Original quantitative research – Shifts in medical cannabis use in Canada during the COVID-19 pandemic

Abstract

Introduction: The COVID-19 pandemic has had widespread secondary negative health impacts including loss of material security and exacerbation of mental illness in at-risk populations. While increases in the nonmedical use of certain substances, including cannabis, have been observed in samples of the Canadian population, no research has documented COVID-concurrent shifts in medical cannabis use in Canada.

Methods: Data were derived from the 2021 Canadian Cannabis Patient Survey, an online survey administered in May 2021 to people authorized to use medical cannabis recruited from one of two Canadian licensed medical cannabis producers. McNemar tests assessed for changes in past 3-month medical cannabis frequency from before to during the pandemic. We explored correlates of increasing frequency of cannabis use since before the pandemic in bivariable and multivariable logistic models.

Results: In total, 2697 respondents (49.1% women) completed the survey. Daily medical cannabis use increased slightly but significantly from before the pandemic (83.2%) to during the pandemic (90.3% at time of survey; p < 0.001). Factors significantly associated with increasing frequency of medical cannabis use included female gender, younger age, pandemic-related job loss, primary cannabis use to manage mental health, prescription drug use and nonmedical cannabis use (p < 0.05).

Conclusion: There were slight shifts towards higher frequency of medical cannabis use after the onset of the COVID-19 pandemic. While short- and long-term impacts of cannabis use on pandemic-related mental distress are unknown, clinicians working with patients who use medical cannabis should be aware of possible changes in use patterns during the pandemic.

Keywords: COVID-19, cannabis, medical cannabis, Canada, survey

Introduction

The COVID-19 pandemic and associated policy response measures have had widespread, long-lasting and compounding direct and indirect impacts on population health and quality of life around the world. In many settings, nonemergency medical procedures were deferred to focus health care resources on the COVID-19 response, and public health measures such as restrictions on in-person gatherings, travel and non-essential service operations were implemented to curb community transmission of the virus. While critical to reducing case rates, these abrupt changes spurred increased feelings of loneliness, boredom and isolation^{Footnote 1} in addition to rising material and income insecurity.^{Footnote 2}

The disruption to everyday life, coupled with fears about contracting or transmitting SARS-CoV-2, became the source of substantial stress for many people.^{Footnote 3Footnote 4Footnote 5Footnote 6Footnote 7} Indeed, early studies observed marked deteriorations in mental health in populations world-wide as well as exacerbated symptoms of anxiety, distress, depression, insomnia and posttraumatic stress disorder (PTSD).^{Footnote 8Footnote 9Footnote 10Footnote 11Footnote 12}

Secondary to these pandemic-induced shifts in social, economic and psychological health states have been changes to the contexts, motivations and patterns of substance use. For example, approximately 1 in 8 adults in a representative sample of the US population reported starting or increasing the use of a substance to cope with pandemic-related negative emotions alongside high rates of self-reported anxiety and/or depression (31%) and trauma- and stress-related disorder symptoms (26%).^{Footnote 13} Studies from Australia, Canada and the US have shown increased alcohol consumption among people experiencing elevated stress, anxiety and depression.^{Footnote 14Footnote 15Footnote 16Footnote 17Footnote 18Footnote 19}

In Canada, where cannabis has been legal and regulated since 2018, the prevalence of current (past-week) nonmedical cannabis use in the overall population does not appear to have shifted noticeably during the COVID-19 pandemic;^{Footnote 20} however, between 30% and 50% of people who used nonmedical cannabis pre-pandemic increased their frequency of use during the pandemic;^{Footnote 3Footnote 20Footnote 21Footnote 22} this estimate is even higher among those with depression, anxiety and/or suicidality.^{Footnote 3}

Surveys found that between 30% and 40% of people who use cannabis do so for both medical and nonmedical purposes.^{Footnote 23Footnote 24} In Canada, clinicians can authorize medical cannabis for a wide range of conditions or symptoms, including pain, anxiety, depression, PTSD and sleep disorders.^{Footnote 25Footnote 26} About 22% of people who use cannabis for medical purposes in Canada are authorized to do so by a health care professional.^{Footnote 27}

Although many overlapping symptoms and conditions commonly managed with medical cannabis (e.g. chronic pain, anxiety, depression, PTSD, insomnia) are reported to have worsened in the population during the COVID-19 pandemic,^{Footnote 28Footnote 29} studies investigating changes in medical cannabis use among Canadians authorized to use it are lacking. Annual data from the Canadian Cannabis Survey suggest that the prevalence of self-reported (i.e. not necessarily authorized) medical cannabis use has remained stable over 2019, 2020 and 2021.^{Footnote 27Footnote 30} We are unaware of any study examining shifts in frequency of medical cannabis use over this time in a single sample of people who use medical cannabis.

