Quality of Life Among Patients Using Cannabis to Manage Anxiety: A Longitudinal Observational Study

Introduction

Cannabis products are increasingly being used to manage mental health conditions.¹ Products derived from cannabis may have therapeutic potential for managing anxiety in patients.^2,3 For example, studies have shown that cannabidiol (CBD), one of the most studied chemicals derived from the cannabis plant, has potential anxiolytic properties.^4-6 In fact, approximately 50% of individuals using medical cannabis report doing so to help manage anxiety.^7,8

Researchers have begun to examine the effects of cannabis on health-related quality of life (HRQoL) in patients with anxiety. HRQoL is a broad construct that is defined as a person’s perceived health status, well-being, and functional status.⁹ A patient’s perception of the impact that medical treatments have on their HRQoL has been recognized as an important aspect of treatment outcomes.^10,11 One well-established approach to measuring HRQoL suggests that there are several distinct facets of HRQoL, including general health, physical functioning, physical and emotional role functioning, bodily pain, vitality, social functioning, and overall mental health.¹²

Studies have suggested that HRQoL is lower among people who are considered heavy cannabis users compared with those who do not use cannabis. A recent empirical synthesis and analysis of 14 studies of recreational cannabis use suggested that heavy recreational users or individuals with a cannabis-use disorder tended to report lower HRQoL than low-level or non-users.¹³

These results suggest a lower HRQoL in heavy users of recreational cannabis compared with non-users. It is perhaps not surprising that patients who take medical cannabis for health-related issues report a lower HRQoL than those who do not seek such treatment. It is important to note, however, that the latter may not have a preexisting condition that adversely affects HRQoL. Importantly, this research does not address the effect of cannabis use on HRQoL, underscoring the need for longitudinal and experimental studies to provide further insight. 

A systematic review and meta-analysis by Goldenberg et al.¹⁴ sought to determine the effects of specific cannabinoids or cannabis on the HRQoL of patients. Among the 20 studies that met the selection criteria, 11 randomized controlled trials, together with cross-sectional and correlational studies, tested the effects of cannabinoids or cannabis on HRQoL. Results of these studies were mixed: Although use of cannabis and cannabinoids was found to have no overall effect on HRQoL, some of the study participants who used cannabinoids for multiple sclerosis, pain, and inflammatory bowel disorders reported small improvements in HRQoL.

Moreover, participants who used cannabinoids for HIV-related symptoms reported small reductions in HRQoL.¹⁴ Similarly, evidence from longitudinal studies also yielded mixed findings, with several suggesting no association between cannabis use and HRQoL.^15-17 These studies focused, in part, on how cannabis-use patterns in general relate to subsequent HRQoL, but they did not examine patterns of HRQoL in patients who used cannabis and cannabinoids to manage specific health issues.

Only one published study to date by Lev-Ran et al.,¹⁸ which analyzed data from Wave 1 of the National Epidemiological Survey of Alcohol and Related Conditions,¹⁹ sought to assess the effect of HRQoL on patients with anxiety disorders who used cannabis compared with individuals with anxiety disorders who were non-users. Patients who used cannabis at least once per week were considered regular users, and those who used cannabis less than once per week were considered occasional users.¹⁸

The regular cannabis users reported significantly worse mental HRQoL than non-users. The reported HRQoL of the occasional cannabis users did not differ from regular users or non-users. This study by Lev-Ran et al. significantly adds to the limited knowledge of the impact of cannabis on HRQoL. However, study limitations constrain the degree to which it informs our understanding of the effects of cannabis when it is taken to manage anxiety. Participants in the study did not describe their reasons for taking cannabis.

