Chronic pain is a widespread health issue broadly defined as pain that lasts or recurs for more than 3 months [27] and is categorized as chronic primary pain—a disease in its own right such as nonspecific low-back pain—or as chronic secondary pain initiated as a symptom of an underlying disease, such as cancer-related pain or neuropathic pain [28]. In Australia, more than 15% of the adult population lives with persistent chronic pain or recurring pain lasting longer than 6 months [29]. In the United Kingdom, as many as 35% of adults experience some level of chronic pain lasting more than 3 months, with more than 10% reporting moderate to severely limiting chronic pain [30]. In 2016, the Centers for Disease Control and Prevention reported that 20% of US adults had chronic pain and 8% experienced chronic pain that severely interfered with daily functioning on most days or every day for a 6-month period [31].

Cancer-related pain is experienced by approximately one-third of patients with cancer at diagnosis and during treatment and by approximately three-fourths of patients with advanced-stage cancer. However, 10%-15% of the patients are nonresponsive to conventional pain therapy [32]. Residual tissue damage from cancer and cancer treatment often results in chronic pain in cancer survivors, which lasts many years after treatment [33].

Neuropathic pain, which is caused by a lesion or disease of the somatosensory nervous system [34], is experienced by approximately 8% of the population [35] and approximately 86% of the patients with MS [36]. Neuropathic pain is associated with higher rates of unemployment [35], poor physical, psychological, and social functioning, significantly impaired overall HRQL and sleep, and higher depression and anxiety than those with other chronic pain and those without pain [37]. Less than 35% of the patients with neuropathic pain respond to conventional therapy [38], whereas others often receive incomplete pain relief along with conventional treatment-related side effects [39].

The prevalence of chronic pain increases dramatically in patients receiving palliative care [40,41]. A study of patients in palliative care clinics in the United States found that most of them were admitted with unrelieved pain and that chronic pain assessment and management were inadequate [42]. In addition, one-third of the patients receiving palliative care who experienced pain were significantly more likely to also suffer from depression [41], report insufficient sleep [43], and were at a higher risk of opioid misuse [40]. Effective pain and symptom management aimed at reducing suffering and improving overall HRQL are the primary goals of palliative care.

MS is a chronic inflammatory and demyelinating neurodegenerative disorder that often involves symptoms such as spasms, tremors, pain, fatigue, bladder dysfunction, cognitive impairment, depression, and impairments in swallowing, speech, vision, and balance [44]. In 2018, MS Research Australia reported an estimated 24,600 Australians with MS, and, on average, their overall HRQL was 31% less than the Australian population norm (measured by the health state utility valuation) [45]. As a currently incurable and often progressive condition, treatment and management largely focus on improving the quality of everyday life by relieving symptoms [44]. A systematic review of reviews conducted in 2018 on the effects of cannabidiol, Δ9-tetrahydrocannabinol, or cannabidiol and Δ9-tetrahydrocannabinol formulations in treating MS symptoms found sufficient evidence supporting cannabinoids in relieving both pain and spasticity symptoms [46].

Epilepsy is a chronic neurological disease characterized by 2 or more unprovoked seizures and affects approximately 50 million people worldwide [47]. High-quality evidence from randomized clinical trials suggests that cannabidiol reduces seizure frequency; however, further examination of PROs is needed to assess whether cannabidiol interacts with other antiseizure medications to produce unwanted side effects [48].

The primary objective of this study is to describe changes in the PROs (HRQL, pain, fatigue, sleep, anxiety, and depression) from baseline to 3 months for a large cohort receiving medicinal cannabis.

The secondary objectives of this study are to describe changes in PROS (HRQL, pain, fatigue, sleep, anxiety, depression, and physical functioning) from baseline up to 12 months and to describe differences between patients accessing medicinal cannabis with different chronic health conditions, including, but not limited to, chronic pain, MS, epilepsy, Parkinson disease, and cancer.

The exploratory objectives of this study are to explore (1) which individuals are more likely to have lower symptom burden and greater HRQL, (2) associations among PROs, with the hypothesis that a high symptom burden is associated with poorer HRQL, and (3) associations between PROs and resource and medication use over time, with the hypothesis that lower symptom burden is associated with reduced health care–resource use and reduced use of opioids and other prescribed medications for managing symptoms.