Using information collected from a large sample of Canadian residents authorized to use medical cannabis, we sought to (1) document changes in the frequency of medical cannabis use; (2) explore concurrent changes in use of alcohol, tobacco, and prescription and unregulated drugs; and (3) identify independent correlates of increasing medical cannabis use during the COVID-19 pandemic.

Methods

Study sample

Data for this study were derived from the 2021 Canadian Cannabis Patient Survey. The survey was developed in consultation with academic partners at institutions across Canada and the United States and administered by Tilray, a Canadian licensed producer of medical cannabis.

All people authorized to use medical cannabis registered with Tilray and/or the licensed producer Aphria were emailed a password-protected link to complete the survey. The survey was available from 7 to 14 May 2021 on REDCap, a HIPAA- (in the USA) and PIPEDA- (in Canada) compliant electronic data capture system. Respondents provided informed consent prior to participating in the study and answered questions on demographics, the reasons they are using medical cannabis and their patterns of use of cannabis and other substances. People who completed the survey and provided a valid Tilray or Aphria patient number were entered into a draw for one of three CAD 1000 credits towards the purchase of medical cannabis from their licensed provider.

Ethics approval for this survey was granted by Advarra, an independent institutional review board (approval number: Pro00050772).

Measures

Medical cannabis use frequency

To explore patterns of medical cannabis use during the COVID-19 pandemic, we added a block of pandemic-related questions to the 2021 Canadian Cannabis Patient Survey.

First, we asked respondents to report their average frequency of medical cannabis use during three time periods that we defined in relation to the onset of the global pandemic: “pre-COVID” was the approximate 3-month period preceding the declaration of the COVID-19 pandemic (i.e. approximately 1 January 2020 to 15 March 2020); “Wave 1” was the initial period after the declaration of the pandemic during which new cases steadily increased, then decreased and remained relatively stable (i.e. approximately 15 March 2020 to 30 August 2020; and “Wave 2” was associated with much more rapid and higher surge of new cases (i.e. approximately 1 October 2020 to the time of data collection).

For each pre-specified period, participants were asked how often they used medical cannabis. Consistent with the Alcohol, Smoking and Substance Involvement Screening Test (ASSIST),^{Footnote 31} the response categories were “never,” “once or twice,” “monthly,” “weekly” and “daily or almost daily.” In anticipation of high rates of daily use among this sample of medical users,^{Footnote 24} respondents authorized to use medical cannabis daily were further prompted to specify the approximate number of times they used cannabis per day: 1 to 2, 3 to 5, 6 to 10, or more than 10.

To obtain a measure corresponding to change and direction of change of medical cannabis use frequency during the COVID-19 pandemic, we ordered the original frequency categories on a scale from 0 to 7 (i.e. never [0] to >10 times/day [7]) and calculated a change from baseline (i.e. “pre-COVID”) score separately for Wave 1 and Wave 2 by subtracting the pre-COVID frequency from the Wave 1 and Wave 2 frequencies, respectively. Change scores of 1 or more, −1 or less and 0, respectively, corresponded to an increase, decrease or no change in medical cannabis use frequency for that period. A change score of 1 or more in either period was considered an increase from baseline; a change score of 1 or less in either period was considered a decrease from baseline.

Reasons for changing medical cannabis use

We asked respondents whether they perceived their medical cannabis use as having changed during the pandemic. Those who responded “yes” were prompted to identify the reason(s) underlying this change. The response categories included adjustment to medical cannabis authorization; change in access or availability of cannabis; cost; social distancing regulations; time at home; anxiety; sleep problems; isolation; loneliness; COVID-19 risk; COVID-19 symptom severity; and other (with a text box for additional responses).

