Among thousands of Americans who later developed dementia, a quiet and puzzling pattern has emerged: those carrying extra weight at the moment of diagnosis tended to live longer than their leaner peers. Researchers at Columbia and New York University, drawing on decades of national survey data, tested whether this could be an illusion created by weight loss or by who manages to survive long enough to be counted, and found the advantage persisted, though its cause remains unknown. The authors are careful to say this is not a reason for anyone to gain weight, but a reminder that for people losin
Obesity at Dementia Diagnosis Linked to Lower Death Risk, but Researchers Reject Weight Gain as Advice
The study was not designed to test whether patients should gain weight.
Why does a finding like this matter if the authors say not to act on it?
Because families are already making food decisions every day. The study's real pull is toward nutrition: people with dementia forget to eat, lose interest in food, or struggle to swallow.
So what exactly was found?
Among 4,523 people diagnosed with dementia, class 1 obesity at diagnosis went with a 27% lower risk of death than normal weight. Overweight was 16% lower, BMI 35 or higher was 22% lower.
Worth saying clearly: that's an association. It's observational, so it can't show that weight protected anyone.
Could the result just be an illusion?
The researchers tested that. Weight can fall for up to 10 years before diagnosis, so they looked at people with stable weight, and the link held, at 33% for those who stayed in class 1 obesity.
Holding up against those checks isn't the same as being explained. The authors say they still don't know why it exists, and unmeasured factors would need to raise both dementia and death risk by about 60% to erase it.
What else limits it?
BMI doesn't separate muscle from fat, results weren't broken out by dementia type, and the benefit shrank with age. Snyder, who wasn't involved, said it can't be a basis for clinical intervention.
And Chang said plainly the study wasn't designed to test whether patients should gain weight.
What should a caregiver take from it?
Track weight, report unplanned loss, and ask about a dietitian or a swallowing evaluation.
Il Polso
- A study of 4,523 Americans with dementia links obesity at diagnosis to a roughly 27% lower risk of death, unsettling long-held assumptions about weight and aging.
- Researchers scrambled to rule out bias, testing pre-diagnosis weight loss and survivor effects, and the link stubbornly held.
- Clinicians and families face a hard truth: the study cannot explain the benefit and explicitly does not recommend weight gain.
- The practical focus has shifted to urgent watchfulness, as unplanned weight loss, forgotten meals, and swallowing trouble threaten patients whose needs cannot always be voiced.
- Caregivers are being pointed toward dietitians, speech therapists, and local aging services, while larger studies of body composition are still needed before the paradox can be resolved.
Among thousands of Americans who later developed dementia, a quiet and puzzling pattern has emerged: those carrying extra weight at the moment of diagnosis tended to live longer than their leaner peers. Researchers at Columbia and New York University, drawing on decades of national survey data, tested whether this could be an illusion created by weight loss or by who manages to survive long enough to be counted, and found the advantage persisted, though its cause remains unknown. The authors are careful to say this is not a reason for anyone to gain weight, but a reminder that for people losing their capacity to eat and swallow, nourishment and vigilance matter deeply.
Among 4,523 Americans who went on to develop dementia, those who were obese at diagnosis were up to 27% less likely to die during follow-up than those at a normal weight. The findings, published in the Journal of Alzheimer's Disease by researchers at Columbia University and New York University, have drawn attention precisely because the authors have been so firm about their limits. The pattern is known as the obesity paradox: extra weight is harmful in midlife, yet often linked to longer survival once illness has taken hold.
The team analyzed the Health and Retirement Study, a nationally representative survey of Americans aged 51 and older, tracking participants from 1998 to 2022. Their average age at diagnosis was 77.2, and most were women. Because the result runs against intuition, the researchers tried to explain it away. They examined whether weight loss before diagnosis, which can begin years earlier, was distorting the picture, and whether survivor bias was at work. The advantage held even among people whose weight stayed stable, and unmeasured factors would have needed to be unusually powerful to account for it.
Still, the study cannot say why the benefit exists, and it cannot prove cause. Body mass index does not distinguish muscle from fat, and results were not broken down by type of dementia. The Alzheimer's Association and the authors themselves stress that patients should not be encouraged to gain weight.