The study includes 3 hypotheses: (1) PROs will improve from baseline to 3 months in patients accessing medicinal cannabis, (2) improvements in PROs at 3 months will be maintained over a 12-month period, and (3) no differences in PROs will be observed between patients being treated for different conditions (eg, chronic pain, MS, epilepsy, Parkinson disease, or cancer).

Our aim is to recruit a large, broad, and representative sample of medicinal cannabis users. Therefore, we will invite every eligible patient treated at each participating center during a 12-month recruitment period. This large real-world cohort will enable several important analyses exploring differences in PROs between disease groups commonly treated with medicinal cannabis, as discussed in the Objectives section.

Following the guidelines [51] for the European Organization for the Research and Treatment of Cancer Quality of Life Questionnaire-30 (EORTC QLQ-C30) [52] and allowing for a 20% loss to follow up, a baseline sample size of 2142 is required with a minimum follow-up of 3 months. This sample size provides 95% power to detect the smallest effect size threshold of 0.1 for the insomnia domain of the QLQ-C30, using a 2-tailed significance level of 1% [53].

Recruitment will take place between November 2020 and November 2021, with the aim of including all eligible patients receiving medicinal cannabis at participating sites during the recruitment period. Figure 1 provides an overview of the patient recruitment and data collection procedures.

LGP, following the World Health Organization Guidelines on Good Agricultural and Collection Practices for Medicinal Plants and European Union-Goods Manufacturing Practices standards, provides its medicinal cannabis products in Australia under the SAS. As part of this process, LGP engages with medical cannabis–focused clinics, authorized prescribers, and other health care professionals prescribing LGP products and will be responsible for identifying recruitment sites for this study. Advertising and information relevant to the study will be disseminated via a dedicated study website and social media platforms. Content on the study website will inform potential participants and recruitment sites of the study objectives, terms, and conditions, including eligibility criteria, and who to contact for more information. Approved recruitment sites will receive study-specific training from the University of Sydney researchers regarding participant screening and recruitment.

Patients from participating centers who meet the eligibility criteria will be invited to participate in the study. Patients will be identified by physicians at recruitment sites and approached to participate by either the physician or site staff. A record of those identified as eligible and invited to participate will be logged in the web-based research database with the patient’s verbal consent. Physicians at recruitment sites will use a generic link to access the study database to create a new record for each patient screened.

The physician or staff at recruitment sites will ask if the patient agrees to have the Participant Information Statement and invitation sent to them through email. Patients will be informed that the process involves recruitment site staff entering the information, as per the Registration and Clinical Data section, into the research database, which then sends an invitation to them automatically. Participants will be informed that email addresses are stored within the Research Electronic Data Capture (REDCap) system and used solely for sending reminders to complete questionnaires at scheduled time points. REDCap [54] is a secure web-based application developed by Vanderbilt University that runs on the servers of the University of Sydney, ensuring that the data stay within the Sydney University data center.

All patients who provide their email address to receive an invitation for the study will receive their medicinal cannabis product at a standardized cost. As medicinal cannabis is still an unregistered product, it has been difficult to control the cost of the product; consequently, there has been considerable variability in what people pay for the product. LGP has partnered with several pharmacies across Australia to ensure that all participants taking part in this study will be charged the same amount for their product, eliminating variability in out-of-pocket costs and enabling a health economic evaluation.

As soon as an email address is entered in the web-based database, the patient will receive an email invitation with a link to the web-based forms for their record. The email will include the Participant Information Statement (Multimedia Appendix 1) to be considered before giving consent on the web. The Participant Information Statement contains detailed information about the rationale, design, and personal implications of the study. Patients will then provide their consent to join the study by checking the consent box before being able to proceed to the data collection forms or they can opt out of study participation at this point. They have as much time as they need to consider their participation. The patient’s right to refuse consent without giving reasons will be respected. If the patient does not respond to the invitation, 2 daily follow-up reminders will be sent via email, after which the system will automatically record the patient as having selected opt out without reason.

Participants will remain free to withdraw from the study at any time without giving reasons and without prejudicing any further treatment. Participants can withdraw responses from the study before the data have been analyzed; otherwise, they will be included.

When physicians at recruitment sites complete the web-based screening form, REDCap will automatically generate a number to be used as the participant’s study ID number for study registration.

Study data will be recorded by physicians at recruitment sites on case report forms and by participants in the questionnaire booklets. These will be completed on the web through the University of Sydney research data capture system, REDCap [54].

