Highlights
In this large population-based Norwegian cohort, greater midlife central adiposity, assessed by waist-to-height ratio, was associated with higher late-life plasma p-tau217 and higher risk of biomarker-verified Alzheimer disease dementia.
The signal was not limited to central adiposity. Body mass index showed a similar life-course pattern, with midlife obesity linked to higher late-life Alzheimer biomarker burden.
By contrast, higher adiposity measured in late life was associated with lower plasma p-tau217 concentrations and lower risk of positive p-tau217 and biomarker-verified Alzheimer dementia, consistent with reverse causation, preclinical weight loss, or frailty-related selection effects.
The findings strengthen the view that midlife, rather than late life, is the key window in which excess adiposity may contribute to Alzheimer pathogenesis and potentially modifiable dementia risk.
Background
Obesity has long been implicated in dementia risk, but its relationship with Alzheimer-specific biology has been harder to define. Epidemiologic studies have repeatedly shown that obesity in midlife is associated with higher risk of later cognitive decline and dementia, whereas overweight or obesity in late life often appears neutral or even protective. This apparent contradiction has generated substantial debate and likely reflects a combination of biological timing, survivor bias, and the tendency for weight loss to occur during the prodromal phase of neurodegenerative disease.
Most prior work has relied on clinical dementia outcomes rather than biomarker-defined Alzheimer disease. That distinction matters. Dementia is syndromic and etiologically heterogeneous, whereas plasma p-tau217 is emerging as one of the most informative blood-based biomarkers of Alzheimer-type tau pathology, closely linked to amyloid and tau processes in the brain. A life-course study linking serial anthropometric measures to plasma p-tau217 therefore addresses an important gap: whether adiposity is associated not simply with dementia broadly, but with biologically anchored Alzheimer disease.
The new HUNT Study analysis is clinically relevant for two reasons. First, it uses repeated anthropometric measures collected over up to 35 years, allowing temporal interpretation that cross-sectional studies cannot provide. Second, it defines biomarker-verified Alzheimer dementia by combining clinical diagnosis with positive plasma p-tau217, thereby bringing etiologic specificity to an otherwise noisy clinical endpoint.
Study Design
Overall design and population
This was a cohort study nested within the fourth wave of the Norwegian Trøndelag Health Study (HUNT4; 2017-2019). Investigators included community-dwelling participants aged 70 years or older who had plasma p-tau217 measured and underwent standardized clinical cognitive assessments at HUNT4.
The final analytical sample consisted of 8,797 participants. Women accounted for 53.5% of the cohort, and the mean age at HUNT4 was 77.8 years (SD 6.2).
Exposure assessment
The study examined both central and general adiposity across adulthood and old age. Central adiposity was assessed by waist-to-height ratio (WtHR), measured three times across HUNT2 to HUNT4 between 1995 and 2019. General adiposity was assessed by body mass index (BMI), measured four times across HUNT1 to HUNT4 between 1984 and 2019.
WtHR is increasingly considered clinically useful because it captures abdominal fat distribution relative to body size and may better reflect cardiometabolic risk than BMI alone. In this study, a midlife WtHR of 0.60 or higher represented marked central adiposity and was compared with WtHR below 0.50.
Outcomes
The primary biomarker outcome was plasma p-tau217 concentration. A threshold of 0.63 pg/mL or greater defined positive p-tau217. Clinical dementia was determined using standardized cognitive assessment. Biomarker-verified Alzheimer dementia was defined as the combination of a clinical dementia diagnosis and positive plasma p-tau217.
In the cohort, 2,649 participants (30.1%) were p-tau217-positive, and 659 participants (7.0%) met criteria for biomarker-verified Alzheimer dementia.
Statistical analysis
The investigators used linear regression for continuous p-tau217 concentration, logistic or relative risk ratio-based models for categorical outcomes, and linear mixed-effects models to examine long-term trajectories of adiposity measures. Analyses were adjusted for demographics, APOE ε4 genotype, lifestyle factors, and mental health variables. This adjustment strategy aimed to reduce confounding, though residual confounding remains possible in any observational study.
Key Findings
Midlife central adiposity predicted higher late-life Alzheimer biomarker burden
The central finding was that excess abdominal adiposity in midlife tracked with higher late-life p-tau217, an increasingly validated blood-based Alzheimer biomarker. Compared with participants whose midlife WtHR was below 0.50, those with midlife WtHR of 0.60 or greater had a 12.8% higher late-life plasma p-tau217 concentration, with a 95% confidence interval of 7.3% to 19.7%.
