[jrapdx3]

After reading the article and the comments on this page, I also think it's unclear what exclusion criteria were used to select the cannabis using and control cohorts. Controlling for likely and unlikely confounding variables is essential in this kind of study. Obesity is a notoriously heterogenous condition with a great number of inheritable and environmental contributors which makes the task especially difficult.

However, a connection of particular interest concerns ADHD, a disorder identified as having a strong link to obesity, including common genetic predisposition [0]. Furthermore, individuals with ADHD are also more likely than non-ADHD peers to develop drug dependence, including cannabis-use disorder [1,2]. If ADHD was not among direct or indirect exclusion criteria, the results of the recent study could be misleading or at least incompletely characterized.

    [0] https://pmc.ncbi.nlm.nih.gov/articles/PMC6097237/ 
    [1] https://pmc.ncbi.nlm.nih.gov/articles/PMC5568505/  
    [2] https://pmc.ncbi.nlm.nih.gov/articles/PMC8025199/

[bb88]

I think the issue I have is simpler. My understanding is those that were obese and have lost weight still have a higher chance of getting type 2 diabetes than those that were slim throughout their lives. If your chance of getting diabetes is 70% while obese, and 2.2% (or 15% or whatever it actually is) after losing weight, how is that not a win?

[jrapdx3]

It is indeed a win. It's long been established that for obese individuals even 5% weight loss reduces comorbidity of obesity-related conditions. Of course greater weight loss, 10-15%, gives better outcomes. Typically the difficulty is maintaining lower weight for the long haul. For those who can do it the payoff is substantial.