Screening and registration and clinical data will be completed by the physician at the recruitment sites. The REDCap database will collect information only identifiable by REDCap-assigned study record ID numbers. Physicians at recruitment sites will maintain a record of participants’ study ID numbers for each screened and registered participant entered. Where required, data can be updated for individual participants by notifying the study project manager of the corresponding study ID numbers (eg, to record participant withdrawal). If 2 consecutive assessments are missed by participants, the study project manager will contact the physician at the recruitment site to determine the reasons for missed assessments, if known.

The following patient screening details will be entered into the REDCap web-based registration form by physicians at recruitment sites: age, sex, country, email address, health indications for accessing medicinal cannabis, neuropathic pain screening using the short-form Douleur Neuropathique en 4 Questions [55,56] (if pain is selected as a health indication), duration of pain (if pain is recorded as a health indication), comorbidities, medicinal cannabis type and dosage, date and dose of any previously prescribed medicinal cannabis, recruitment physician ID, and reasons for declining study participation (if applicable).

Patients who provide email addresses will receive an email with a link to the patient consent and demographic form corresponding to their study record ID. Patient consent and demographic questions include the following: consent to participate (or opt out), reasons for declining study participation (if applicable), ethnicity and cultural background, education, living arrangements, marital status, height, weight, gender identity, work status, access to health services, any medication other than medicinal cannabis taken during the last 4 weeks for health indication, and previous history of cannabis use.

Baseline PROMs will be presented to participants on the web automatically after completing the demographic questions. Participants self-complete the questionnaires through the web-based platform at home, accessible on a computer or other device with an internet connection, depending on their preference. All the questionnaires will be administered in the same order. It is anticipated that completion of baseline questionnaires may take up to 30 minutes. Follow-up questionnaires may take approximately 25 minutes. We have estimated the time to complete the questionnaires (including demographic questions) based on 10-12 seconds per item [57].

All participants will complete the following PROMs.

The following questionnaires will only be administered to patients with identified conditions or health status.

Participants will receive automatic reminders from the REDCap system to their email addresses at scheduled follow-up assessment time points (Table 1). Follow-up questionnaires can be completed using computers or mobile devices depending on their preference. Up to 2 email reminders to complete the follow-up questionnaires will be sent within the assessment time windows (Table 1). The following questions will be added to the front page of the follow-up questionnaires: current cannabis product and dose, any reduction in other medications taken for health indication because of using medicinal cannabis (including brand, strength, and dose), and work status.

Prospective assessment of newly prescribed patients before and after treatment is required to assess changes in PROs over time. PROs will be completed at baseline before starting medicinal cannabis, at 2-3 weeks after starting medicinal cannabis (end of titration period), and then at 1, 2, 3, 5, 7, 9, and 12 months after the titration period.

The acceptable PRO assessment time windows are indicated in Table 1.

All analyses will be performed using SPSS software (version 26.0; IBM Corp). Baseline demographic and clinical data will be summarized descriptively for all patients registered for the study. Categorical data will be presented as frequencies and percentages. For continuous scale data, mean, SD, median, 25th and 75th percentiles, and minimum and maximum scores will be presented.

PRO analyses will explore changes over time using mixed linear models. Subgroup analyses will compare differences in PROs between underlying conditions, dose, and duration of pain and over time using linear mixed models.

A comprehensive PRO-specific statistical analysis plan will be produced by the study statistician. The key considerations include the following:

Missing items within the PRO questionnaires are not expected. This is because of the web-based administration platform alerting participants of missed items and the requirement to complete those items before progressing to the next page. Only those participants who complete at least 2 questionnaires (baseline and one other) will be included in the analysis. Single missed assessments will be imputed using the last value carried forward technique, that is, no change from that individual’s last assessment. Pattern mixture models will be used to impute scores for missed assessments based on recorded reasons [84].

The economic evaluation will use collected data around pharmaceutical and other medical costs to explore the drivers of patient-level costs. As this study is not a comparative randomized trial, we are not proposing to conduct a formal economic evaluation, resulting in a cost per quality-adjusted life year. We will instead use baseline resource use as an indicator of typical care and contrast resource use throughout the study with baseline data. We will explore the relationship between HRQL and resource use across the cohort, which is potentially important information for future economic evaluation of medicinal cannabis.