This association also translated into categorical biomarker risk. Midlife WtHR of 0.60 or greater was associated with higher risk of positive p-tau217, with a relative risk ratio of 1.55 (95% CI 1.21-1.99). The same exposure was associated with higher risk of biomarker-verified Alzheimer dementia, with a relative risk ratio of 1.84 (95% CI 1.27-2.65).
These are clinically meaningful effect sizes. They suggest that central obesity in midlife is not merely associated with vascular cognitive burden or nonspecific dementia risk, but with later evidence of Alzheimer-type tau pathology itself.
Late-life central adiposity showed the opposite pattern
In sharp contrast, higher central adiposity measured in late life was associated with lower p-tau217 and lower probability of biomarker positivity. Participants with late-life WtHR of 0.60 or greater had 15.6% lower plasma p-tau217 concentration compared with those with WtHR below 0.50, with the reported confidence interval indicating a statistically robust inverse association. The same late-life WtHR category was linked to lower risk of positive p-tau217, with a relative risk ratio of 0.49 (95% CI 0.41-0.59), and lower risk of biomarker-verified Alzheimer dementia, with a relative risk ratio of 0.63 (95% CI 0.46-0.86).
This reversal is one of the most important interpretive features of the study. It argues strongly against a simplistic reading that “higher weight protects against Alzheimer disease.” Instead, it supports a life-course model in which excess adiposity may be harmful in midlife but lower body weight in old age may partly reflect underlying neurodegeneration, reduced appetite, sarcopenia, systemic illness, or frailty.
BMI mirrored the WtHR findings
BMI produced a similar pattern. Midlife obesity was associated with higher late-life p-tau217 concentration and elevated risk of biomarker-verified Alzheimer dementia, whereas late-life overweight and obesity were associated with lower p-tau217 concentrations and lower risk of biomarker-verified Alzheimer dementia.
This concordance strengthens the overall inference. However, the WtHR findings may be especially important because central adiposity is metabolically more active and more strongly linked to insulin resistance, systemic inflammation, dyslipidemia, and vascular injury than BMI alone.
Trajectory analyses supported a reversal over time
Among participants with positive p-tau217 or biomarker-verified Alzheimer dementia, mixed-effects regression showed higher adiposity in midlife followed by reversal in late life. In practical terms, individuals with later Alzheimer biomarker positivity or Alzheimer dementia tended to have greater adiposity earlier in adulthood but lower adiposity by old age.
This trajectory is biologically plausible and aligns with previous observations that body weight often declines during the preclinical and prodromal phases of dementia, sometimes years before diagnosis. The HUNT data therefore help reconcile decades of conflicting epidemiology by demonstrating that timing matters substantially.
Clinical Interpretation
Why might midlife adiposity promote Alzheimer pathology?
Several mechanisms are plausible. Central adiposity is closely linked to chronic low-grade inflammation, insulin resistance, altered adipokine signaling, endothelial dysfunction, and atherosclerotic vascular disease. Each of these may contribute directly or indirectly to neurodegeneration. Hyperinsulinemia may impair amyloid clearance; inflammation may accelerate tau phosphorylation; vascular dysfunction may reduce brain resilience; and obesity-related metabolic disturbances may amplify neurodegenerative cascades over decades.
Abdominal fat may be particularly relevant because visceral adipose tissue is endocrinologically active. It produces cytokines and inflammatory mediators that can influence blood-brain barrier integrity, cerebrovascular health, and neuronal stress responses. The stronger emphasis on WtHR in this study therefore has translational value: it is inexpensive, easy to measure, and may more closely reflect pathogenic adiposity than BMI.
Why do late-life associations appear protective?
The inverse late-life association likely reflects several overlapping processes. Reverse causation is the leading explanation: preclinical Alzheimer disease may cause weight loss through hypothalamic dysfunction, reduced olfaction, declining appetite, behavioral change, or emerging functional impairment. Frailty and sarcopenia may also shift body composition in ways not fully captured by BMI or WtHR. Selective survival is another consideration, because individuals with longstanding obesity who survive into older age may differ biologically from those who do not.
For clinicians, the implication is straightforward: excess weight in an older adult should not be interpreted as protective against Alzheimer disease in a causal sense, and low or falling weight in later life may warrant attention as a possible marker of underlying illness, including neurodegeneration.
Strengths and Limitations
Major strengths
The study has several notable strengths. It is large, population-based, and uses repeated anthropometric measures spanning up to 35 years. It evaluates both BMI and WtHR, allowing comparison between general and central adiposity. It also uses plasma p-tau217, one of the most promising blood-based biomarkers for Alzheimer disease, thereby improving biologic specificity compared with studies based only on clinical dementia diagnosis.