Therefore, this cross-sectional study leaves open the possibility that more severe anxiety psychopathology led to increased cannabis use, and regular cannabis users were more likely to meet criteria for comorbid problems than non-users. More severe anxiety and elevated comorbidity—in contrast to cannabis use per se—could account for the lower HRQoL observed in this study.¹⁸

The present study aimed to expand on previous research by examining changes in self-reported HRQoL among patients using medical cannabis to manage anxiety symptoms over the course of approximately 1 year. We hypothesized that mental HRQoL would improve over time in patients who used cannabis to alleviate anxiety. Exploratory analyses also were conducted to examine changes in self-reported physical HRQoL across time.

Methods

Study Design

This was a prospective, open-label, observational study conducted between April 2015 and April 2019. Patients from several Canadian specialized medical cannabis clinics were recruited by their physicians to participate. Upon initiation of the study, all the participants were required to register online through a digital platform created by DSG, Inc. At each participating clinic, enrolled participants were given instructions on how to create an ePRO portal account and were required to enter all baseline data in the presence of their physician.

Follow-up data were self-reported electronically on a monthly basis for up to 1 year. The study was approved by Quorum Review Institutional Review Board, an independent ethics review board, and was conducted in accordance with research regulations, applicable international standards of Good Clinical Practice, institutional research policies and procedures, and with the ethical principles laid out in the Declaration of Helsinki. 

Study Sample

The target population of the present study consisted of Canadian adults who incorporate medical cannabis into their treatment strategy for anxiety. For this study, medical cannabis is defined as physician-authorized patient purchases of cannabis (dried flower or ingestible oil format with varying cannabinoid content) from Canopy Growth Corporation. Potential participants were eligible if they met the following criteria:

1. ≥18 years of age; 
2. Authorized for treatment with medical cannabis; 
3. Sufficiently comprehended English or French; 
4. Had sufficient cognitive ability to understand the questionnaires (based on physician assessment); 
5. Had internet access; and 
6. Provided a signed informed consent form. 

The study excluded participants who were pregnant or who were employed at a licensed producer of cannabis. The study’s treating physicians received a reimbursement of up to $300 CAD per year, and patients received a 10% discount on medical cannabis while participating in the study. During the study period, 769 patients were approved for cannabis treatment across 10 participating specialized medical cannabis clinics.

This longitudinal observational study enrolled participants who took medical cannabis for a wide array of reasons. For inclusion in the present study, participants from this larger sample reported using medical cannabis primarily to reduce anxiety symptoms and completed an online assessment at least once during each of the following time points: baseline (T1), after 4 to 6 months of treatment (T2), and after 9 to 12 months of treatment (T3). 

Procedure

Treatment Regimen 

All participants underwent a clinical evaluation by a physician, which included a review of their medical history and health status. Medical cannabis authorization was accompanied by a personalized treatment plan, detailing specific product name, dosing instructions, and route of administration.
To assist with treatment planning, literature support and recommendations regarding specific indications^20,21 were provided to some physicians based on their level of experience in authorizing medical cannabis. 

Data Collection

DSG, Inc. monitored all captured data via an ePRO portal, which was accessible by secure email and password login. All data entered into the portal received a time stamp. The following data were collected at baseline (ie, before patients commenced treatment with cannabis): patient demographics, completed patient–physician interview questionnaire, completed cannabis regimen questionnaire, and a completed 36-Item Short Form Health Survey (SF-36).

Thereafter, participants were instructed to use the ePRO portal to provide self-report data (eg, SF-36 questionnaire) every 30 days. Additionally, physicians encouraged patients to report any potential adverse events (AEs) as soon as they occurred either by logging into the ePRO portal or by directly contacting Canopy Growth’s Customer Care Division.  

Measures

Baseline Characteristics 

At baseline, we recorded standard demographic information on all participants; their previous experience with cannabis; and their medical history, including clinical diagnosis, and primary and secondary symptoms for which they were using medical cannabis. 

SF-36 Health Survey, Version 2

We used version 2 of the SF-36 Health Survey, a self-report questionnaire, to evaluate and collect HRQoL data at each follow-up. Standardized scores for each of the 8 subscales were calculated as recommended.²² The transformed scale scores yielded a mean of 50 and an SD of 10, with higher scores reflecting better health. Version 2 of the SF-36 is a well-established measure of HRQoL with good psychometric properties. 