Changes in other substance use

We asked respondents to report how their use of other substances, including alcohol, tobacco, prescription drugs (opioids and non-opioids) and unregulated drugs (cocaine or crack, methamphetamine, unregulated opioids) changed during the pandemic (i.e. after 15 March 2020). Respondents could report an increase, decrease, no change or no use of that substance.

Statistical analysis

First, we examined the number and proportion of respondents who reported less than weekly, weekly and daily medical cannabis use in each assessment period. We used a McNemar–Bowker test to assess for overall within-group changes in frequency of use, and McNemar tests with Bonferoni correction to examine pairwise changes in frequency between assessment periods (e.g. less than weekly to weekly from pre-COVID to Wave 1, less than weekly to weekly from Wave 1 to Wave 2, etc.). We then descriptively examined the reasons given for increased or decreased frequency of use since the pre-COVID assessment period.

Next, we used Chi-square tests (or Fisher exact tests, as appropriate) to examine changes in secondary substance use during the pandemic (i.e. stable or decrease or increase in frequency), stratified by increased medical cannabis use during the pandemic (yes vs. no). We followed up significant (p < 0.05) results with post hoc pairwise Chi-square tests with Bonferroni correction.

Finally, we built exploratory bivariable and multivariable logistic regression models to examine sociodemographic, substance use and health-related correlates of increasing medical cannabis use during the COVID-19 pandemic. The analytic sample for the model was restricted to respondents who were eligible to increase their medical cannabis use frequency (i.e. those who used medical cannabis less than 10 times per day during the pre-COVID period).

We considered the following covariates: gender (female vs. male); age (per 5-year increase); household annual income (in CAD, <40 000, 40 000–99 999, 100 000–159 999, ≥160 000); employment status (part/full-time, unemployed/on disability, retired); lost employment during the pandemic (yes vs. no); community setting (urban/suburban vs. rural); nonmedical cannabis use (yes vs. no); alcohol use (yes vs. no); tobacco use (yes vs. no); use of cocaine/crack, illicit opioids (e.g. heroin) or methamphetamine (pooled together into a category for “unregulated drugs”; yes vs. no); prescription opioids, non-opioid prescription drugs (pooled into a category for prescription drugs; yes vs. no); and primary symptom(s) treated with medical cannabis. The information about primary symptoms treated with medical cannabis was derived from a checklist of symptom categories including an option to describe additional symptoms.

We derived the following binary (yes vs. no) groupings from the predetermined and self-described symptoms: (1) pain, inflammation, nerve damage; (2) sleep problems; (3) anxiety, stress, hypersensitivity; (4) depression, low mood; (5) addiction, dependence, withdrawal; (6) attention deficit, memory loss; (7) nausea, appetite loss or gastrointestinal issues; (8) seizures, muscle spasms, tremors; and (9) other (responses that could not be reasonably re-categorized into any of the above groups).

We adopted a conservative multivariable model-building approach in which all these covariates with a bivariable significance level (p < 0.20) were included in multivariable analyses.

As some participants may have increased their medical cannabis use because they received medical authorization to do so, we conducted a sensitivity analysis. Participants whose sole underlying reason for their increased use of medical cannabis was a change to their medical cannabis authorization were recoded to 0 for the outcome.

All analyses were conducted in R Version 1.4.1106 using RStudio (R Foundation for Statistical Computing, Vienna, AT). All p-values are two-sided.

Results

Of the 27 431 people registered with Tilray or Aphria (or both) and who were sent a link to the survey, 2697 (9.8%) provided complete responses. Of these, 1325 (49.1%) were women (mean [SD] age: 54.3 [14.0] years). Most respondents (91.3%) were White and living in central Canada or the Prairies (see Table 1).

Table 1. Demographic characteristics of a sample of Canadian residents authorized to use medical cannabis (n = 2697)

Characteristic	n	%
Mean age (SD), in years	54.3 (14.0)	N/A
Gender
Male	1352	50.1
Female	1325	49.1
OtherFootnote a	10	0.4
Did not disclose	10	0.4
Race/ethnicityFootnote b
White	2463	91.3
Black	36	1.3
Hispanic	27	1.0
Asian	60	2.2
Indigenous	67	2.5
Métis	64	2.4
Other	70	2.6
Geographical region
Atlantic Canada	251	9.3
Central Canada	1276	47.3
Prairies	930	34.5
West Coast	235	8.7
Northern Territories	5	0.2
Community setting
Urban	2077	77.0
Rural	620	23.0
Annual household income, CAD
<40 000	607	22.5
40 000–99 999	1281	47.5
100 000–159 999	599	22.2
≥160 000	210	7.8
Employment
Full-time or part-time	1202	44.6
Unemployed	608	22.5
Retired	887	32.9
Nonmedical cannabis use
Yes	923	34.2
No	1764	65.4