Another strength is the use of biomarker-verified Alzheimer dementia, which reduces some diagnostic misclassification. Adjustment for APOE ε4 status and multiple lifestyle and mental health factors further improves internal validity.
Important limitations
As with any observational study, causality cannot be established. Residual confounding remains possible, including by unmeasured comorbidity, medication use, socioeconomic factors, and changes in diet or physical activity over time. Anthropometric measures are practical but imperfect surrogates for body composition; they cannot distinguish visceral from subcutaneous fat, nor do they capture lean mass or sarcopenic obesity.
The biomarker threshold for p-tau217 is operationally useful, but blood biomarkers continue to evolve, and assay standardization across populations is still an active field. The cohort is drawn from Norway and is predominantly older adults of a specific population background, so generalizability to more diverse populations requires further confirmation. Finally, although biomarker-verified Alzheimer dementia improves specificity, mixed pathologies remain common in late-life cognitive impairment.
Implications for Practice and Prevention
The practical message is that midlife adiposity deserves attention as a potentially modifiable contributor to Alzheimer disease risk. This aligns with broader dementia prevention frameworks, including the 2024 Lancet Commission, which identifies obesity among modifiable risk factors for dementia prevention. The HUNT data add an important mechanistic dimension by linking obesity not only to clinical dementia but also to plasma evidence of Alzheimer-type pathology.
WtHR may be especially appealing in routine practice and public health screening. It is simple to calculate, requires only waist circumference and height, and may offer a more informative measure of central fat than BMI in some adults. Whether WtHR should be formally incorporated into dementia risk stratification tools remains uncertain, but this study supports further evaluation.
At the same time, weight management advice in older adults should be individualized. Intentional weight reduction in midlife for cardiometabolic benefit remains well supported. In contrast, unintentional weight loss in late life should prompt clinical assessment rather than reassurance. Preserving nutritional status, muscle mass, and function remains essential in older populations.
Context Within the Literature
The findings are directionally consistent with prior epidemiology showing that midlife obesity increases later dementia risk, while late-life obesity often shows attenuated or inverse associations. They also complement work on blood biomarkers of Alzheimer disease suggesting that p-tau217 performs well in detecting Alzheimer pathology and may be useful for large-scale risk stratification.
What is particularly novel here is the combination of life-course anthropometric trajectories with a contemporary blood biomarker and a biomarker-verified clinical endpoint. This helps move the field beyond the long-standing debate over whether obesity is linked to dementia, toward the more precise question of when adiposity matters and through what biological pathways.
Conclusion
The HUNT Study provides compelling life-course evidence that both central and general adiposity in midlife are associated with higher late-life plasma p-tau217 and higher risk of biomarker-verified Alzheimer dementia. The opposite association observed in late life likely reflects disease-related weight loss, frailty, and survivor effects rather than a true protective effect of excess adiposity.
For clinicians and health systems, the study reinforces midlife obesity prevention as part of long-horizon brain health strategy. For researchers, it highlights the importance of timing, repeated measurements, and biomarker-defined outcomes in dementia epidemiology. Future studies should test whether reducing midlife central adiposity can meaningfully alter downstream Alzheimer biomarker trajectories and clinical risk.
Funding and Trial Registration
The abstract provided does not report a ClinicalTrials.gov registration number, which is expected given the observational cohort design. Specific funding details should be confirmed from the full article in Neurology.
Citation
Zotcheva E, Strand BH, Sunde A, Deckers K, Aarsland D, Ashton NJ, Zetterberg H, Skirbekk VF, Borda MG, Livingston G, Singh-Manoux A, Selbaek G. Associations of Anthropometry Measures Across 35 Years With Late-Life Plasma P-tau217 and Dementia: The HUNT Study. Neurology. 2026-06-02;106(12):e218093. PMID: 42228892. https://pubmed.ncbi.nlm.nih.gov/42228892/
Selected References
Livingston G, Huntley J, Liu KY, et al. Dementia prevention, intervention, and care: 2024 report of the Lancet standing Commission. Lancet. 2024;404(10452):572-628.
Ashton NJ, Janelidze S, Al Khleifat A, et al. A multicentre validation study of the diagnostic value of plasma neurofilament light and p-tau217 for Alzheimer’s disease. Nat Med. 2024;30:???
Note: Readers should verify final bibliographic details for biomarker method references directly from PubMed or the full Neurology article, as the present summary is anchored primarily to the indexed study citation supplied above.