Thanks to a new computational model, doctors will now be able to calculate the risk of older patients with type 2 diabetes developing dementia. Doctors can now estimate if patients have a higher risk of developing dementia or if they are less at risk.

The fact that, on average, diabetes patients are twice as likely to develop dementia as people without diabetes was already known. However, there are major differences between patients and it is unknown which factors determine the risk of diabetes patients developing dementia. Insight into who has a high risk of developing dementia is important, as it may be possible to influence these factors in the early stage of diabetes.

Eight factors

The study showed that there are eight deciding risk factors (i.e. older age, cerebrovascular conditions, acute severely irregular blood sugar levels, a history of depression, microvascular conditions, diabetic foot deformities, cardiovascular conditions, and a lower level of education). Using these factors, the researchers developed a computational model based on statistical analyses. Thanks to this model, it is now possible to predict the risk of an individual diabetes patient developing dementia in the next ten years.

The study showed that the lowest risk group (which comprises approximately 20% of the patients) have a less than 10% chance of developing dementia within ten years. At the other end of the spectrum, the study showed that a small group of patients (approximately 4% of the total group) have a 65% chance or higher of developing dementia.

Lieza Exalto is a trainee neurologist at University Medical Center (UMC) Utrecht and the lead author of the paper. She explains: “Doctors can use the model to help them make decisions in the case of early-stage cognitive disorders that make people vulnerable to the side effects of diabetes treatment. The risk score will also help us to understand the causes of the increased risk of developing dementia among diabetes patients because we can study them in the early stages of the dementia process.”

“The computational model can also be useful when selecting the high risk patients to participate in trials exploring new therapies for dementia,” says Prof. Dr. Geert Jan Biessels, a neurologist affiliated with the UMC Utrecht Brain Center and co-author of this publication.

The database

The so-called ‘Diabetes-Specific Dementia Risk Score’ was published this week in the first issue of Lancet Diabetes & Endocrinology, a new publication from the prestigious medical journal The Lancet. The researchers developed the model by using the data of nearly 30,000 diabetes patients over 60 years who have been monitored since 1994 and are included in the database of Kaiser Permanente Northern California, a major US health insurer. This database contains detailed and well documented demographic and medical data of more than three million insured parties, each of whom has been monitored for at least ten years. As such, this database is ideal for healthcare studies.

more about computational model

more about The University Medical Center