reason (nonmedical-only, medical-only, and medical and nonmedical), and use characteristics. Past-year users had higher rates of any ED visit (30.0%) and hospitalization (14.7%) than prior-to-past-year users and never users. However, negative binomial regression models showed that past-year users did not differ from never users on numbers of ED visits and nights hospitalized, although they had more ED visits than prior-to-past-year users (IRR = 1.21, 95% CI = 1.10–1.34). Medical-only users had more ED visits (IRR = 1.38, 95% CI = 1.02–1.88) than nonmedical-only users. Cannabis use and use characteristics were not associated with nights hospitalized. The study findings provide insights into older cannabis users’ healthcare utilization.

Keywords

cannabis marijuana: ED visit hospitalization cannabis use disorder mental disorder

Introduction

Individuals age 50 and older are the fastest growing group of cannabis users due in part to the increasing numbers of states legalizing or decriminalizing cannabis for medical and/or nonmedical use, decreasing cannabis risk perceptions, and the aging of the baby boom generation (Kaskie et al., 2017; Lloyd & Striley, 2018). Individuals in this age group appear to be especially attracted to cannabis’ purported therapeutic effects as one-third to more than half of medical cannabis registrants in medical cannabis legal (MCL) states are age 50+ (Fairman, 2016, Brown et al., 2020). Others are self-treating as they report medical use but lack medical authorization, and most report that cannabis has beneficial effects on their physical and psychiatric symptoms (Reynolds et al., 2018; Yang et al., 2021). State medical cannabis laws are associated with significant reductions in opioid prescribing in the Medicare Part D population and lower pain and better self-assessed health and increased labor supply among those age 51+ (Bradford et al., 2018; Nicholas & Maclean, 2019).

Research shows that cannabis is somewhat effective for treating different types of pain, chemotherapy-induced nausea and vomiting, multiple sclerosis spasticity symptoms, and sleep disturbances associated with some chronic conditions (Johal et al., 2020; National Academies, 2017; Rabgay et al., 2020). However, cannabis use has also been associated with both acute and long-term adverse physical, psychiatric, and cognitive health effects and physical injury, especially with high-potency Δ⁹-tetrahydrocannabidiol (THC), heavy use, hyperemesis, and interactions with medications, alcohol, and illicit substances (Bahorik et al., 2018; Balachandran et al., 2021; Banerjee et al., 2020; Broyd et al., 2016; Campeny et al., 2020; Damkier et al., 2019; Hasin & Walsh, 2020; Lev-Ran et al., 2014; Matheson & Le Foll, 2020; Porr et al., 2020). Although overall harms from cannabis tend to be lower than harms from alcohol, tobacco, and other illicit drugs (Bonomo et al., 2019), research shows that cannabis use is associated with disproportionately increased physical and mental health-related emergency department (ED) visits and hospitalizations among both adolescents and adults (Hall et al., 2018; Hall & Lynskey, 2020; Perisetti et al., 2020; Roberts, 2019; Wang et al., 2018). A study of Colorado’s statewide ED visit data found that primary diagnoses of mental disorders (e.g., mood and anxiety disorders, schizophrenia and other psychotic disorders, and suicide/intentional self-injury) comprised 31.0% of ED visits with cannabis-associated diagnostic codes among adults (Hall et al., 2018). Data on hospitalizations among cannabis users also indicate a significant healthcare burden and increased odds of in-hospital mortality (Desai et al., 2017).

Compared to those age 50+ who do not use cannabis, past-year users have two-to-three times higher rates of mental and substance use disorders (Vacaflor et al., 2020), nearly ten times greater odds of reporting opioid dependence, and sixfold greater odds of past-year nonmedical opioid use, especially heroin (Ramadan et al., 2021). As expected, older cannabis users who misuse opioids have been found to have more chronic medical conditions in addition to a higher rate of major depressive episode (Choi et al., 2021). While cross-sectional studies cannot establish the time order of the associations between cannabis use and behavioral health problems, many long-term cannabis users likely have long-standing behavioral health problems given cannabis’ long-as well as short-term effects on users’ cognition, motivation, and psychosis (Volkow et al., 2016).

A study based on 2012-2013 National Epidemiologic Survey on Alcohol and Related Problems data also found that compared to nonusers, past-year cannabis users age 50+ had higher odds of ED visits through increased odds of injury (Choi et al., 2018). Other than this study, research on the associations between cannabis use and ED visits and hospitalizations among older adults is scant. Given increasing nonmedical and medical cannabis use in the 50+ age group, more research on the associations of healthcare utilization with cannabis use status, use reasons, and use characteristics is needed. In the present study, we examined associations of numbers of ED visits and inpatient hospital nights with: (1) cannabis use status (never use, prior-to-past-year use, and past-year use) in the 50+ age group; and (2) cannabis use reason (nonmedical-only, medical-only, and both medical and nonmedical), cannabis initiation age, cannabis use frequency, and cannabis use disorder (CUD) among past-year cannabis users. Andersen’s Behavioral Model of Health Services Use (Andersen & Newmann, 1973; Andersen, 1995) posits that healthcare service use is determined by predisposing factors (individual biological and demographic and social-structural characteristics and health beliefs), enabling resources (that facilitate access to care), and need (perceived/evaluated health problem severity). Of these, need factors tend to have the greatest influence, especially among medically vulnerable population groups (e.g., older adults and poor/uninsured individuals) (Babitsch et al., 2012). We posited that cannabis use, medical use in particular, early-onset use, frequent use, and CUD are indications of older adults’ overall poorer health, which contributes to their healthcare use. Study hypotheses were: controlling for predisposing, enabling, and physical and behavioral health need factors, past-year cannabis use, compared to never use and prior-to-past-year use, will be associated with greater numbers of (H1a) ED visits and (H1b) nights hospitalized, (H2a) any medical use compared to nonmedical use only, (H2b) cannabis initiation before age 21, and (H2c) high frequency use, and (H2d) CUD will be associated with greater numbers of ED visits and nights hospitalized. Study findings will aid in increasing knowledge of healthcare service use among the growing numbers of older cannabis users.

Methods

Data and Sample

Data came from the public use files of the 2015-2019 National Survey on Drug Use and Health (NSDUH), the largest, annual population-based survey measuring the prevalence of substance use, mental and substance use disorders, and behavioral health treatment among the U.S. civilian, non-institutionalized, population aged 12+. The survey also includes questions about physical and functional health and healthcare use. To ensure respondents’ privacy and confidentiality in responding to questions and to increase honest reporting of sensitive behaviors such as illicit drug use, respondents were interviewed in private at their residence using computer-assisted and audio computer-assisted self-interviewing and computer-assisted personal interviewing. The number of respondents completing the survey was 57,146 in 2015, 56,897 in 2016, 56,278 in 2017, 56,313 in 2018, and 56,136 in 2019, with a total of 44,007 respondents age 50+ during the five years. For a detailed description of the NSDUH’s multi-stage stratified sampling design, see the Center for Behavioral Health Statistics and Quality (2020).

