Persistence of use of prescribed cannabinoid medicines in Manitoba, Canada: a population-based cohort study.

BACKGROUND AND AIMS: To estimate prevalence of continuous use (persistence) of prescribed cannabinoid medications for up to 1 year from initial prescription in Manitoba, Canada and predictors of duration of use.

DESIGN AND SETTING: A retrospective, population-based, cohort study using administrative data from the Manitoba Population Research Data Repository located at the Manitoba Centre for Health Policy, Canada.

PARTICIPANTS: People without a record of a previous prescription who were prescribed a cannabinoid medication from 1 April 2004 to 1 April 2016 followed for 1 year from the date of first prescription.

MEASUREMENTS: Continuous prescribed cannabinoid medication use was defined as use without a gap exceeding 60 days between prescriptions. The primary outcome was prevalence of continuous prescribed cannabinoid medication use for up to 1 year. A secondary outcome was duration of continuous use. Predictors were socio-demographic characteristics, medical diagnoses and type of cannabinoid medication.

FINDINGS: Among 5452 new users, 18.1% [95% confidence interval (CI) = 17.08-19.12] were still using cannabinoids at 1 year. Median duration of use was 31 days [interquartile range (IQR) = 25-193]. This was highest for nabilone (33 days, IQR = 25-199) and lowest for nabiximols (20 days, IQR = 7-30). Use was longest among 19-45- and 46-64-year-old users and those with the highest socio-economic status. Fibromyalgia [hazard ratio (HR) = 0.89, 95% CI = 0.84-0.95], osteoarthritis (HR = 0.91, 95% CI = 0.82-0.97) and substance use disorder (HR = 0.85, 95% CI = 0.76-0.94) diagnoses were associated with longer use (HR for discontinuation-HR < 1 less discontinuation and longer use). A diagnosis of cancer was associated with shorter use (HR = 2.73, 95% CI = 2.02-3.67).

CONCLUSIONS: In Manitoba, Canada approximately 18% of people prescribed cannabinoid medication continue using for at least 1 year. Duration of use varies with type of cannabinoid medication, age, socio-economic status and dagnosis.