
A study linked higher BMI at dementia diagnosis with lower mortality. Learn why this concerns survival, not prevention or advice to gain weight.

In October 2026, Columbia Mailman School of Public Health reported a study linking higher body mass index at dementia diagnosis with lower subsequent mortality. Led by Virginia Chang of NYU and Yuan Zhang of Columbia, “Body mass index and mortality in incident dementia” was published in the Journal of Alzheimer’s Disease.
According to the Columbia study release, the research team used Health and Retirement Study data collected from 1998 through 2022. The analysis included 4,523 people who were dementia-free when they entered the study and later developed dementia.
The researchers examined body mass index, or BMI, at diagnosis in relation to deaths afterward. Compared with the normal-weight group at diagnosis, they reported:
The reported associations remained after the team adjusted for chronic conditions, smoking, and demographic characteristics. That means the pattern was still present after accounting for those measured differences between participants.
Each percentage is a comparison with the normal-weight group. Keeping that reference group in view helps prevent a striking figure from becoming detached from the actual study question: how weight at diagnosis related to survival afterward.
The release also states that the analysis received National Institutes of Health support and that the authors declared no competing financial interests.
For people living with dementia and those supporting them, the immediate value is a clearer focus on nutrition and weight history during care conversations. A BMI category is only part of that discussion.
Chang noted that people with dementia may experience wasting and difficulties eating or feeding themselves. Her comments put the survival findings in the context of challenges that patients and care partners may face.
The practical response is to bring unintentional weight changes or eating difficulties to a clinician or dementia care team. The discussion can focus on the person's nutrition needs and wider health rather than on reaching a category described in a headline.
Useful questions for that conversation include:
These questions focus attention on changes a care team can assess in the context of a person's health. They also give patients and care partners a clear way to raise nutrition concerns without having to interpret the research statistics on their own.
For someone who has recently been diagnosed, that focus can be more useful than trying to judge whether a familiar weight label is reassuring or worrying. The same conversation can include what has changed over time and whether eating has become difficult. No single label needs to carry the whole discussion.
The study was observational. It found an association rather than showing that higher weight caused longer survival. As Forbes’s coverage also reported, the researchers said their analysis was not designed to determine whether people with dementia should gain weight.
Nor did it test whether higher weight prevents dementia. Everyone in this analysis eventually developed dementia, so the question was survival after diagnosis rather than the chance of developing the condition. Readers interested in the separate prevention question can turn to FitBrainLab’s discussion of dementia risk reduction.
The percentages also do not say how many additional years any individual might live. They describe differences in mortality risk between groups, not a personal forecast or the outcome of a weight-gain program.
The reported outcome was mortality, meaning death during follow-up. That keeps this study focused on survival. The figures are not measures of memory, attention, independence or quality of life.
The researchers discussed energy stores and muscle mass as possible explanations for the association. They did not establish the reason for it. These are ideas for further study rather than confirmed explanations for the mortality differences.
For related context on the muscle question, FitBrainLab also covers strength and brain health.
The release notes that dementia-related weight loss may begin years before diagnosis and accelerate closer to it. This means a normal-range BMI at diagnosis may reflect an earlier change in weight rather than a person's longer-term pattern.
The authors examined potential biases involving weight loss before diagnosis, selection effects from studying people who developed dementia, and survivor bias. They reported that additional analyses did not suggest these issues explained the association. That is narrower than saying bias was ruled out.
The concern is that the people included in a study may differ from those who are not included. Here, everyone had developed dementia and survived to diagnosis. The authors' checks therefore matter for interpreting this particular group, rather than treating it as a sample of every older adult.
The release does not provide confidence intervals or p-values for the estimates. It also does not provide full model specifications or detailed results for each sensitivity analysis. Those gaps limit how closely readers can assess the precision of the findings and the checks for bias.
This finding matters because cognitive-health reporting needs to follow people beyond the moment of diagnosis. Research on survival asks a different question from research on avoiding disease. Keeping both questions visible can make dementia coverage more useful to people who are already living with the condition.
It also makes weight history a more meaningful subject than a single label. A number recorded at diagnosis sits within a person's health and care needs. The direction of weight change and the ability to eat are part of that wider picture.
For researchers, the next task is to explain the observed association more clearly. For patients and families, the useful response is a conversation about nutrition and concerning changes with qualified care. The enduring lesson is to read a surprising result for the question it answers, then keep the person and their needs at the center of care.
A surprising link between weight and dementia survival needs context rather than a more frightening headline, which fits FitBrainLab’s calm approach to explaining brain-aging evidence. For adults 60+ unsettled by dementia coverage, the publication separates supported findings from overstated claims so readers can understand the research without treating it as a personal prediction.
Follow FitBrainLab for research-led insights on memory, focus, brain aging, nutrition and mental fitness after 60. Stay connected for new articles, practical guidance and ideas for a sharper, more engaged life.



Build habits that support memory, focus and a curious, connected life.
Read the Blog