Measures

Past-year healthcare use: NSDUH respondents reported the numbers of times in the past 12 months they were treated at an ED and the number of nights hospitalized for any reason (mental or physical). Responses were coded as 0 to 31 times, with 31 representing 31 or more visits. To measure the rates of any ED visit and hospitalization for descriptive purposes, we created dichotomous variables (0 = no ED visit or hospitalization and 1 = one or more ED visits or hospitalizations). To describe the sample, we also report any outpatient visit (about their own health) to a doctor, nurse, physician assistant or nurse practitioner at a doctor’s office, clinic, or other place (0 = no visit and 1 = one or more visits).

Past-year cannabis use status, use reasons, and use characteristics: Cannabis use was assessed with the question, “have you ever, even once, used?”. Those who answered yes were asked about the age of their first and last use and whether they used in the past year and the past month. Based on responses to these questions, we identified never use, prior-to-past-year use (i.e., used in the past but not during the past year), and past-year use. Past-year users were also asked whether a healthcare professional recommended any or all of their past-year use (regardless of the cannabis laws in the state where they lived). Answers to this question were used to categorize use reasons as nonmedical-only, medical-only, or both medical and nonmedical (hereinafter medical/nonmedical). Past-year users’ use frequency was determined by the number of days they used. Those who met criteria for either past-year cannabis abuse or dependence based on the DSM-IV (American Psychiatric Association, 1994) were classified as having CUD.

Predisposing and enabling factors: Predisposing factors were age (50–64 years and 65+ years [to protect respondents’ anonymity, NSDUH’s public-use data files do not provide chronological age]), gender, race/ethnicity, and residence in a MCL state. Enabling factors were marital status, education (bachelor’s degree vs. no degree), employment status (full- or part-time work vs. not working), family income (living in poverty, up to 2 times poverty, and more than 2 times poverty), and health insurance (0 = no and 1 = yes).

Physical and behavioral health need factors: Physical health conditions were: (1) the number (0-10) of chronic medical conditions (hypertension, heart disease, diabetes, asthma, chronic bronchitis or COPD, kidney disease, HIV/AIDS, hepatitis B or C, cirrhosis of the liver, and cancer) ever diagnosed by a healthcare professional; (2) functional impairment: serious difficulty in concentration, memory, and decision making; walking or climbing stairs; dressing or bathing; and doing errands alone (0 = no, 1 = yes for each; total score = 0–4, Cronbach’s α = 0.68 for the study sample); and (3) sensory impairment: serious difficulty in hearing or seeing (0 = no hearing or vision difficulty; 1=either hearing or vision difficulty; and 2 = both hearing and vision difficulty).

Behavioral health conditions included mental illness and substance use/use disorders. Mental illness in the NSDUH was estimated based on predictive probability models using K-6 psychological distress scale scores (and classified as no, mild, moderate, and serious) (Kessler et al., 2003), serious thoughts of suicide, major depressive episode, and the World Health Organization Disability Assessment Schedule (WHODAS) score (Novak et al., 2010). The WHODAS was used to measure the level of difficulty respondents experienced in doing eight daily activities in the one month during the past year when they were at their worst emotionally (CBSHQ, 2020). Serious mental illness was defined as any mental, behavioral, or emotional disorder (excluding developmental and substance use disorders) that substantially interfered with or limited one or more major life activities (Substance Abuse and Mental Health Services Administration, 2020). Mental health treatment in the past year included pharmacotherapy, outpatient treatment/counseling, and/or inpatient treatment/counseling.

Substance use/use disorder measures included: (1) alcohol use disorder (per DSM-IV alcohol abuse or dependence criteria); (2) nicotine dependence (based on Fagerström test scores [Heatherton et al., 1991]); and (3) illicit drug use other than cannabis (e.g., cocaine, crack, heroin, LSD, PCP, or misuse of prescription pain relievers and other prescription psychotherapeutics [i.e., tranquilizers, stimulants, and sedatives]).

Analysis

Following NSDUH guidelines, we created adjusted person-level analysis weights by dividing the final person-level analysis weights by the number of years of pooled data (i.e., five) used in the present study. In all analyses, we used Stata/MP 17’s svy function (College Station, TX) and subpop command to account for NSDUH’s multi-stage, stratified sampling design and to ensure that variance estimates incorporate the full sampling design. All estimates presented in this study are weighted except sample sizes. We used Pearson’s χ² tests and one-way ANOVA to describe the distribution of the study variables among (1) never, prior-to-past-year, and past-year cannabis users, and (2) nonmedical-only, medical/nonmedical, and medical-only users. We also used χ² tests to examine differences between groups (i.e., prior-to-past-year vs. past-year users; medical/nonmedical vs. medical-only users). We fit two negative binomial regression models to test H1a (associations of the numbers of ED visits with cannabis use status). The first model compared prior-to-past-year and past-year use to never use, and the second model compared past-year use to prior-to-past-year use). We fit a negative binomial regression model to test H1b (association of the numbers of nights hospitalized with cannabis use status). We fit two negative binomial regression models to test H2a-d (associations of the numbers of ED visits and nights hospitalized with cannabis use reasons and use characteristics among users). All negative binomial regression models excluded cases with missing data on the outcome variables (≤2.0%). Following Gelman and Hill’s (2007) recommendations, we included a variety of covariates to adjust for missing data under the missing-at-random assumption. Negative binomial regression model results are presented as adjusted incident rate ratios (IRR) with 95% confidence intervals (CI). Survey year (2015-2019) was not included in the final models because it was not significant in preliminary analyses. Significance was set at p<.05.

Results

Characteristics of Never Users, Prior-To-Past-Year Users, and Past-Year Users

Table 1 shows that 57.7%, 34.8%, and 7.5% of those age 50+ reported never, prior-to-past-year, and past-year cannabis use, respectively. Past-year cannabis use increased from 6.3% in 2015 to 9.6% in 2019. Compared to never users and prior-to-past-year users, past-year users included significantly larger proportions of the 50-64 age group, men, non-Hispanic Blacks, residents of MCL states, and those living in poverty, but significantly smaller proportions of married individuals and those who had any health insurance.

Table 1.

Characteristics of Individuals Age 50+ by Cannabis Use Status.