The prevalence of daily medical cannabis use in the sample was 83.2% (n = 2245) during the pre-COVID period; 85.9% (n = 2317) during Wave 1; and 90.3% (n = 2422) during Wave 2 (see Table 2). A McNemar–Bowker test confirmed significant changes in the frequency of medical cannabis use during the pandemic (p < 0.001), with pairwise post hoc tests showing a significant increase from less than weekly use to weekly and daily use between the pre-COVID period and Wave 1 (both adjusted p < 0.001), and a significant increase from less than weekly to daily use from Wave 1 to Wave 2 (adjusted p < 0.001; Table 2).

Table 2. Frequency of medical cannabis use before and during the COVID-19 pandemic reported by a sample of Canadian residents authorized to use medical cannabis (n = 2697)

Frequency	n (%)
	Pre-COVIDFootnote a	Wave 1Footnote b	Wave 2Footnote c
<Weekly	294 (10.9)	243 (9.0)	135 (5.0)
Weekly	158 (5.9)	137 (5.1)	125 (4.7)
Daily	2245 (83.2)	2317 (85.9)	2422 (90.3)

About 546 (18.4%) respondents increased their medical cannabis use frequency since the pre-COVID period. Common reasons for increasing use frequency included anxiety (n = 280 [51.3%]); sleep problems (n = 206 [37.7%]); social distancing regulations and/or more time at home (n = 194 [35.5%]); isolation and/or loneliness (n = 149 [27.3%]); and change to medical cannabis authorization (n = 91 [16.7%]; see Table 3).

Table 3. Reasons for changing medical cannabis use, by direction of frequency change from baseline, given by a sample of Canadian residents authorized to use medical cannabis (n = 2697)

Reason for change from the pre-COVID period	n (%)
	Increase (n = 546; 18.4%)	Decrease (n = 123; 4.6%)
Change to medical cannabis authorization	91 (16.7)	5 (4.1)
Change in access or availability	33 (6.0)	5 (4.1)
Cost	45 (8.2)	19 (15.4)
Social distancing regulations / time at home	194 (35.5)	19 (15.4)
Anxiety	280 (51.3)	23 (18.7)
Sleep problems	206 (37.7)	26 (21.1)
Isolation / loneliness	149 (27.3)	10 (8.1)
COVID-19 risk or symptom severity	50 (9.2)	6 (4.9)
Other
Change in medical need	15 (2.7)	2 (1.6)
Stress	3 (0.6)	0
Started a different treatment	0	4 (3.3)
Self-guided experimentation	3 (0.5)	5 (4.1)
Not achieving desired effect	0	3 (1.6)
Negative side effects	0	5 (4.1)
Uncategorized other	2 (0.7)Footnote a	3 (3.3)Footnote b
Not reportedFootnote c	155 (28.4)	49 (39.8)

Another 123 (18.4%) respondents decreased their use frequency. The top reasons given for changing use frequency were anxiety and sleep problems, reported by 18.7% (n = 23) and 21.1% (n = 26) of the group, respectively.

In total, 50 (9.2%) of the respondents who increased their use and 6 (4.9%) of those who decreased their use cited risk of COVID-19 infection or symptom severity as a reason for changing frequency (see Table 3 for reasons for changing frequency of medical cannabis use during the COVID-19 pandemic).

Significantly more respondents who increased their medical cannabis frequency during the pandemic reported an increase in alcohol use, while significantly more respondents who did not increase their medical cannabis frequency reported no change in alcohol use (both p < 0.001). The groups did not differ in terms of perceived reductions in alcohol consumption. Similar group differences were noted for perceived changes in tobacco consumption (see Table 4). Reported changes in consumption of prescription opioids, unregulated opioids, non-opioid prescription drugs, crack/cocaine and methamphetamine did not differ during the pandemic between the groups.

