Human study · Exercise · 30 Oct 2026
Very high blood p-tau217 marked a 38 percent five-year chance of cognitive impairment in older adults who were unimpaired at the start
Source: ScienceDaily
What the research found
ScienceDaily posted this on 30 October 2026. The journal paper is older: Buckley RF and colleagues, JAMA, DOI 10.1001/jama.2026.12556, journal date 15 July 2026. A University College London repository record lists it as JAMA volume 336, issue 11, pages 950 to 959. The investigators pooled six cohorts in North America, Japan, and Australia. At entry, 2,684 older adults were cognitively unimpaired. Blood p-tau217 and amyloid PET were measured at baseline. Follow-up ran from as early as 2004 through 2025, and some people were followed as long as 13.5 years. About 478 later developed cognitive impairment. A very high p-tau217 level, more than 2.5 standard deviations above the reference, carried an estimated 38 percent absolute risk at five years (95 percent confidence interval 33 to 43 percent, from the UCL and JAMA abstract). A high band, 1.1 to 2.4 standard deviations, was 24 percent (20 to 28 percent) at five years. The 10-year risk in the very-high group was estimated up to 78 percent, but the data at that horizon are sparse. Per 1 standard deviation higher p-tau217, the hazard ratio was 1.38 (1.30 to 1.46), and it remained 1.32 (1.24 to 1.41) after accounting for amyloid PET. The association went beyond APOE4. The author group, including Reisa Sperling, said the test is not recommended now for routine screening of people without symptoms, because there is no proven disease-modifying treatment for asymptomatic people at high risk. The release still points to physical activity, food, sleep, and general wellness regardless of the test. The cohorts were selected and may not represent everyone. A blood result is not a diagnosis. Sperling consults widely, including for Novo Nordisk and Lilly, and trial funding named in the disclosures includes Eisai and Lilly. Not medical advice.
Evidence level: Human study. A study in people, but an association isn't proof of cause.
More exercise news
- In 13,423 adults ages 65 and older, higher balance, strength, and march scores lined up with a lower chance of dying over seven years
- In 1,917 Mexican adults ages 50 and older, a weaker handgrip lined up with a higher chance of dying, and the link was stronger in men
- In a rodent model, tongue exercise raised output from surviving motor neurons
- Both orders of complex training improved fitness measures in 25 male university athletes
- Resistance exercise plus essential amino acid intake beat either one alone for muscle fitness in women over 65