	Never user24,988 (57.68%)	Prior-to-past-year user15,551 (34.80%)	Past-year user3468 (7.52%)	P (all three groups)	P (excluding never users)
Year (%)				<.001	<.001
2015	19.87	19.46	16.34
2016	20.39	19.15	16.77
2017	19.96	20.27	19.42
2018	19.85	20.70	21.29
2019	19.93	20.41	26.18
Age group (%)				<.001	<.001
50−64	43.99	70.10	77.85
65+	56.01	29.90	22.15
Female (%)	58.32	47.90	39.13	<.001	<.001
Race/ethnicity (%)				<.001	.001
Non-Hispanic White	66.83	80.64	76.58
Non-Hispanic Black	10.81	9.56	12.19
Hispanic	14.16	5.86	6.71
Other	8.20	3.94	4.52
Married (%)	62.54	62.50	48.31	<.001	<.001
Bachelor’s degree (%)	29.47	36.50	31.12	<.001	<.001
Working full- or part-time (%)	39.85	58.04	54.58	<.001	.006
Income (%)				<.001	<.001
Living in poverty	10.80	7.61	14.95
Up to 2X poverty	21.79	14.25	18.08
>2X poverty	67.41	78.15	66.96
Have health insurance (%)	95.26	95.40	91.95	<.001	<.001
Resident of medical cannabis legal state (%)	56.58	61.64	68.06	<.001	<.001
No. of chronic medical conditions (0−10), M (SE)^a	1.05 (0.01)	1.01 (0.01)	0.92 (0.02)	<.001	.001
No. of functional impairments (0−4), M (SE)^b	0.35 (0.01)	0.30 (0.01)	0.48 (0.02)	<.001	<.001
No. of sensory impairments (0−2), M (SE)^c	0.17 (0.00)	0.13 (0.00)	0.16 (0.01)	<.001	.001
Mental illness (%)				<.001	<.001
None	89.29	83.10	72.36
Mild	6.61	8.65	12.52
Moderate	2.49	4.39	8.12
Serious	1.61	3.86	7.0
Any past-year mental health treatment (%)	10.46	20.19	28.21	<.001	<.001
First used cannabis <age 21 (%)		71.23	81.25	n/a	<.001
Alcohol use disorder (%)	1.32	4.83	12.19	<.001	<.001
Nicotine dependence (%)	4.66	11.49	24.19	<.001	<.001
Any illicit drug (other than cannabis) use (%)	1.77	5.37	19.88	<.001	<.001
Any past-year outpatient healthcare visit (%)				<.001	<.001
No	9.64	9.46	12.66
Yes	87.33	88.86	85.20
Don’t know/refused	3.03	1.67	2.15
Any past-year emergency department (ED) visit (%)				<.001	<.001
No	72.16	74.87	68.54
Yes	25.82	24.21	30.0
Don’t know/refused	2.02	0.92	1.47
No. of ED visits, M (SE) (n = 43,284)^d	0.51 (0.01)	0.42 (0.01)	0.63 (0.03)	<.001	<.001
No. of ED visits among those with any ED visit, M (SE) (n= 11,505)^e	1.94 (0.04)	1.71 (0.03)	2.06 (0.08)	<.001	<.001
Any past-year hospitalization (%)				<.001	<.001
No	86.23	87.83	85.02
Yes	13.45	12.02	14.66
Do not know/refused	0.32	0.16	0.31
No. of nights hospitalized, M (SE) (n = 43,785)^f	0.68 (0.02)	0.59 (0.02)	0.81 (0.06)	<.001	.001
No. of nights hospitalized among those with any hospitalization, M (SE) (n = 11,505)	5.11 (0.12)	4.92 (0.16)	5.59 (0.30)	.170	.061

Note. For categorical variables, probability values were calculated using χ². For continuous variables^a-f , probability values were calculated using ANOVA (with Bonferroni corrections) for comparisons of all three groups (never users, prior-to-past-year users, and past-year users) and t-tests for comparisons between prior-to-past-year users and past-year users.

Past-year users had fewer chronic medical conditions but more functional impairments than never users and prior-to-past-year users. Past-year users included the largest proportions of those with any and serious past-year mental disorders. More past-year users reported receiving past-year mental health treatment, and they had the highest rates of past-year alcohol use disorder, nicotine dependence, and any illicit drug use.

Past-year users had the lowest proportion of those with any outpatient care visit but the highest proportions of those with any ED visit (30.0%) and inpatient hospitalization (14.7%). Of those who had any ED visit, prior-to-past-year users had fewer visits than never users or past-year users.

Characteristics of Past-Year Nonmedical-Only, Medical-Only, and Medical/Nonmedical Users

Table 2 shows that of past-year users, 83.4% reported nonmedical use only, 12.1% medical use only, and 4.5% medical/nonmedical use, without significant change in these rates during the five survey years (F_{(5.42, 271.25)} = 1.18, p = .319). The three groups did not differ significantly on age, gender, marital status, education, income, and health insurance. However, compared to nonmedical-only users, medical-only users included a significantly lower proportion of non-Hispanic Whites but the highest proportion of MCL state residents, and medical/nonmedical users included a significantly higher proportion of non-Hispanic Whites.

Table 2.

Characteristics of Past-Year Cannabis Users by Cannabis Use Reason.

	Nonmedical use only2886(83.44%)	Medical use only431 (12.05%)	Medical/nonmedical use151 (4.52%)	P (all three groups)	P (excluding nonmedical users)
Year (%)				.319	.323
2015	16.47	15.08	12.27
2016	16.70	19.72	10.14
2017	20.12	15.19	17.76
2018	21.21	21.67	21.68
2019	25.49	28.35	33.16
Age group (%)				.258	.982
50−64	78.59	73.13	74.06
65+	21.41	25.87	25.94
Female (%)	38.56	45.78	32.06	.057	.016
Race/ethnicity (%)				.005	.015
Non-Hispanic White	76.60	72.45	87.30
Non-Hispanic Black	12.86	9.32	7.51
Hispanic	6.28	10.85	3.48
Other	4.26	7.38	1.72
Resident of medical cannabis legal state (%)	63.85	92.18	81.53	<.001	.001
Married (%)	48.56	48.28	43.75	.636	.490
Bachelor’s degree (%)	30.53	32.40	38.49	.199	.250
Working full- or part-time (%)	57.13	38.54	50.35	<.001	.041
Income (%)				.636	.995
Living in poverty	15.03	14.49	14.83
Up to 2X poverty	17.47	21.13	21.21
>2X poverty	67.50	64.38	63.97
Have health insurance (%)	91.40	94.51	95.19	.165	.815
No. of chronic medical conditions (0–10), M (SE)^a	0.88 (0.02)	1.10 (0.07)	1.23 (0.12)	.001	.262
No. of functional impairments (0−4), M (SE)^b	0.40 (0.02)	0.84 (0.08)	0.82 (1.00)	<.001	.922
No. of sensory impairments (0−2), M (SE)	0.15 (0.01)	0.18 (0.03)	0.25 (0.04)	.062	.151
Mental illness (%)				<.001	.387
None	73.92	64.31	65.01
Mild	12.44	13.40	11.61
Moderate	7.48	10.03	14.95
Serious	6.16	12.25	8.43
Any past-year mental health treatment (%)	25.73	40.03	42.41	<.001	.670
Alcohol use disorder (%)	13.04	7.27	9.56	.019	.370
Nicotine dependence (%)	24.87	19.40	24.33	.129	.243
Any illicit drug (other than cannabis) use (%)	21.31	10.73	17.99	.005	.121
First used cannabis before age 21 (%)	81.89	74.68	87.0	.004	.003
Cannabis use frequency (%)				<.001	.700
1–29 days	44.29	21.46	25.69
30–99 days	17.10	20.69	16.99
100–199 days	14.10	17.85	16.35
200–365 days	24.51	40.0	40.97
Cannabis use disorder	4.08	3.62	6.19	.541	.277
Any past-year outpatient healthcare visit (%)				<.001	.012
No	13.88	7.96	2.48
Yes	84.14	88.37	96.19
Don’t know/refused	1.97	3.67	1.33
Any past-years emergency department (ED) visit (%)				.014	.744
No	70.30	59.97	58.75
Yes	28.27	38.11	40.24
Do not know/refused	1.43	1.92	1.0
No. of ED visits, M (SE) (n = 3417)^c	0.58 (0.03)	0.90 (0.14)	0.85 (0.14)	.037	.792
No. of ED visits among those with any ED visit, M (SE) (n = 1061)	2.01 (0.10)	2.32 (0.27)	2.09 (0.30)	.612	.561
Any past-year hospitalization (%)				.004	.282
No	86.19	80.88	74.45
Yes	13.45	19.0	25.55
Do not know/refused	0.36	0.12	0
No. of nights hospitalized, M (SE) (n = 3445)	0.75 (0.07)	1.15 (0.26)	0.96 (0.19)	.349	.549
No. of nights hospitalized among those with any hospitalization, M (SE) (n = 494)^d	5.65 (0.41)	6.19 (1.05)	3.80 (0.52)	.021	.059