Table 4. Changes in secondary substance use during the COVID-19 pandemic, stratified by medical cannabis use change (increase/no increase) reported by a sample of Canadian residents authorized to use medical cannabis (n = 2697)

Secondary substanceFootnote a	Increased medical cannabis useFootnote b n (%)		p value
	Yes	No
Prescription opioids (n = 364)	79 (21.7)	285 (78.3)	0.199
Stable	53 (67.1)	203 (71.2)
Decrease	14 (17.7)	58 (20.4)
Increase	12 (15.2)	24 (8.4)
Unregulated opioids (n = 49)	13 (26.5)	36 (73.5)	0.086Footnote c
Stable	8 (61.5)	29 (80.6)
Decrease	2 (15.4)	0 (0.0)
Increase	3 (23.1)	7 (19.4)
Non-opioid prescription drugs (n = 1017)	223 (21.9)	794 (78.1)	0.119
Stable	174 (78.0)	666 (83.8)
Decrease	29 (13.0)	71 (8.9)
Increase	20 (9.0)	57 (7.2)
Crack or cocaine (n = 46)	15 (32.6)	31 (67.4)	1.000Footnote c
Stable	9 (60.0)	20 (64.5)
Decrease	3 (20.0)	5 (16.1)
Increase	3 (20.0)	6 (19.4)
Methamphetamine (n = 42)	10 (23.8)	32 (76.2)	0.391
Stable	8 (80.0)	24 (75.0)
Decrease	1 (10.0)	1 (3.1)
Increase	1 (10.0)	7 (21.9)
Alcohol (n = 1538)	339 (22.0)	1199 (78.0)	<0.001
Stable	182 (53.7)	853 (71.1)	<0.001
Decrease	77 (22.7)	208 (17.3)	0.149
Increase	80 (23.6)	138 (11.5)	<0.001
Tobacco (n = 471)	116 (24.6)	355 (75.4)	<0.001
Stable	55 (47.4)	246 (69.3)	<0.001
Decrease	29 (25.0)	65 (18.3)	0.705
Increase	32 (27.6)	44 (12.4)	<0.001

The odds of increasing medical cannabis use during the COVID-19 pandemic were significantly elevated among women (adjusted odds ratio [aOR] = 1.67; 95% confidence interval [CI]: 1.36–2.05); respondents who lost their jobs during the pandemic (1.38; 1.06–1.80); those who use cannabis for nonmedical purposes (1.35; 1.09–1.68); those who use prescription drugs (1.24; 1.00–1.54); and those who reported anxiety/stress (1.43; 1.12–1.81) and/or depression/low mood (1.36; 1.05–1.76) as a primary symptom treated with medical cannabis (see Table 5).

Table 5. Characteristics associated with increasing medical cannabis use during the COVID-19 pandemic in a sample of Canadian residents eligible to increase their medical cannabis use (n = 2622)^{Footnote a}

