Danish adults who started a statin within the first year of a type 2 diabetes diagnosis had a 15% lower relative risk of dementia over 10 years than those who never started one, according to a study of 132,585 people presented today at ESC Congress 2026 in Munich and published in The Lancet Regional Health, Europe. People who waited and started between one and five years after diagnosis had a 10% lower relative risk, which points to timing mattering.
The study was observational. It cannot show that statins prevent dementia.
Key takeaways
- Among 132,585 statin-naive Danes who developed type 2 diabetes from 2006 through 2019, 2.7% developed dementia over a median 7.1 years of follow-up.
- Starting a statin within one year of diagnosis tracked with a 15% lower relative risk of dementia over 10 years; starting one to five years later tracked with a 10% lower relative risk.
- Results were similar in women and men, and the researchers say their design reduces, but does not remove, the bias built into observational data.
What the study found
The team, led by Dr. Tummas Ternhamar at Copenhagen University Hospital, pulled 132,585 people from Danish national registers who had never taken a statin and who developed type 2 diabetes between 2006 and 2019, according to the European Society of Cardiology. Follow-up ran to the end of 2021, and medical records were searched for any diagnosis of dementia from any cause.
Everyone was sorted into three groups: no statin started within five years of diagnosis, an early start within one year, and a late start between one and five years.
The analysis used a clone-censor-weight design. In plain terms, that is a statistical method built to imitate the structure of a randomized trial using data from people who were never randomized. Ternhamar said the team tested the idea “in a large nationwide study using analysis methods to emulate a randomised trial and reduce some of the bias associated with observational studies.”
Over a median 7.1 years, 2.7% of the group developed dementia. Against never starting, early statin treatment was linked to a 15% lower relative risk of dementia over 10 years, and late treatment to a 10% lower relative risk.
Dr. Kumar’s take
My skepticism about how widely statins get handed out is on record, and this study does not change it. What it does do is describe a group where the biology lines up better than usual.
Most dementia in people with diabetes is not textbook Alzheimer’s disease. It is small-vessel disease and mixed disease, meaning the tiny arteries that feed deep brain tissue stiffen, narrow, and drop out, and the damage stacks up over years without a single dramatic event to mark it. If a drug protects those small vessels, the brain is exactly where you would expect a signal to show up. The timing pattern fits that picture too. Vascular injury in diabetes starts accruing at diagnosis, not five years after it, so an earlier start would be expected to beat a later one.
Now the trap. This is confounding by indication, and it is the reason this is an association and not a prescription. A person who fills a statin prescription within a year of a new diabetes diagnosis is, on average, a different person from one who never fills it. They showed up to the appointment. They are more likely to take their other medications, to have their blood pressure checked, and to have a doctor who follows up. Those same habits independently lower dementia risk, and so does staying strong and active. A target trial emulation narrows that gap. It does not close it.
The effect size deserves the same flat reading. A 15% lower relative risk in a group where 2.7% developed dementia is a modest shift, not a shield.
What it means for you
If you have type 2 diabetes and your clinician has raised a statin, this is one more data point for that conversation, and the timing part is the useful piece: the study found the larger association with starting sooner rather than later. Bring it up at the visit rather than treating it as a settled answer.
If you do not have diabetes, nothing here argues for a statin. Note that the same small vessels this study is circling respond to blood sugar control, blood pressure control, and what you eat over decades, none of which require a prescription.
And treat the headline for what it is. Professor Isabel Goncalves of the ESC Communication Committee, commenting on the findings, put the limit plainly: “this was an observational study, so it cannot prove that statins prevent dementia.”