Note. For categorical variables, probability values were calculated using χ². For continuous variables^a-d, probability values were calculated using ANOVA (with Bonferroni corrections) for comparisons of all three groups (nonmedical users, medical-only users, and medical/nonmedical users) and t-tests for comparisons between medical-only users and medical/nonmedical users.

Compared to nonmedical-only users, medical-only and medical/nonmedical users had significantly more chronic medical conditions and functional impairments and a higher rate of mental disorders. The three groups did not differ on the rate of nicotine dependence; however, medical-only and medical/nonmedical users had significantly lower rates of alcohol use disorder and other illicit drug use than nonmedical-only users. There was no significant difference in the two groups of medical users on physical, functional, and mental health statuses and rates of alcohol use disorder and other illicit drug use.

Compared to 74.7% of medical-only users, 81.9% of nonmedical-only users and 87.0% of medical/nonmedical users reported cannabis initiation before age 21. Additional analysis showed that only 1.9%, 4.9%, and 1.2% of nonmedical-only, medical-only, and medical/nonmedical users, respectively, initiated use within the preceding five years. Significantly larger proportions (40-41%) of both groups of medical users used cannabis on 200-365 days than nonmedical-only users (25%); however, the three groups did not differ significantly on CUD rates.

A higher proportion of medical/nonmedical users than nonmedical-only and medical-only users had any outpatient visit, but higher proportions of medical-only and medical/nonmedical users (without any significant difference between them) than nonmedical-only users had any ED visit or hospital stay. Of those with any hospitalization, medical/nonmedical users spent fewer nights hospitalized than nonmedical-only users.

Association of Number of ED Visits and Nights Hospitalized With Cannabis Use/Nonuse: Negative Binomial Regression Results

The first column of Table 3 shows that prior-to-past-year cannabis use, compared to never use, was associated with a lower IRR for the number of past-year ED visits (IRR = 0.88, 95% CI = 0.82−0.95). The second column of Table 3 shows that past-year cannabis use, compared to prior-to-past-year use, was associated with a higher IRR for the number of ED visits (IRR = 1.21, 95% CI = 1.10−1.34). However, the third column of Table 3 shows that cannabis use status was not associated with the number of nights hospitalized. These findings partially support H1.

Table 3.

Associations of Cannabis Use Status With Number of Past-year ED Visits and Nights Hospitalized: Negative binomial regression results.

	No. of ED visitsModel 1^a IRR (95% CI)	No. of ED visitsModel 2^b IRR (95% CI)	No. of nights hospitalizedIRR (95% CI)
Cannabis use status: versus never use
Prior-to-past-year use	0.88 (0.82−0.95)**		1.04 (0.93−1.16)
Past-year use	1.05 (0.95−1.17)		1.21 (0.97−1.51)
Past-year cannabis use versus prior-to-past year use		1.21 (1.10−1.34)***
65+ years versus 50–64 years	1.06 (0.98−1.14)	0.98 (0.89−1.08)	1.40 (1.24−1.58)***
Female versus male	0.95 (0.90−1.00)	1.02 (0.93−1.12)	0.70 (0.63−0.78)***
Race/ethnicity: versus non-Hispanic White
Non-Hispanic Black	1.32 (1.21−1.43)***	1.57 (1.38−1.78)***	1.13 (0.98−1.31)
Hispanic	1.04 (0.92−1.18)	1.09 (0.88−1.34)	0.88 (0.72−1.08)
Other	0.91 (0.78−1.07)	1.14 (0.94−1.38)	0.80 (0.58−1.12)
Resident of MCL state versus non-MCL state	1.07 (1.01−1.13)*	1.05 (0.97−1.14)	1.01 (0.92−1.11)
Married versus not married	0.83 (0.77−0.89)***	0.85 (0.77−0.93)**	0.76 (0.69−0.84)***
Bachelor’s degree versus no degree	0.82 (0.76−0.88)***	0.81 (0.75−0.86)***	0.88 (0.80−0.97)*
Working full- or part-time versus not working	0.78 (0.73−0.83)***	0.85 (0.77−0.93)***	0.56 (0.49−0.63)***
Income vs. living in poverty
Up to 2X poverty	0.82 (0.73−0.91)**	0.83 (0.73−0.95)**	0.90 (0.72−1.14)
>2X poverty	0.61 (0.55−0.67)***	0.60 (0.52−0.69)***	0.72 (0.58−0.88)**
Have health insurance versus no insurance	1.13 (0.97−1.31)	1.26 (1.03−1.55)*	1.79 (1.39−2.31)***
No. of chronic medical conditions	1.20 (0.17−1.23)***	1.23 (1.18−1.27)***	1.55 (1.49−1.62)***
No. of functional impairments	1.24 (1.20−1.29)***	1.24 (1.16−1.32)***	1.46 (1.36−1.57)***
No. of sensory impairments	1.16 (1.08−1.26)***	1.15 (1.04−1.26)**	1.09 (0.94−1.25)
Mental illness: versus no mental illness
Mild	1.38 (1.23−1.55)***	1.29 (1.14−1.46)***	1.49 (1.27−1.76)***
Moderate	1.50 (1.28−1.75)***	1.41 (1.14−1.74)**	1.99 (1.46−2.70)***
Serious	1.35 (1.17−1.55)***	1.37 (1.15−1.63)**	1.78 (1.31−2.43)***
Alcohol use disorder versus no disorder	0.98 (0.87−1.11)	0.96 (0.84−1.09)	1.10 (0.85−1.42)
Nicotine dependence vs. no dependence	1.13 (1.00−1.28)*	1.07 (0.96−1.20)	1.19 (0.97−1.46)
Any illicit drug (other than cannabis) use versus no use	1.25 (1.12−1.40)***	1.27 (1.13−1.43)***	1.15 (0.92−1.43)
Model statistics	N = 43,284;	N = 18,800;	N = 43,785;
	design df = 50;	design df = 50;	design df = 50;
	F (23,28) = 172.18;	F (22,29) = 118.71;	F (23,28) = 84.92;
	p<.001	p<.001	p<.001

