Background and purpose
Denmark launched a pilot program of medical cannabis in January 2018. The aim was to establish whether medical cannabis and cannabis-based medicine (MC/CBM) were superior and safe compared to conventional treatment, regardless of the indications for which people received such medication.

Materials and methods
People (cases) were identified who had redeemed at least one prescription of MC/CBM according to the nationwide, unselected Danish registers. These were propensity-score matched to controls with the same indications who had not used MC/CBM. Potential participants were contacted electronically, and if willing to participate filled in various survey instruments online. Participants were also interviewed in person in order to investigate symptoms of depression, anxiety, and to assess cognitive levels. Different sets of analyses were conducted, handling potential confounders in different ways.

Results
In the primary analyses, cases were more satisfied with their treatment than were controls (mean (SD) 29.2 (4.8) versus 26.5 (4.5) on the CSQ, p = 0.006), and scored lower on depression (3.3 (3.0) versus 4.6 (2.9), p = 0.03). Cases reported higher levels of pain than controls when measured on the SF-36 bodily-pain subdomain (36.3 (23.0) versus 48.7 (30.1), p = 0.01). There were indications of worse symptoms of multiple sclerosis in cases compared to controls. Reported side-effects were generally mild.

Conclusion
Both potential effects and harms of MC/CBM were observed. Randomized trials are required to establish if these are true effects and harms, or due to confounding by indication.