Characteristic	Bivariable OR (95% CI)	p value	Multivariable OR (95% CI)	p value
Sociodemographic characteristics
GenderFootnote b
Male	1.00 (Reference)	NA	1.00 (Reference)	NA
Female	1.54 (1.27–1.85)	<0.001	1.67 (1.36–2.05)	<0.001
Age
Per 5-year increase	0.85 (0.82–0.88)	<0.001	0.89 (0.85–0.94)	<0.001
Household income, CAD
<40 000	1.20 (0.91–1.60)	0.189	NA	NA
40 000–99 999	1.06 (0.84–1.36)	0.604	NA	NA
100 000–159 999	1.00 (Reference)	NA	NA	NA
≥160 000	0.94 (0.63–1.42)	0.785	NA	NA
Employment
Part-/full-time	1.00 (Reference)	NA	1.00 (Reference)	NA
Unemployed	1.15 (0.91–1.45)	0.235	1.13 (0.87–1.45)	0.361
Retired	0.54 (0.43–0.68)	<0.001	1.10 (0.80–1.51)	0.553
Lost job during the pandemic
No	1.00 (Reference)	NA	1.00 (Reference)	NA
Yes	1.64 (1.28–2.11)	<0.001	1.38 (1.06–1.80)	0.017
Community setting
Urban / suburban	1.00 (Reference)	NA	1.00 (Reference)	NA
Rural	0.80 (0.63–1.01)	0.052	0.81 (0.63–1.03)	0.086
Substance use
Nonmedical cannabis useFootnote c
No	1.00 (Reference)	NA	1.00 (Reference)	NA
Yes	1.76 (1.45–2.14)	<0.001	1.35 (1.09–1.68)	0.007
Alcohol use
No	1.00 (Reference)	NA	1.00 (Reference)	NA
Yes	1.26 (1.04–1.53)	0.016	1.22 (0.99–1.49)	0.061
Tobacco use
No	1.00 (Reference)	NA	1.00 (Reference)	NA
Yes	1.43 (1.13–1.81)	0.004	1.24 (0.96–1.59)	0.105
Unregulated drug use
No	1.00 (Reference)	NA	NA	NA
Yes	1.19 (0.66–2.15)	0.561	NA	NA
Prescription drug use
No	1.00 (Reference)	NA	1.00 (Reference)	NA
Yes	1.21 (1.10–1.47)	0.048	1.24 (1.00–1.54)	0.047
Primary symptoms treated with medical cannabis (yes vs. no)
Pain, inflammation, nerve damage	0.75 (0.61–0.92)	0.007	0.88 (0.70–1.10)	0.254
Sleep problems	1.41 (1.17–1.72)	<0.001	0.94 (0.75–1.17)	0.553
Anxiety, stress, hypersensitivity	2.20 (1.82–2.66)	<0.001	1.43 (1.12–1.81)	0.004
Depression, low mood	2.18 (1.77–2.69)	<0.001	1.36 (1.05–1.76)	0.020
Addiction, dependence, withdrawal	1.74 (0.82–3.70)	0.043	0.83 (0.36–1.92)	0.662
Attention deficit, memory loss	1.85 (1.29–2.64)	0.001	1.09 (0.73–1.64)	0.678
Nausea, appetite loss, gastrointestinal issues	1.88 (1.49–2.38)	<0.001	1.18 (0.89–1.54)	0.246
Seizures, muscle spasms, tremors	1.21 (0.87–1.68)	0.265	NA	NA
Other symptoms	1.20 (0.48–3.03)	0.700	NA	NA

For every 5-year increase in age, the odds of increasing medical cannabis use decreased by approximately 11% (aOR: 0.89; 95% CI: 0.85–0.94; Table 5). All significant findings remained after recoding the outcome for 42 respondents whose only self-reported reason for increasing medical cannabis use was a change to medical cannabis authorization (data not shown).

Discussion

In this study of Canadian residents authorized to use medical cannabis, we sought to examine changes in frequency of medical cannabis use and identify correlates of increasing use during the first two waves of the COVID-19 pandemic. While other Canadian surveys have documented increases in nonmedical cannabis use during the pandemic,^{Footnote 3Footnote 19Footnote 20Footnote 21Footnote 22} to our knowledge this is the first Canadian study to examine changes in authorized medical cannabis use.

Daily medical cannabis use increased by 7 percentage points, from 83.2% pre-pandemic to 90.3% in Wave 2. Given that prevalence of daily use pre-pandemic was already relatively high, the modest increase observed may reflect a ceiling effect. The magnitude of this change is similar to what we recently documented in a web-based sample of people who use cannabis for medical purposes in the United States, in which daily cannabis use increased from 16.2% pre-pandemic to 20.7% in the first few months of the pandemic.^{Footnote 32} Our previous study’s sample comprised people who self-reported using cannabis for therapeutic purposes.^{Footnote 32} We suspect that the current study recorded a substantially higher prevalence of daily medical cannabis use because the sample consisted solely of people authorized to use cannabis for medical purposes.