Prior-to-past-year users and past-year users vs. never users.

Past-year cannabis users vs. prior-to-past-year users.

p<.05; **p<.01; ***p<.001.

Of predisposing, enabling, and need factors, being non-Hispanic Black, number of chronic medical conditions, functional impairment, and sensory impairment, all levels of mental disorder, and any illicit drug use were associated with higher IRRs for ED visits; being married, having a college degree, working full- or part-time, and higher income (compared to living in poverty) were associated with lower IRRs for ED visits in both Models 1 and 2. MCL state residence and nicotine dependence were associated with higher IRRs for ED visits in Model 1, and having health insurance was associated with a higher IRR for ED visits in Model 2.

Age 65+, having health insurance, numbers of chronic medical conditions, functional impairment, and all levels of mental disorder were positively associated with the number of nights hospitalized; and being female, married, having a college degree, working full- or part-time, and higher income were negatively associated. Additional analyses showed no change in the main findings when mental health treatment receipt was added as a covariate.

Association of Number of ED Visits and Nights Hospitalized with Cannabis Use Reasons and Characteristics among Past-year Cannabis Users: Negative Binomial Regression Results

Table 4 shows that medical-only use compared to nonmedical-only use (IRR = 1.38, 95% CI = 1.02–1.88) and 30–99 days of use compared to 1–29 days (IRR = 1.29, 95% CI = 1.04–1.61) were positively associated with the number of ED visits. Cannabis initiation age and CUD were not associated with the number of ED visits. Cannabis use reason and other use characteristics were not significantly associated with the number of nights hospitalized. These findings partially support H2.

Table 4.

Associations of Cannabis use Characteristics With Number of past-year ED Visits and Nights Hospitalized Among Past-Year Cannabis Users: Negative Binomial Regression Results.

	No. of ED visits B (SE)	No. of nights hospitalized B (SE)
Cannabis use reason: versus nonmedical reason
Medical-only	1.38 (1.02−1.88)*	1.42 (0.73−2.78)
Medical/nonmedical	1.15 (0.83−1.61)	1.15 (0.68−1.97)
First used cannabis age <21 versus >21	1.06 (0.83−1.37)	1.31 (0.88−1.96)
Cannabis use frequency: versus 1–29 days
30–99 days	1.29 (1.04−1.61)*	0.74 (0.47−1.15)
100–199 days	0.90 (0.70−1.15)	1.13 (0.70−1.82)
200–365 days	0.91 (0.73−1.15)	0.81 (0.52−1.27)
Cannabis use disorder versus no disorder	1.11 (0.83−1.48)	0.62 (0.33−1.17)
65+ years vs. 50–64 years	1.00 (0.78−1.27)	1.25 (0.87−1.79)
Female versus male	1.12 (0.92−1.36)	0.70 (0.51−0.96)*
Race/ethnicity: versus non-Hispanic White
Non-Hispanic Black	1.45 (1.10−1.90)**	2.15 (1.42−3.26)**
Hispanic	0.91 (0.61−1.37)	0.84 (0.43−1.64)
Other	0.99 (0.65−1.53)	0.84 (0.40−1.76)
Resident of MCL state versus non-MCL state	1.25 (1.03−1.52)*	1.15 (0.79−1.65)
Married versus not married	0.83 (0.67-1.03)	0.58 (0.43−0.79)**
Have a bachelor’s degree versus no degree	0.59 (0.49−0.71)***	0.62 (0.43−0.88)**
Working full- or part-time versus not working	0.82 (0.64−1.04)	0.78 (0.54−1.13)
Income: versus living in poverty
Up to 2X poverty	0.89 (0.68−1.16)	0.63 (0.33-1.18)
>2X poverty	0.65 (0.49−0.85)**	0.56 (0.31−1.02)
Have health insurance versus no insurance	1.22 (0.83−1.79)	1.33 (0.74−2.40)
No. of chronic medical conditions	1.61 (1.06−1.28)**	1.49 (1.29−1.72)***
No. of functional impairments	1.14 (1.00−1.30)*	1.61 (1.37−1.89)***
No. of sensory impairments	1.30 (1.07−1.57)**	1.25 (0.89−1.75)
Mental illness: versus no mental illness
Mild	1.27 (1.02−1.59)*	1.88 (1.41−2.51)***
Moderate	1.67 (1.96−2.64)*	2.86 (1.39−5.89)**
Serious	1.36 (0.95−1.95)	1.37 (0.78−2.41)
Alcohol use disorder versus no disorder	0.86 (0.69−1.07)	0.94 (0.64−1.40)
Nicotine dependence versus no dependence	1.04 (0.86−1.24)	0.79 (0.57−1.08)
Any illicit drug (other than cannabis) use versus no use	1.26 (1.06−1.49)**	0.94 (0.65−1.38)
Model statistics	N = 3404;	N = 3432;
	design df = 50;	design df = 50;
	F (28,23) = 11.38;	F (28,23) = 25.97;
	p<.001	p<.001

*p<.05; **p<.01; ***p<.001

Of predisposing, enabling, and need factors, being non-Hispanic Black, MCL state residence, numbers of chronic medical conditions, functional impairment, sensory impairment, mild and moderate levels of mental illness, and any illicit drug use were positively associated with ED visits; having a college degree and higher income (compared to living in poverty) were negatively associated. Being non-Hispanic Black, number of chronic medical conditions, functional impairment, and mild and moderate mental disorder were positively associated with the number of nights hospitalized; and being female and married and having a college degree and higher income were negatively associated. Additional analyses showed that (1) medical-only and medical/nonmedical users did not differ significantly in associations with number of ED visits and nights hospitalized; and (2) mental health treatment receipt added as a covariate did not change the findings.

