José A. Luchsinger, MD, is a professor of medicine and epidemiology at Columbia University Irving Medical Center, New York. As an expert in aging and cognition, he shares his insights on the relationship between diabetes and cognitive impairment, and how health care professionals can adapt treatment for patients with cognitive problems.

Q: What is cognitive impairment, and why does it matter for people with diabetes?

A: Cognitive impairment is a decline from usual cognitive abilities. There are different types. The one most people are concerned about is forgetfulness, which is the inability to remember events or learn material. Other types of cognitive impairment are not related to memory or can co-occur with memory problems. These include problems with attention and executive function—the ability to start, conduct, and finish a complex task.

When these problems are mild and do not impact a person’s ability to function by themselves, they're called mild cognitive impairment. When cognitive impairment is so severe that it affects people’s ability to live independently—meaning they need help to remember things, keep track of things, or conduct activities of daily living—then we call that dementia.

People with diabetes are at risk of developing cognitive impairment, which presents challenges for following treatment.

Q: Are people with diabetes more likely to develop cognitive impairment than people without diabetes?

A: Yes. Studies have demonstrated that people who have diabetes, compared with people without diabetes, are more likely to develop cognitive problems. Older adults with diabetes have higher incidences of dementia, Alzheimer’s disease, and vascular dementia than those with normal glucose tolerance.

All cognitive domains may be affected. But the nonmemory domains, like attention, executive function, and psychomotor speed (how fast a task is done), are affected more strongly or are affected first, studies show. People with diabetes need to follow a relatively complex treatment regimen. Anything that affects their ability to remember or organize themselves to begin and carry out an action related to their treatment is important.

Why people with diabetes are more likely to develop cognitive impairment is not well understood. One explanation is that people with diabetes age just like people without diabetes. Cognitive problems increase with age, so people with diabetes are bound to have the same problems as the general population. We also know that certain comorbid conditions, such as high blood pressure, can contribute to cognitive decline.

We know that diabetes causes cerebrovascular disease—that is, vascular disease in the brain—just as it causes vascular disease in the heart or in the legs in peripheral arterial disease, for example. This damage in the brain can lead to different cognitive problems. There are studies showing that diabetes can affect memory, but whether this is caused by vascular problems or other mechanisms is unclear. Investigators are interested in figuring this out.

Q: Which groups are most at risk of cognitive impairment?

A: Some studies suggest that the more severe the diabetes or the longer the duration of diabetes, the higher the risk of developing cognitive impairment. But not all studies show this. In general, older adults are at higher risk of developing diabetes or cognitive impairment, or both, than younger adults.

Risk does vary among racial and ethnic groups. Non-Hispanic Black and Hispanic people have a greater risk of cognitive impairment than non-Hispanic white people. That’s true for diabetes, too. My colleagues and I looked at whether these disparities in diabetes could account for disparities in cognitive impairment. Our study of 941 adults age 65 and older in northern Manhattan, New York, suggested that at least part of the disparity in cognitive impairment between racial and ethnic groups could be accounted for by the disparities in diabetes prevalence.

Q: What signs of cognitive impairment should health care professionals look for in patients with diabetes? Should they screen for cognitive impairment? 

A: Anyone on the diabetes care team, family members, and patients themselves who notice problems keeping up with treatment—for example, forgetting appointments, forgetting to take medication, or trouble following a complex treatment regimen—should recognize there's a possible cognitive issue underlying those behaviors. Another sign is that diabetes control worsens despite good treatment. We have to be careful not to judge the patient, to think they’re being irresponsible.

It's important to go the extra mile to make sure that prescriptions are being filled. If they're being filled, is medication actually being taken? For that, it's necessary to do pill counts, to communicate with nurses, and to have patients bring in their medications.

As for screening, the U.S. Preventive Services Task Force does not recommend screening older adults for cognitive impairment because there is no evidence it improves patient or caregiver outcomes. However, the American Diabetes Association (ADA) 2021 practice guidelines do call for screening adults age 65 and older with diabetes for cognitive impairment. That is covered by Medicare and can be done at the initial visit and annually as appropriate.

Now, if somebody has signs of concern, then we’re talking about detection, not screening. There are various tools, such as the Mini-Mental State Examination (PDF, 29 KB) , the Montreal Cognitive Assessment, the Mini-Cog, or the Memory Impairment Screen (PDF, 104 KB) . The use of one or the other depends on availability and time. When you have a patient with complex issues, taking 5 or 10 minutes to do this is easier said than done. Depending on the results, the patient might need further testing and referral to a neurocognitive specialist.

Q: How can health care professionals work successfully with patients who have both diabetes and cognitive impairment?

A: Successful management of diabetes requires a high degree of patient involvement. Patients with cognitive impairment may have trouble adhering to treatment, including monitoring blood glucose, taking medications as directed, timing meals, and recognizing and managing hypoglycemia. Patients who live alone are at particular risk, as are those with complex treatment regimens.

Health care professionals can simplify treatment regimens, if possible; make sure patients have the support they need to comply with treatment; and avoid complications such as hypoglycemia. This could involve setting appropriate but not overly ambitious glycemic targets. The current ADA recommendation in elderly people with diabetes and cognitive impairment is a hemoglobin A1C level between 7% and 8%.

Depending on the stage of cognitive decline, patients may be able to perform some tasks independently and need help with others. Caregivers should be educated to help with blood glucose monitoring and to recognize signs of hyper- and hypoglycemia. Health care professionals can help arrange professional services for the patient—a visiting nurse service or other type of support system to make sure the patient can comply with treatment.

More research is needed to determine the best approaches to assess cognition and address cognitive decline in the health care setting. At what stage do you assess cognition? Do you do it when people are asymptomatic or already symptomatic? What do you do if cognitive impairment is detected? What are the best strategies to support patients? It seems that a team approach, such as a medical home, would be the ideal solution, with more support given to people who have cognitive impairment.

Q: Can improved glucose control or diabetes treatments help prevent cognitive impairment?

A: There are studies suggesting that better glucose control—through medications or other interventions—can prevent cognitive impairment, but it has not been established with certainty. For example, metformin has been tested to treat or prevent Alzheimer’s dementia, often in mildly impaired subjects. Negative and conflicting results point to the need for larger clinical trials.

ACCORD is a landmark study of tight glycemic control versus regular control in people with type 2 diabetes. A sub-study called MIND compared cognitive outcomes; no differences were found between the two groups. That is, tight glycemic control did not help prevent cognitive impairment or improve cognition.

In addition, the Diabetes Control and Complications Trial (DCCT) compared the effects of intensive insulin therapy and standard therapy in people with type 1 diabetes. It showed no relationship between hypoglycemic status and cognitive-test performance. The follow-up Epidemiology of Diabetes Interventions and Complications (EDIC) study found similar results.

Do you screen your patients for cognitive impairment? Share below in the comments.