This JAMA letter discusses the association between self-reported cannabis use and DKA (diabetic ketoacidosis):
To read the full article, you need to sign up for a free account—or you can read the summary on WebMD:
The letter’s data is based on self-reporting statistics of DKA patients. It establishes no causality between the two—although I could see that heavy use might impair the sound judgment required at all times to make insulin dosage choices. But I can see other issues, such as different A1c average and socio-economic level (as reported in the WebMD summary), that could be responsible for at least some, if not all of the difference in DKA rates.