Discussion

With almost one in ten people age 50+ reporting past-year cannabis use in 2019 and the proportion likely to continue to increase in the future, this study illuminates older cannabis users’ healthcare utilization. Thirty percent of past-year users, compared to a quarter of never users and prior-to-past-year users, had any ED visit in the past year, and nearly 15% of past-year users and slightly lower proportions of never users and prior-to-past-year users had a hospital stay. The high rates of ED visit and hospitalization suggest that cannabis users have substantial healthcare needs. In multivariable models controlling for predisposing, enabling, and need factors, past-year users did not differ from never users in numbers of ED visits and nights hospitalized, but they had significantly more ED visits than prior-to-past-year users.

As expected, among past-year users, medical (medical-only and medical/nonmedical) users, compared to nonmedical-only users, reported more physical, functional, and mental health problems, which were significantly associated with more ED visits. Higher rates of mental disorders among medical than nonmedical users have been found in previous studies (Lin et al., 2016; Park & Wu, 2017). However, our findings show lower rates of other illicit drug use and alcohol use disorder among medical users, likely due to their poorer physical and functional health. Older adults tend to reduce their drinking as their health declines (Moos et al., 2010). Illicit drug, other than cannabis, use also tends to decline with age (Wu & Blazer, 2014). Nevertheless, the rates of illicit drug use, alcohol use disorder, and nicotine dependence were still significantly higher among medical users than never or prior-to-past-year users. This suggests that a substantial proportion of medical users are long-term cannabis users who may have transitioned to medical use from nonmedical-only use. The finding that less than 5% of medical users initiated use within the preceding five years supports this possibility.

Consistent with the primacy of health-related need factors in the Behavioral Model (Babitsch et al., 2012), greater health-related needs among past-year users, especially medical users, may at least partially explain their higher number of ED visits. These positive associations, especially regarding the number of ED visits and behavioral health problems, show that older adults with behavioral health problems have higher healthcare needs. Our study shows that compared to their nonusing peers, cannabis users, regardless of use reasons, tend to have significantly higher rates of mental illness, especially serious mental illness, and other substance use/use disorders. These high rates of behavioral health problems and associated healthcare needs among older cannabis users have important public health implications as the numbers/proportions of older adults who use cannabis will continue to rise with the aging of young and middle-aged adults who are accustomed to using cannabis as increasing numbers of states have legalized cannabis.

Of cannabis use characteristics, only 30-99 days, compared to 1-29 days, of use was associated with more ED visits. The lack of association between number of ED visits and higher frequency use and between number of ED visits and nights hospitalized with CUD (which also included more high frequency users) are notable and calls for more research. A cross-sectional study of urban primary care clinic patients also found no significant associations between cannabis use frequency and ED visits or hospitalizations (Fuster et al., 2014). The authors speculated that cannabis had little additive impact above other drug use; however, they recognized that explanation would not apply to cannabis-only users.

Though high frequency cannabis use was not associated with healthcare service utilization in our cross-sectional examination, it may exacerbate mental and cognitive health problems, especially given potential safety issues and adverse effects of commercially available cannabis and cannabis products and increasing THC potency (Elsohly et al., 2016; Matheson & Le Foll, 2020). Older cannabis users are heterogeneous in their pathways to cannabis use (Arora et al., 2021), and those with low health literacy and those who used cannabis earlier in life, in particular, may be unaware of the high THC level in cannabis products today and related potential safety concerns.

The Behavioral Model also posits the facilitating or impeding roles of predisposing and enabling factors in healthcare use. More nights of hospitalization among the 65+ age group compared to the 50-64 age group may have been in part due to the greater healthcare needs associated with aging. More ED visits and hospitalizations among Black older adults also likely reflect systemic racism in the healthcare system and resulting health disparities throughout these older adults’ lives. Regardless of cannabis use status and use characteristics, those with SES advantages more likely used preventive care on a regular basis, reducing serious health crises requiring ED visits or hospitalizations.

Study limitations are: (1) The validity of respondents’ self-reported healthcare use and cannabis use characteristics were not ascertained. (2) NSDUH uses the term “marijuana” without distinguishing between THC and cannabidiol (some older cannabis users may use only cannabidiol products, which lack THC’s psychoactive effects). (3) NSDUH’s cross-sectional data allow examination of correlation but not causation. (4) Despite pooling five years of data, the number who used for both medical and nonmedical reasons was small, which may explain nonsignificant associations between this category of use and the outcome variables. (5) There may have been unmeasured covariates in the associations examined.

In conclusion, the study shows that older adults who use cannabis tend to have greater healthcare needs than their peers who do not use cannabis. Older cannabis users’ poorer physical/functional and behavioral health statuses were significantly associated with greater numbers of ED visits. The higher rates of behavioral health problems among older cannabis users are especially concerning for their physical and mental well-being and healthcare needs. The study’s clinical and policy implications are: (1) The role of EDs in educating patients on cannabis, cannabinoids, and illicit drugs’ physical and mental health effects and potential safety issues and on preventive healthcare is essential. (2) Improved access to preventive healthcare for older substance users is needed to reduce health crises leading to ED visits. (3) With increasing cannabis and other illicit drug use among older adults, specialized behavioral health treatment services are needed.

Footnotes

Author Contributions

Study conceptualization: NGC, DMD, and BYC; data management: NGC; data analysis and interpretation: NGC, CNM, and BYC; manuscript draft: NGC; final editing: NGC, DMD, CNM, and BYC

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by grant, P30AG066614, awarded to the Center on Aging and Population Sciences at The University of Texas at Austin by the National Institute on Aging. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Ethical Approval

This study based on de-identified public-domain data was exempt by the University of Texas at Austin’s Institutional Review Board.

ORCID iD

Namkee G. Choi

References

American Psychiatric Association . (1994). Diagnostic and statistical manual of mental disorders. 4th ed. American Psychiatric Association.

Andersen

Newman

J. F.

(1973). Societal and individual determinants of medical care utilization in the United States. The Milbank Memorial Fund QuarterlyHealth and Society, 51(1), 95–124. https://doi.org/10.2307/3349613.

Andersen

R. M.

(1995). Revisiting the behavioral model and access to medical care: Does it matter? Journal of Health and Social Behavior, 36(1), 1–10. https://doi.org/10.2307/2137284.

Arora

Qualls

S. H.

Bobitt

Milavetz

Kaskie

(2021). Older cannabis users are not all alike: Lifespan cannabis use patterns. Journal of Applied Gerontology, 40(1), 87–94. https://doi.org/10.1177/0733464819894922.

Babitsch

Gohl

&von Lengerke

(2012). Re-revisiting Andersen’s Behavioral Model of Health Services Use: A systematic review of studies from 1998-2011. Psychosocial Medicine, 9, Doc11. https://doi.org/10.3205/psm000089.

Bahorik

A. L.

Campbell

C.I.