We also asked about perceived changes in other prescribed and non-prescribed substance use during the pandemic and observed a few differences according to concurrent changes in medical cannabis frequency. We found that significantly more respondents whose medical cannabis use increased also self-reported increases in alcohol and tobacco use during the pandemic. Studies exploring the impact of the COVID-19 pandemic on alcohol consumption suggest that boredom, loneliness, depression, stress and anxiety may be particularly pertinent to upwards shifts in individual usage trajectories.^{Footnote 16Footnote 17Footnote 18Footnote 33Footnote 34} While we did not assess for motives for alcohol or tobacco use, respondents’ reasons for shifting medical cannabis use reveals certain contexts that may trigger coping-oriented substance use, such as anxiety, isolation/loneliness and boredom (time at home). In contrast, approximately 20% of respondents reported reducing their use of alcohol and tobacco regardless of any shift in medical cannabis use frequency, possibly reflecting shared motive-related reductions in alcohol and tobacco use across the increasing, stable and decreasing medical cannabis use groups. This will be an important area to monitor in longer-term evaluations of the impact of the pandemic on the mental health of vulnerable populations. We found no differences for use of prescription opioids, unregulated opioids, non-opioid prescription drugs, crack/cocaine or methamphetamine, but low numbers were recorded for unregulated opioids, crack/cocaine and methamphetamine, reducing power to detect a change in these substances.

Our exploratory multivariable model highlighted several markers of higher susceptibility to increasing medical cannabis use during the pandemic. First, women in our study had approximately 67% higher odds of increasing medical cannabis use than men, consistent with a growing number of studies documenting disproportionate effects of the pandemic on women’s mental health and substance use,^{Footnote 6Footnote 9Footnote 10Footnote 11Footnote 14Footnote 29Footnote 35} Research focussing on sex/gender-based health disparities has shown how reinforcement of gender roles during the pandemic likely contributed to increased stress among women,^{Footnote 36Footnote 37} with increased household and childcare burdens resting disproportionately on women.^{Footnote 38Footnote 39}

In our previous analysis of the US-based sample, the odds of increasing medical cannabis use for anxiety early in the pandemic approximately doubled among women.^{Footnote 32} This current study examined medical cannabis use more broadly; although anxiety was the most common reason cited among those who increased their use (51.3% overall, and 52.3% among women), we cannot attribute the observed increases solely to mental health or stress reasons. More research is needed to further contextualize underlying drivers of sex differences in substance use and mental health changes during the pandemic.

In this current study, respondents who lost their jobs during the pandemic had higher odds of increasing their medical cannabis use. It is plausible that they used cannabis more often out of boredom and reduced work-related responsibilities. However, taking into account evidence linking income loss and material insecurity to increasing alcohol use during the pandemic,^{Footnote 7Footnote 11Footnote 14Footnote 40} increased medical cannabis use in this group may also indicate a coping-related response to emergent stress, anxiety, depression and other health problems exacerbated by income instability.

Survey respondents who reported using cannabis to alleviate symptoms of anxiety and depression were more susceptible to increasing their medical cannabis use during the pandemic, aligning with a previously documented association in an online sample of 1200 self-identified medical cannabis users in the United States.^{Footnote 41} Considering the well-documented effects of the pandemic on mental health more broadly, this finding is not surprising. While survey research involving people who use medical cannabis shows a high level of consensus that cannabis is an effective treatment for anxiety and depression,^{Footnote 42} more experimental research is needed to confirm the therapeutic potential of cannabis in these areas.^{Footnote 43} As our study did not track symptom relief due to cannabis use in response to increased mental health challenges, future research should seek to understand potential long-term therapeutic and/or adverse health outcomes.

For health care providers, the results of this study should serve as a reminder to check in with patient–clients who are using cannabis about potential shifts in their use in response to changes and stresses induced by the pandemic. Of note, given that respondents could report symptom management with medical cannabis for conditions secondary to their primary reason for authorization, clinicians should be aware that usage may have also increased among those who were not initially authorized to use cannabis for anxiety or depression.

We also observed a significant association with increasing medical cannabis use among respondents who were taking concurrent prescription drugs, possibly signifying complex morbidity, heavier disease burden or worsening of disease over time. This indicated susceptibility to increasing intensity of medical cannabis use. Further research will be required to understand whether those taking medical cannabis concurrently with prescription medications were differentially impacted by the COVID-19 pandemic (e.g. through increased symptom flare-ups, managed with cannabis).

Finally, younger age and nonmedical use of cannabis was significantly associated with increasing medical cannabis use during the pandemic, reflecting the high degree of overlap between medical and nonmedical use, particularly among younger people and those using medical cannabis for mental health needs.^{Footnote 24} Although shifting frequency of use of cannabis for medical purposes was the focus of the current analysis, given that one-third of our sample reported at least some use of cannabis for nonmedical purposes (see Table 1), additional research is needed to understand the overlap with medical cannabis changes during the pandemic.

