How do you counsel non-diabetic patients who wish to start metformin to reduce the risk of developing dementia?

There have been some interesting observational studies evaluating the reduction of cognitive decline in patients with type 2 DM. First, they are all in patients WITH diabetes, so not yet generalizable to patients without DM. Second, they are observational trials, which means that they cannot adjust for the important differences of confounding by indication. For example, elders who are healthy enough to tolerate metformin (i.e., no advanced CKD) are inherently different than those who cannot. Because this is such a standard of DM care, I would worry the patients NOT on metformin are very different in ways that may predict the outcome. So in summary, I would not use a drug that has not yet been validated in the population you are interested i...

Multi-domain lifestyle interventions, such as those used in the FINGER trial, have stronger evidence for reducing and partially reversing cognitive impairment, are safe, and have co-benefits to all organ systems compared to metformin. I tell patients to focus on those instead, and that Metformin data is not robust, and that diet, exercise, good sleep, purpose, avoiding drugs and tobacco, and emphasizing social connection are essential to reduce their risk of cognitive impairment. We don’t yet have a magic pill that comes anywhere close to reducing cognitive impairment compared to living well.