Sterling

S. A.

Leibowitz

Travis

Weisner

C. M.

Satre

D. D.

(2018). Adverse impact of marijuana use on clinical outcomes among psychiatry patients with depression and alcohol use disorder. Psychiatry Research, 259, 316-322. https://doi.org/10.1016/j.psychres.2017.10.026.

Balachandran

Elsohly

Hill

K. P.

(2021). Cannabidiol interactions with medications, illicit substances, and alcohol: A comprehensive review. Journal of General Internal Medicine, 36(7), 2074–2084. https://doi.org/10.1007/s11606-020-06504-8.

Banerjee

Gandhi

A. B.

Antony

Alexander

Hisbulla

Kannichamy

Kaleem

Mishra

Khan

(2020). Role of cannabis in the incidence of myocardial infarction: A review. Cureus, 12(10), e11097. https://doi.org/10.7759/cureus.11097.

Bonomo

Norman

Biondo

Bruno

Daglish

Dawe

Egerton-Warburton

Karro

Kim

Lenton

Lubman

D. I.

Pastor

Rundle

Ryan

Gordon

Sharry

Nutt

Castle

(2019). The Australian drug harms ranking study. Journal of Psychopharmacology, 33(7), 759–768. https://doi.org/10.1177/0269881119841569.

10.

Bradford

A. C.

Bradford

W. D.

Abraham

Bagwell Adams

(2018). Association between US state medical cannabis laws and opioid prescribing in the Medicare Part D population. JAMA Internal Medicine, 178(5), 667–672. https://doi.org/10.1001/jamainternmed.2018.0266.

11.

Brown

J. D.

Costales

Van Boemmel-Wegmann

Goodin

A.J.

Segal

Winterstein

A.G.

(2020). Characteristics of older adults who were early adopters of medical cannabis in the Florida medical marijuana use registry. Journal of Clinical Medicine, 9(4), 1166. https://doi.org/10.3390/jcm9041166.

12.

Broyd

S. J.

van Hell

H. H.

Beale

Yücel

Solowij

(2016). Acute and chronic effects of cannabinoids on human cognition-a systematic review. Biological Psychiatry, 79(7), 557–567. https://doi.org/10.1016/j.biopsych.2015.12.002.

13.

Campeny

López-Pelayo

Nutt

Blithikioti

Oliveras

Nuño

Maldonado

Florez

Arias

Fernández-Artamendi

Villalbí

J. R.

Sellarès

Ballbè

Rehm

Balcells-Olivero

M. M.

Gual

(2020). The blind men and the elephant: Systematic review of systematic reviews of cannabis use related health harms. European Neuropsychopharmacology, 33, 1-35. https://doi.org/10.1016/j.euroneuro.2020.02.003.

14.

Center for Behavioral Health Statistics and Quality . (2019). National Survey on Drug Use and Health Public Use File Codebook. Substance Abuse and Mental Health Services Administration.

15.

Choi

N. G.

DiNitto

D. M.

Choi

B. Y.

(2021). Prescription pain reliever use and misuse among cannabis users aged 50+ years. Clinial Gerontologist, 44(1), 53–65. https://doi.org/10.1080/07317115.2020.175754.

16.

Choi

N. G.

Marti

C. N.

DiNitto

D. M.

Choi

B. Y.

(2018). Older adults’ marijuana use, injuries, and emergency department visits. American Journal of Drug and Alcohol Abuse, 44(2):215–223. https://doi.org/10.1080/00952990.2017.1318891.

17.

Damkier

Lassen

Christensen

M. M. H.

Madsen

K. G.

Hellfritzsch

Pottegård

(2019). Interaction between warfarin and cannabis. Basic and Clinical Pharmacology and Toxicology, 124(1):28–31. https://doi.org/10.1111/bcpt.13152.

18.

Desai

Patel

Sharma

Amin

Bhuva

Patel

M. S.

Sharma

Shah

Patel

Savani

Batra

Kumar

(2017). Nonmedical marijuana use and acute myocardial infarction: Insights from Nationwide Inpatient Sample in the United States. Cureus, 9(11), e1816. https://doi.org/10.7759/cureus.1816.

19.

ElSohly

M. A.

Mehmedic

Foster

Gon

Chandra

Church

J. C.

(2016). Changes in cannabis potency over the last 2 decades (1995-2014): Analysis of current data in the United States. Biological Psychiatry, 79(7), 613–619. https://doi.org/10.1016/j.biopsych.2016.01.004.

20.

Fairman

B. J.

(2016). Trends in registered medical marijuana participation across 13 US states and District of Columbia. Drug and Alcohol Dependence, 159, 72–79. https://doi.org/10.1016/j.drugalcdep.2015.11.015.

21.

Fuster

Cheng

D. M.

Allensworth-Davies

Palfai

T. P.

Samet

J. H.

Saitz

(2014). No detectable association between frequency of marijuana use and health or healthcare utilization among primary care patients who screen positive for drug use. Journal of General Internal Medicine, 29(1), 133–139. https://doi.org/10.1007/s11606-013-2605-z.

22.

Gelman

Hill

(2007). Data analysis using regression and multilevel/hierarchical models. Cambridge University Press.

23.

Hall

Lynskey

(2020). Assessing the public health impacts of legalizing nonmedical cannabis use: The US experience. World Psychiatry, 19(2), 179–186. https://doi.org/10.1002/wps.20735.

24.

Hall

K. E.

Monte

A. A.

Chang

Fox

Brevik

Vigil

D. I.

Van Dyke

James

K. A.

(2018). Mental health-related emergency department visits associated with cannabis in Colorado. Academic Emergency Medicine, 25(5), 526–537. https://doi.org/10.1111/acem.13393.

25.

Hasin

Walsh

(2020). Cannabis use, CUD, and comorbid psychiatric illness: A narrative review. Journal of Clinical Medicine, 10(1), 15. https://doi.org/10.3390/jcm10010015.

26.

Heatherton

T. F.

Kozlowski

L. T.

Frecker

R. C.

Fagerström

K. O.

(1991). The fagerström test for nicotine dependence: A revision of the fagerström tolerance questionnaire. British Journal on Addiction, 86(9), 1119–1127. https://doi.org/10.1111/j.1360-0443.1991.tb01879.x.

27.

Johal

Devji

Chang

Simone

Vannabouathong

Bhandari

(2020). Cannabinoids in chronic non-cancer pain: A systematic review and meta-analysis. Clinical Medicine Insights: Arthritis and Musculoskeletal Disorders, 13, 1179544120906461. https://doi.org/10.1177/1179544120906461.

28.

Kaskie

Ayyagari

Milavetz

Shane

Arora

(2017). The increasing use of cannabis among older Americans: A public health crisis or viable policy alternative? The Gerontologist, 57(6), 1166–1172. https://doi.org/10.1093/geront/gnw166.