Strengths and limitations

The inclusion of almost 3000 people authorized to access regulated cannabis for medical purposes is a major strength of this study, but our findings should be interpreted in light of certain limitations. First, respondents were a self-selected group of authorized medical cannabis users who were registered with two medical cannabis companies. Although the respondents lived in Canada, generalizability is limited as the sample does not represent all registered medical cannabis users across the provinces and territories.

Second, the pandemic was a shared experience across the whole sample; while respondents could indicate whether they perceived that the pandemic affected their medical cannabis use, we cannot rule out the possibility of observing similar changes in use outside of the pandemic. However, it is worth noting that previous longitudinal studies of medical cannabis use show relatively stable frequency and dosage patterns over 6- to 12-month periods.^{Footnote 44Footnote 45Footnote 46}

Third, the study relied on respondents self-reporting frequency patterns, introducing the possibility of recall deficiencies—particularly for the pre-pandemic and Wave 1 periods. Frequency of medical cannabis use in each period was captured on a categorical scale. As these categories are not linearly equidistant, we created a binary outcome that corresponds with any frequency increase, and this approach does not reflect the magnitude of increase. In addition, incremental increases within each ASSIST category (e.g. from 2 to 4 times per week) could not be measured.

Our use of a predetermined set of reasons for changing frequency of medical cannabis use during the COVID-19 pandemic may have missed some important factors, including possible shifts to nonmedical sources for ease of accessibly; however, we tried to mitigate these oversights by including a free-form text box for participants to add more information if they thought their reasons were not accurately represented by the pre-determined categories. Not all respondents for whom we detected a change in medical cannabis use frequency during the pandemic self-perceived a change in their use. Accordingly, 28% (n = 155) of those who increased and 40% (n = 49) of those who decreased medical cannabis use did not provide a reason for the change (see Table 3). The reasons provided, expressed as the proportion of respondents who increased or decreased their medical cannabis use frequency, may be underestimated and are only meant to generate hypotheses, not confirm underlying connections. Nonmedical use of cannabis was not captured in the question assessing secondary substance shifts since the onset of the pandemic. Nonmedical use of cannabis deserves detailed consideration in future research of pandemic-concurrent trends in this population.

Finally, the current findings represent self-reported changes that occurred early in the pandemic. Further research is needed to characterize ongoing shifts in medical cannabis use later in the pandemic and associated long-term impacts.

Conclusions

We documented modest but significant and persistent increases in the frequency of medical cannabis use early in the COVID-19 pandemic in this sample of authorized medical cannabis users in Canada. We observed increased use of alcohol and tobacco more often among those who increased their medical cannabis use frequency. Women, younger people, those who lost employment during the pandemic, those who used nonmedical cannabis and prescription drugs and those who used medical cannabis to manage depression and anxiety had greater odds of increasing their medical cannabis use during the pandemic.

Our findings highlight a subset of the people authorized to use medical cannabis who may be particularly vulnerable to increased substance use and adverse mental health outcomes during the pandemic, underscoring the need for clinicians to check in with patients who use medical cannabis.

Acknowledgements

The authors would like to thank the respondents who shared their time and experiences with us via the Canadian Cannabis Patient Survey; Tilray for providing the primary funding for the study; and Broadstreet Health Economics & Outcomes Research for the initial data cleaning, coding and analysis.

Portions of this study were funded by National Institute on Drug Abuse grant DA047296 and the Semel Charitable Foundation.

Conflicts of interest

The Canadian Cannabis Patient Survey was sponsored by Tilray, a licensed producer of medical cannabis. At the time of data collection, KO was Clinical Research Manager and PL was Vice-President, Global Patient Research and Access at Tilray. Over the past 12 months, ZDC reports receiving honorariums from Canopy Growth Corporation.

Authors’ contributions and statement

SL – Conceptualization, Investigation, Methodology, Formal analysis, Data curation, Writing – Original draft

ZDC – Methodology, Writing – Review and Editing

KO – Software, Resources, Project Administration, Writing – Review and Editing

PL – Conceptualization, Investigation, Methodology, Administration, Writing – Review and Editing

The content and views expressed in this article are those of the authors and do not necessarily reflect those of the Government of Canada.

Previous | Table of Contents | Next