The real lesson, the researchers suggest, is about nutrition. People with dementia may forget to eat, struggle with utensils, or have trouble swallowing, and they often cannot describe what hurts. Caregivers can track weight and report unplanned loss, and doctors can look for treatable causes such as dental pain, depression, or medication side effects. Dietitians, speech therapists, and local Area Agencies on Aging can offer practical help. For people without dementia, the established advice remains unchanged: maintain a healthy weight, control blood pressure, and stay active in midlife.
Among 4,523 Americans who went on to develop dementia, those who were obese when the diagnosis came were up to 27% less likely to die during follow-up than people at a normal weight. The finding, published Oct. 6 in the Journal of Alzheimer's Disease, comes from researchers at Columbia University and New York University. It drew wider coverage this week, and its authors have been unusually firm about what it does not mean.
The pattern is known as the obesity paradox. Obesity in midlife raises the risk of heart disease, diabetes, and dementia itself. Yet extra weight later in life has often been tied to longer survival among people who are already ill.
The team drew on the Health and Retirement Study, a nationally representative survey of Americans 51 and older, using data from 1998 to 2022. Participants had no dementia when they entered the study and were diagnosed later. Their average age at diagnosis was 77.2, and 57.5% were women. Compared with people at a normal weight, those with class 1 obesity, a body mass index of 30 to 34.9, had a 27% lower risk of death. Overweight people had a 16% lower risk, and those with a BMI of 35 or higher had a 22% lower risk. The results were adjusted for smoking, existing medical conditions, and demographic factors. The National Institutes of Health funded the work, and the authors reported no conflicts of interest.
Because the paradox has long attracted skepticism, the researchers tested several explanations. "Because this finding is somewhat counterintuitive, an important question is whether it could simply be an artifact of bias," said corresponding author Yuan Zhang of Columbia's Mailman School of Public Health. One concern was weight loss before diagnosis. The same authors have previously found that weight loss can start up to 10 years before dementia is diagnosed, so a single measurement might put people with more advanced disease in the normal-weight group. When the team looked only at people whose weight stayed stable before diagnosis, the link held: those who remained in class 1 obesity had a 33% lower risk of death than those who remained at a normal weight.
They also considered survivor bias, the possibility that people with obesity who live long enough to develop dementia are an unusually healthy group, along with distortions that arise when a study includes only people who already have a disease. Unmeasured factors would have to raise the risk of both dementia and death by about 60% to explain the result. The researchers concluded that the biases they examined do not account for the survival advantage, though they still cannot say why it exists. The apparent benefit also shrank with age. First author Virginia Chang of NYU's School of Global Public Health told Healio the team was "both surprised and not surprised."
The limits are plain. This was an observational study, which can show a link but not prove cause. Weight was measured by BMI, which does not separate muscle from fat, and results were not reported by dementia type. Heather M. Snyder of the Alzheimer's Association, who was not involved, said the findings "cannot serve as the basis for clinical intervention." Chang was direct: "This doesn't mean that patients with dementia should attempt to become overweight or obese, as our study was not designed to test that question."
The authors say the practical lesson concerns nutrition. "Nutrition can be a big problem for people with dementia," Chang said. People with the condition may forget to eat, lose interest in food, struggle with utensils, or have trouble swallowing. Caregivers can weigh the person regularly and tell the care team about unplanned loss. A doctor can look for treatable causes such as dental pain, depression, medication side effects, or swallowing problems, which are easy to miss when someone cannot describe them. Families can ask whether a medicine affecting appetite or alertness could be adjusted, and whether finger foods or more frequent small meals would help.
Help exists beyond the doctor's office. A registered dietitian can suggest higher-calorie, easy-to-eat foods, and a speech therapist can evaluate swallowing. Medicare covers medical nutrition therapy for people with diabetes or kidney disease, and the Eldercare Locator can connect families with local Area Agencies on Aging and home-delivered meals.
For people without dementia, nothing changes. A healthy weight, blood pressure control, and physical activity in midlife remain among the best-supported ways to lower dementia risk. Larger studies of body composition and diet after diagnosis would help explain the paradox.
Citazioni salienti
This doesn't mean that patients with dementia should attempt to become overweight or obese, as our study was not designed to test that question.— Virginia Chang, NYU School of Global Public Health
The findings cannot serve as the basis for clinical intervention.— Heather M. Snyder, Alzheimer's Association