To test our primary hypotheses, we examined the following 4 subscales within a mental HRQoL component: 

1. Social functioning (2 items: eg, “During the past 4 weeks, how much of the time has your physical health or emotional problems interfered with your social activities [like visiting with friends, relatives, etc]?”); 
2. General mental health (5 items: eg, “Have you been happy?”); 
3. Role limitations due to emotional problems (3 items: eg, “Have you accomplished less than you would like?”); and 
4. Vitality (4 items: eg, “Did you feel full of life?”). 

Exploratory analyses employed the following 4 subscales within a physical HRQoL component: 

1. Physical functioning (10 items: eg, “vigorous activities, such as running, lifting heavy objects, participating in strenuous sports”); 
2. Role limitations due to physical problems (4 items: eg, “cut down the amount of time you spent on work or other activities”); 
3. Bodily pain (2 items: 1 past-4-week pain rating and 1 work interference rating); and 
4. General health perceptions (5 items: eg, “my health is excellent”).

Safety

Study participants were instructed to report all AEs through the ePRO portal, which were then reported to the Customer Care Division; alternatively, participants could report AEs directly to Canopy Growth’s Customer Care Division, which follows strict standard operating procedures complying with the requirements of Canada’s Access to Cannabis for Medical Purposes Regulations (SOR/2016-230). 

Analytic Approach

Descriptive and inferential statistics were calculated using SPSS version 24 (IBM, Armonk, NY, USA). AEs were coded using MedDRA system organ class, and descriptive statistics were applied. Repeated measures of analysis of variance (ANOVA) using the Greenhouse-Geisser correction with SF-36 subscale scores at T1, T2, and T3 (entered as a repeated measure) were conducted to determine if change over time differed from zero. Partial eta squared (ƞp²) was used to index effect size. When an ANOVA suggested significant change over time, we controlled for simple effects by conducting paired-sample t tests to identify significant differences between measurement points, which generated a 95% CI. Significance tests were 2-tailed and adopted an α of 0.05. 

Results

Of the 769 study participants who met the screening criteria, 60 reported anxiety as their primary reason for seeking medical cannabis; these participants also completed an assessment during each of the specified time points. As a result of these inclusion criteria, analyses were conducted with a sample size of 60 participants. Figure 1 presents a Consolidated Standards of Reporting Trials flowchart.

Descriptive Information

Table 1 shows the participants’ demographic characteristics. Most participants were cannabis-experienced (75%) and men (58%), and the average age of participants was 45 years (SD = 12.1 years). There were substantial missing data for patients’ clinical diagnoses (75%). Data that were available included the following diagnoses among 15 participants: fibromyalgia (7%), chronic pain (5%), anxiety disorder (5%), cancer (2%), Sjögren’s syndrome (2%), irritable bowel syndrome (2%), and hypothyroidism (2%). 

Adverse Events

In the cohort of patients who used cannabis primarily for anxiety, 2 participants reported experiencing non-serious AEs: drowsiness, hunger, and upset stomach. No serious AEs were reported during the study period (April 2015–April 2019). 

Mental Health-Related Quality of Life

Figure 2 displays the results for the SF-36 mental health domains and Table 1 shows the results from analyses. Vitality subscale scores significantly increased over time. Improvements in the vitality domain from T1 to T2 were approaching traditional levels of statistical significance [t(59) = –1.91; 95% CI, –5.20 to 0.11; P=0.06]. Significant improvements in vitality were detected from T1 to T3 [t(59) = –2.54; 95% CI, –4.65 to –0.55; P<0.05]. Significant changes over time were not observed in social functioning, role functioning due to emotional problems, or mental health subscales.   

