A five-minute test of balance, walking and standing up was linked to later dementia risk

Doctors and researchers have a quick, cheap way to measure how well an older person's body works: the Short Physical Performance Battery, or SPPB. It has three parts. First, a balance test: standing with the feet side by side, then with one foot half in front of the other, then heel to toe. Second, standing up from a chair five times as fast as possible. Third, walking 2.44 metres (8 feet) at a normal pace. Each part gives 0 to 4 points, so the total runs from 0 (worst) to 12 (best). No special equipment is needed.

Low scores on this test are already known to go with falls, frailty, hospital stays and earlier death. A team at Henan Medical University in China asked whether they also go with a higher risk of developing dementia, the loss of memory and thinking that makes daily life hard.

What they did

The team used the English Longitudinal Study of Ageing (ELSA), a long-running survey meant to represent people aged 50 and over in England. In 2012/2013, nurses gave the SPPB to participants aged 60 and older. After removing people with missing test results, those who already had dementia and those who never came back for a follow-up interview, 4,927 people remained. Their median age was 69, and 55% were women.

Scores were sorted into four groups: high (10–12 points), moderate (7–9), low (4–6) and very low (0–3). Most people, 70%, scored high; 7.4% scored low and 6.2% very low.

Participants were interviewed again about every two years, up to 2021/2023. Someone was counted as having dementia if they said a doctor had diagnosed dementia or Alzheimer's disease, or if a relative or carer filled in a questionnaire showing a clear decline in memory and thinking. The statistics were adjusted for age, sex, education, marriage, ethnicity, smoking, drinking, weight, physical activity, several chronic diseases, blood pressure and diabetes medicines, trouble with daily tasks and depression.

What they found

• 252 people (5.1%) developed dementia over a median of 8.7 years.
• Lower scores, higher risk. Compared with the high-scoring group, the low group had a 71% higher rate of dementia and the very low group about two and a half times the rate. The moderate group did not differ clearly.
• Point by point. Each extra point on the 12-point test went with an 11% lower dementia rate.
• Ages and sexes. In people aged 60 to 79, both the low and very low groups had higher risk; in those aged 80 and over, only the very low group did. In women, both low and very low scores were linked to higher risk; in men, only very low scores. The authors note that the difference between the sexes was not statistically significant and should be read with caution.
• Not just early dementia. When people diagnosed in the first two years, or the first five years, were left out, the link remained. When the team also accounted for memory scores at the start, the link became weaker, but a very low score was still tied to a higher risk.

What to keep in mind

• Observational. The authors stress that this kind of study cannot show cause and effect. Dementia could, for example, start damaging the brain years before diagnosis and slow people down physically, although the sensitivity analyses argue against that being the whole story.
• Dementia was not diagnosed by the study. It came from what participants reported and from relatives' questionnaires, not from clinical examinations, so some cases were probably missed.
• Healthier sample. The people left out of the analysis were older, less educated, less active and sicker than those included, which may make the results less representative.
• Deaths not fully accounted for. People with poor physical performance may die before dementia appears; the team could not analyse this because death dates were incomplete. The authors think this would, if anything, weaken the link they saw.
• Missing factors. Diet and genes such as APOE were not taken into account.
• One country. The findings come from older adults in England and may not apply everywhere.

Why it matters

The SPPB takes only a few minutes and can be done in a clinic or a community centre. If low scores reliably flag a higher dementia risk, the test could help identify older people who might benefit from closer attention. The link held even after accounting for how physically active people said they were, which suggests the test captures something beyond exercise habits. What the study cannot say is whether improving these scores, for example through exercise, would lower dementia risk; the authors call for intervention studies to test exactly that.