Dementia Patients With Obesity Had Up to 27% Lower Death Risk

People who were overweight or obese when they were diagnosed with dementia had a 16 percent to 27 percent lower risk of death than people at a normal weight, according to a study of 4,523 older Americans. Researchers from Columbia University’s Mailman School of Public Health and NYU School of Global Public Health published the work in the Journal of Alzheimer’s Disease.

The authors say the result gives no reason for people with dementia to try to gain weight. The study was not designed to test that, and the cause of the survival difference is still unknown.

How much lower the risk was in each weight group

Weight was measured with body mass index (BMI), a ratio of weight to height. Compared with people in the normal range, the risk of death was 16 percent lower for people classed as overweight (BMI of 25 to 29.9). It was 27 percent lower for class 1 obesity (BMI of 30 to 34.9), and 22 percent lower for class 2 and 3 obesity (BMI of 35 or higher).

The figures held after the researchers accounted for smoking, existing medical conditions and demographic characteristics.

How the team followed people from 1998 to 2022

The data come from the Health and Retirement Study, a nationally representative project that follows Americans ages 51 and older. The team analyzed records collected between 1998 and 2022 and focused on 4,523 participants who did not have dementia when they joined the study but developed it later.

Because the records span so many years, the researchers could see how each person’s weight changed before diagnosis, record the weight at diagnosis and follow survival afterward. This is an observational study: the researchers analyzed existing records and did not assign anyone to gain or lose weight.

Could bias explain a result that looks backward?

Obesity is an established risk factor for heart disease and diabetes, and being overweight or obese in middle age is associated with a greater likelihood of developing dementia later. Earlier research had already suggested that this relationship reverses after dementia develops, with higher weight tied to better survival.

Corresponding author Yuan Zhang of Columbia called the finding “somewhat counterintuitive” and asked whether it could be a statistical artifact. Several kinds of bias could produce one.

• Weight loss before diagnosis. Earlier work by Zhang and Virginia Chang found that people can begin losing weight as much as 10 years before a dementia diagnosis. Someone at normal weight at diagnosis may have been heavier before and may already be more ill, which would make higher weight look protective by comparison.
• Survivor bias. People with obesity face a higher risk of dying from certain illnesses before reaching the age when dementia typically appears, so those who reach a diagnosis may be an unusually healthy group.
• Selection. Looking only at people who develop dementia can create misleading links between obesity and death, even when the underlying connection is more complicated.

The team ran additional analyses for these problems. According to Zhang, the major sources of bias they examined did not explain the survival advantage.

Why extra weight might help, and what is still unknown

The researchers point to energy reserves as one possibility. People with more stored fat or muscle may be better able to get through periods of poor nutrition and the physical decline that can come with dementia. Dementia can reduce appetite or the drive to eat, raise energy needs, and make it harder for people to feed themselves, and over time these problems can lead to weight loss and muscle wasting.

The study did not establish that energy reserves explain the survival differences. The researchers say they cannot say for certain why overweight and obese participants had lower mortality.

What this means for patients and families

Chang, an associate professor at NYU and the study’s first author, said the work was not designed to test whether people with dementia should become overweight or obese. She and Zhang said weight status at diagnosis may serve as a signal for how long a person is likely to survive, and that the results support the importance of nutrition in people with dementia.

Co-authors are Eleanor Hayes-Larson of the University of Southern California and Hui Zheng of the University of Toronto. The National Institutes of Health funded the research. The paper, titled “Body mass index and mortality in incident dementia,” carries the DOI 10.1177/13872877261491694.

Source: ScienceDaily