Physical Health-Related Quality of Life

Figure 3 shows the results for the SF-36 physical health domains and Table 2 presents the details from the analyses. There was a significant change in physical functioning, with a decrease in score across repeated assessments. Physical functioning did not differ from T1 to T2 [t(59) = 0.12; 95% CI, –1.93 to 2.18; P>0.05]. The decline from T1 to T3 approached statistical significance [t(59) = 1.98; 95% CI, –0.02 to 5.52; P=0.052]. There was no statistically significant change in role functioning due to physical problems, bodily pain, or general health.   

Discussion

Evidence is accumulating for medical cannabis as a treatment for the management of anxiety.^2,4,8 However, little information exists regarding the effects of medical cannabis on HRQoL when it is used to manage anxiety. In the present longitudinal observational study, we examined changes in self-reported mental HRQoL over the course of 1 year among Canadian adults using medical cannabis primarily to manage anxiety symptoms. We found significant improvements in vitality, with no change in other aspects of mental HRQoL. We also noted a significant decline in self-reported physical functioning during the study. 

Results across mental health domains were only partially consistent with our hypotheses. Contrary to expectations, there were no changes in social functioning, role functioning due to emotional problems, or general mental health. Figure 2 illustrates minor trends toward improvement in role functioning related to emotional health, as well as general mental health. It is possible that examining these patterns using a larger sample may detect these small improvements over time. By the end of the study, reported trends showed a worsening of social functioning among participants. None of these trends, however, were statistically significant.

As such, no inferences can be made regarding possible effects of cannabis on these aspects of HRQoL. As real-world evidence continues to accumulate, tests of these trends in larger studies should be conducted to determine reliable, small effects on these mental HRQoL domains.

The present study demonstrated a significant improvement in vitality, although the magnitude of the observed improvement was small.²³ Improvements in vitality reflect increases in energy and reductions in fatigue. Vitality was measured in the current study with a 4-item subscale of the SF-36, with questions focusing on feeling full of life, having a lot of energy, feeling worn out, and feeling tired.¹² The methods used in this study did not allow us to test the mechanisms underlying these improvements. Vitality may have improved as a result of reductions in anxiety, which is consistent with literature suggesting such effects of cannabis.^5,6

Although not directly tested in the present study, it also is possible that changes in vitality were attributable to improvements in sleep resulting from cannabis use, as suggested by preliminary research.²⁴ These possible explanations are not mutually exclusive, given well-established links between anxiety and sleep.²⁵ The change in vitality also could be a direct result of the effects of cannabinoids on neurobiological systems underlying vitality. For example, cannabinoids may improve energy and motivation associated with the endocannabinoid system’s modification of hedonic tone through its effect on the prefrontal cortex.²⁶

Future research should replicate these findings and further examine any possible causal relationship between cannabis use and vitality to enhance insight into this observation. For example, direct measurement and mediational analysis of fatigue, sleep parameters, and anxiety may help explain cannabis use-related changes in vitality. 

Research has typically examined correlations between cannabis use and general HRQoL.¹³ In contrast, we examined prospective correlations between the effects of cannabis when used specifically for anxiety symptoms and changes in HRQoL across time. With respect to mental HRQoL, use of cannabis products to manage anxiety symptoms was related to improvements in vitality. This contrasts with cross-sectional research demonstrating a link between regular cannabis use and lower HRQoL among people with anxiety disorders.¹⁸

Our study employed a within-subjects design, which provided added information to the group comparisons examined previously. Although a sample of regular cannabis users with anxiety disorders reported lower HRQoL than both occasional and non-users,¹⁸ individuals taking cannabis for anxiety may experience improvements (albeit limited) in mental HRQoL. Notably, this pattern aligns with multiple studies that have documented improvements in certain HRQoL domains when using cannabis products for other specific conditions, including neuropathic pain and inflammatory bowel disease.^27,28 