29.

Kessler

R. C.

Barker

P. R.

Colpe

L. J.

Epstein

J. F.

Gfroerer

J. C.

Hiripi

Howes

M. J.

Normand

S. L.

Manderscheid

R. W.

Walters

E. E.

Zaslavsky

A. M.

(2003). Screening for serious mental illness in the general population. Archives of General Psychiatry, 60(2), 184–189. https://doi.org/10.1001/archpsyc.60.2.184.

30.

Lev-Ran

Roerecke

Le Foll

George

T. P.

McKenzie

Rehm

(2014). The association between cannabis use and depression: A systematic review and meta-analysis of longitudinal studies. Psychological Medicine, 44(4), 797–810. https://doi.org/10.1017/S0033291713001438.

31.

Lin

L. A.

Ilgen

M. A.

Jannausch

Bohnert

K. M.

(2016). Comparing adults who use cannabis medically with those who use nonmedically: Results from a national sample. Addictive Behaviors, 61, 99–103. https://doi.org/10.1016/j.addbeh.2016.05.015.

32.

Lloyd

S. L.

Striley

C. W.

(2018). Marijuana use among adults 50 years or older in the 21st century. Gerontology & Geriatric Medicine, 4, 2333721418781668. https://doi.org/10.1177/2333721418781668.

33.

Matheson

Le Foll

(2020). Cannabis legalization and acute harm from high potency cannabis products: A narrative review and recommendations for public health. Frontiers in Psychiatry, 11, 591979. https://doi.org/10.3389/fpsyt.2020.591979.

34.

Moos

R. H.

Brennan

P. L.

Schutte

K. K.

Moos

B. S.

(2010). Older adults’ health and late-life drinking patterns: A 20-year perspective. Aging & Mental Health, 14(1), 33–43. https://doi.org/10.1080/13607860902918264.

35.

National Academies of Sciences . (2017). The health effects of cannabis and cannabinoids: The current state of evidence and recommendations for research. National Academies Press (US).Engineering, and Medicine

36.

Nicholas

L. H.

Maclean

J. C.

(2019). The effect of medical marijuana laws on the health and labor supply of older adults: Evidence from the Health and Retirement Study. Journal of Policy Analysis and Management, 38(2), 455–480.

37.

Novak

S. P.

Colpe

L. J.

Barker

P. R.

Gfroerer

J. C.

(2010). Development of a brief mental health impairment scale using a nationally representative sample in the USA. International Journal of Methods in Psychiatric Research, 19(Suppl 1), 49–60. https://doi.org/10.1002/mpr.313.

38.

Park

J. Y.

L. T.

(2017). Differences in behavioral health disorders and unmet treatment needs between medical marijuana users and nonmedical marijuana users: Results from a national adult sample. Drug and Alcohol Dependence, 180, 311-318. https://doi.org/10.1016/j.drugalcdep.2017.08.028.

39.

Perisetti

Gajendran

Dasari

C. S.

Bansal

Aziz

Inamdar

Tharian

Goyal

(2020). Cannabis hyperemesis syndrome: An update on the pathophysiology and management. Annals of Gastroenterology, 33(6), 571–578. https://doi.org/10.20524/aog.2020.0528.

40.

Porr

C. J.

Rios

Bajaj

H. S.

Egan

A. M.

Huot

Batten

Bishop

Ryan

Davis

Darvesh

Rahman

Asghari

Acheampong

Tricco

A. C.

(2020). The effects of nonmedical cannabis use on glycemic outcomes and self-management behaviours in people with type 1 and type 2 diabetes: A rapid review. Systematic Reviews, 9(1), 187. https://doi.org/10.1186/s13643-020-01411-9.

41.

Rabgay

Waranuch

Chaiyakunapruk

Sawangjit

Ingkaninan

Dilokthornsakul

(2020). The effects of cannabis, cannabinoids, and their administration routes on pain control efficacy and safety: A systematic review and network meta-analysis. Journal of the American Pharmacists Association, 60(1), 225–234. https://doi.org/10.1016/j.japh.2019.07.015.

42.

Ramadan

M. M.

Banta

J. E.

Bahjri

Montgomery

S. B.

(2021). Marijuana users are likely to report opioid misuse among adults over 50 years in representative sample of the United States (2002-2014). Journal of Addictive Disorders, 39(1), 66–73. https://doi.org/10.1080/10550887.2020.1816117.

43.

Reynolds

I. R.

Fixen

D. R.

Parnes

B. L.

Lum

H. D.

Shanbhag

Church

Linnebur

S. A.

Orosz

(2018). Characteristics and patterns of marijuana use in community-dwelling older adults. Journal of the American Geriatrics Society, 66(11), 2167–2171. https://doi.org/10.1111/jgs.15507.

44.

Roberts

B. A.

(2019). Legalized cannabis in Colorado emergency departments: A cautionary review of negative health and safety effects. Western Journal of Emergency Medicine, 20(4), 557–572. https://doi.org/10.5811/westjem.2019.4.39935.

45.

Substance Abuse and Mental Health Services Administration . (2020). Key substance use and mental health indicators in the United States: Results from the 2019 National Survey on Drug Use and Health. HHS Publication No. PEP20-07-01-001, NSDUH Series H-55.

46.

Vacaflor

B. E.

Beauchet

Jarvis

G. E.

Schavietto

Rej

(2020). Mental health and cognition in older cannabis users: A review. Canadian Geriatrics Journal, 23(3), 242–249. https://doi.org/10.5770/cgj.23.399.

47.

Volkow

N. D.

Swanson

J. M.

Evins

A. E.

DeLisi

L .E.

Meier

M. H.

Gonzalez

Bloomfield

M. A.

Curran

H. V.

Baler

(2016). Effects of cannabis use on human behavior, including cognition, motivation, and psychosis: A review. JAMA. Psychiatry, 73(3):292–297. https://doi.org/10.1001/jamapsychiatry.2015.3278.

48.

Wang

G. S.

Davies

S. D.

Halmo

L. S.

Sass

Mistry

R. D.

(2018). Impact of marijuana legalization in Colorado on adolescent emergency and urgent care visits. Journal of Adolescent Health, 63(2), 239–241. https://doi.org/10.1016/j.jadohealth.2017.12.010.

49.

L-T.

Blazer

D. G.

(2014). Substance use disorders and psychiatric comorbidity in mid and later life: A review. International Journal of Epidemiology, 43(2), 304–317. https://doi.org/10.1093/ije/dyt173.

50.

Yang

K. H.

Kaufmann

C. N.

Nafsu

Lifset

E. T.

Nguyen

Sexton

Han

B. H.

Kim

Moore

A. A.

(2021). Cannabis: An emerging treatment for common symptoms in older adults. Journal of the American Geriatrics Society, 69(1), 91–97. https://doi.org/10.1111/jgs.16833.