Face validity of the ICD‐10 criteria of substance abuse and dependence for patients prescribed cannabis‐based medicines for chronic pain—A survey of pain medicine physicians in Canada, Germany and Israel

A major concern with cannabis-based medicines (CbM) and medical cannabis (MC) is the risk of abuse and dependence. The face validity of the International Classification of Diseases (ICD-10) criteria for cannabis dependence in patients prescribed CbM for chronic pain has not been assessed.Physicians from Canada, Germany and Israel were recruited via the mailing lists of national pain societies. To be eligible, physicians had to have prescribed CbM for chronic pain treatment in the past 12 months. Participants were asked to rate the appropriateness of items adapted from the substance use module of the Composite International Diagnostic Interview Version 3.0 to assess dependence in the context of chronic pain treatment with CbM and the appropriateness of two additional items. In case of disagreement, participants were asked to give reasons. Furthermore, they were asked to suggest potential additional criteria.On average 69.0% of 178 participating physicians indicated agreement with the appropriateness of the ICD-10 criteria, while 20.6% indicated disagreement. The highest disagreement rate was observed for the item on repeated use despite legal problems (35.4% disagreement or strong disagreement). Reasons for disagreement were summarized into six content categories using qualitative methods of analysis. Additional criteria suggested by participants were summarized into 10 content categories.A collaboration of psychiatrists and pain physicians is required to define criteria and develop instruments to capture abuse and dependence of CbM in chronic pain patients.Sixty-nine per cent of 178 pain medicine physicians in Canada, Germany and Israel who participated in a survey on the appropriateness of the ICD-10 criteria for cannabis abuse and dependence for patients prescribed cannabis-derived products for chronic pain assessed the criteria as appropriate, whereas 20.6% deemed the criteria as not appropriate.