The present study suggested a decrease in self-reported physical functioning primarily during the second half of the observational period, with no change in other physical HRQoL. The magnitude of decline in physical functioning was small, and the finding requires replication and further study to understand why such change may have occurred. It is possible that cannabis use caused problems with physical functioning, although relatively few physical health problems have been linked to cannabis use.²⁹

Alternatively, participants may have been experiencing progression of physical problems during the study. Indeed, several participants reported the presence of physical health problems, such as fibromyalgia (7%), chronic pain (5%), cancer (2%), Sjögren’s syndrome (2%), irritable bowel syndrome (2%), and hypothyroidism (2%). Therefore, it remains possible that the progression of physical health problems accounted for declines in physical functioning-related QoL. 

Methodological challenges inherent to studying naturalistic use of cannabis products limit direct comparison of the current findings to previous studies. One such challenge is the substantial variability across cannabis products. Medical cannabis preparations that include several cannabinoids and terpenes may have greater therapeutic effects and impacts on HRQoL compared with preparations that include only a single cannabinoid.¹⁴ 

The present study did not track which specific cannabis products were used by patients, thereby introducing possible significant variability in dosage, cannabinoid content, and method of administration (eg, oral vs inhaled). Additional research is needed to examine whether the changes in HRQoL observed in this study follow use of a particular cannabis formulation (eg, CBD-dominant oral formulations) or dose strength. For example, dosing and tolerance of medical cannabis depends on prior patient experience with cannabinoids, along with other factors such as patient’s body mass index, concomitant medication use, and daily routine.²⁰

Several studies have observed anxiolytic effects of doses of 300 to 400 mg of CBD for anxiety disorders.^30,31 Fusar-Poli et al.³ observed anxiolytic effects in healthy volunteers who were administered 10 mg delta-9-tetrahydrocannabinol:600 mg CBD of ingestible cannabis. Further studies are needed to understand the anxiolytic effects of cannabis in real-world settings and to determine the degree to which medical cannabis not only aligns with doses administered in controlled research studies but also whether a link exists between dose strength and HRQoL.³

Similarly, identifying the effects of recreational cannabis on anxiety and mental HRQoL is difficult due to the wide variance in cannabinoid content among products.¹³ Designs that carefully measure dose of specific cannabinoids are needed to determine the impact of medical cannabis use on HRQoL. 

Study Limitations

The present study had several limitations, in addition to those previously discussed, which warrant careful consideration. First, because the SF-36 is a self-report tool, there is the potential for responder bias. In combination with the absence of a control group, it is possible that demand characteristics may have influenced the study, resulting in biased responses of endorsed improvements.

The lack of improvement across most subscales is inconsistent with a general response bias, but assessment strategies that go beyond reliance on questionnaires are needed. Although the naturalistic nature of the study offered important insight into outcomes among actual patients engaging in self-selected use patterns, it also resulted in lack of control over sample characteristics. Participants in the present study had to have access to the internet, were required to speak either English or French, and had to be able to respond to questionnaires.

Moreover, most of the participants in the study were experienced cannabis users. Accordingly, the degree to which our results generalize to the broader population of individuals using medical cannabis to manage anxiety requires additional testing. 

The study’s small sample size also limited the degree to which more detailed hypotheses could be examined. For example, there is some evidence that among individuals with anxiety disorders who use cannabis, women report lower mental HRQoL than men.¹⁸ Larger datasets are needed to examine whether the effects of cannabis use for the management of anxiety are sex-dependent and play a role in predicting QoL.

Conclusion

The present study extends previous research on associations between cannabis use and HRQoL by employing a 12-month naturalistic follow-up period to examine the use of medical cannabis specifically for managing anxiety. Preliminary findings indicate self-reported improvements in vitality and declines in physical functioning.

Future investigations with larger sample sizes should expand on these preliminary results to more closely track (or even control) the specific cannabis products used to manage anxiety. In theory, it remains likely that if medical cannabis does alleviate anxiety, these anxiolytic effects will lead to a concomitant improvement in mental HRQoL.