Calf down 5+ cm in 3 years: 32.5% higher odds of a fall
A Chinese cohort of 7,068 adults aged 60 and over links a shrinking calf measurement to higher odds of a fall three years later, with larger losses tracking larger odds, plus what the study cannot say about cause or about any one person's risk.
By Gale Staff · October 24, 2026 · JAMA Network Open
Not reviewed by a clinician.

The short answer
Older adults whose calf circumference shrank over three years had higher odds of reporting a fall, and the odds rose with the size of the loss. Among 7,068 participants in the Chinese Longitudinal Healthy Longevity Survey, a decrease of 5 cm or more was associated with 32.5% higher odds of a fall than no decrease (odds ratio 1.33, 95% confidence interval 1.10 to 1.60), and a relative decrease of 20% or more with 48.1% higher odds (1.48, 1.20 to 1.84). This is an observational association from self-reported falls in a single follow-up wave, so it cannot show that calf loss causes falls, and the abstract does not give the fall rate in each group.
Who to see
Geriatrician
Older adults whose calves shrank over three years reported more falls, more so with bigger losses; a geriatrician can check your fall risk.
Find care →Does a shrinking calf say anything about falling?
A pair of trousers that once fit snugly around the lower leg now hangs loose, though the number on the bathroom scale has barely moved. Nobody in the household thinks much of it. The calf is part of the lower limb, and a tape measure wrapped around it is one of the cheapest ways to put a number on how much of it is left.
That is the idea behind a paper in JAMA Network Open, published on 23 October 2026 by Chi Zhang and colleagues at Beijing Hospital. The authors describe calf circumference as a simple, noninvasive proxy for lower-limb muscle mass, and they say that evidence on how a change in that measurement relates to later falls had been scarce. The question they set out to answer is whether a progressive decline in the measurement goes with higher odds of an incident fall among community-dwelling older adults.
How the researchers tracked calf size and falls
The data come from the Chinese Longitudinal Healthy Longevity Survey, a nationally representative prospective cohort. The team included participants aged 60 or older who had a complete calf measurement at baseline in 2018 and again at follow-up in 2021. That left 7,068 people with a mean age of 79.66 years (standard deviation 10.32), of whom 3,835, or 54.26%, were female.
The outcome was an incident fall, self-reported in the 2021 wave, and the median follow-up was 3.07 years (interquartile range 2.91 to 3.17). The change in calf circumference was expressed two ways: in absolute centimetres and as a percentage of the starting measurement. Each was analysed as a continuous number and in bands, with absolute change grouped as no decrease, under 3 cm, 3 to under 5 cm, and 5 cm or more, and relative change grouped as no decrease, under 10%, 10% to under 20%, and 20% or more. The reference group, labelled no decrease, combined people whose calf stayed the same with people whose calf grew.
Because the analysis only includes people who could be measured twice, those who died or dropped out between 2018 and 2021 are missing by design. The authors say they used inverse probability weighted logistic regression to address that survival bias. In plain terms, the weighting gives extra influence to participants who resemble the people who vanished, so the measured group stands in for a fuller one. The models were adjusted for sociodemographic characteristics, fall history, health behaviours, functional status and chronic diseases.
What they found: more centimetres lost, higher odds of a fall
Of the 7,068 participants, 1,671 (23.64%) reported an incident fall during follow-up. In the fully adjusted models, each 1 cm absolute decrease in calf circumference was associated with 3.2% higher odds of a fall (odds ratio 1.03, 95% confidence interval 1.02 to 1.05). Each 10% relative decrease was associated with 10.3% higher odds (odds ratio 1.10, 95% confidence interval 1.06 to 1.15).
The banded comparisons point the same way at the extreme end. Compared with no decrease, an absolute loss of 5 cm or more was associated with 32.5% higher odds of a fall (odds ratio 1.33, 95% confidence interval 1.10 to 1.60). A relative loss of 20% or more was associated with 48.1% higher odds (odds ratio 1.48, 95% confidence interval 1.20 to 1.84). Both intervals sit entirely above 1, so the association is unlikely to be a product of chance alone in this sample.
The abstract does not report the results for the two middle bands in each scheme, the under-3 cm and 3 to under-5 cm groups or the under-10% and 10% to under-20% groups. Whether the climb in odds is steady from band to band, or whether most of it sits in the largest losses, is therefore not something the text read for this story can settle.
The caveat: an odds ratio is not a fall rate
Odds ratios describe how the odds of an event compare between groups, and they are easy to read as risk. When the event is uncommon the two are close. A fall reported by nearly one participant in four is not uncommon, and in that situation an odds ratio of 1.33 overstates the ratio of the actual fall rates. The abstract gives the overall fall figure of 23.64% but not the fall rate within the no-decrease group or within the group that lost 5 cm or more, so the absolute difference a person might actually face cannot be worked out from what was published in the abstract.
The measurement itself carries a second caveat. A tape around the calf captures muscle, fat, fluid and bone together, so a smaller number can reflect muscle loss, fat loss or less swelling, and the abstract does not say how the authors separated these. The authors also describe the finding as an association. Falls are self-reported in one wave, and people who are unwell or less active may lose calf size and fall more for reasons that run in parallel, which even a long list of adjustments does not fully remove.
Why it matters
The authors conclude that monitoring calf circumference over time may serve as a practical, low-cost measure to support fall risk assessment. The new element is the trend over time, not any single reading.
The authors describe calf circumference as a proxy for lower-limb muscle mass. This paper does not test a muscle pathway directly, since it measures circumference and not muscle, and it does not test whether anything that slows calf loss changes the number of falls. What it adds is a large national sample in which the direction of change in the measurement tracked with later falls.
What this study can't tell you
- Whether shrinking calves cause falls: this is an observational cohort, so the analysis shows an association after adjustment, not that the loss of calf size produces the fall.
- How large the difference in actual fall rates is: the abstract reports an overall rate of 23.64% and odds ratios, but not the proportion who fell in the no-decrease group or in the groups that lost 5 cm or 20% of calf size. With an outcome this common, odds ratios overstate rate ratios.
- What the calf loss was made of: circumference includes muscle, fat, fluid and bone, and the abstract does not say how muscle loss was separated from the other components.
- Whether the middle bands behave like the extreme ones: results for losses under 3 cm, 3 to under 5 cm, under 10% and 10% to under 20% are not in the abstract that was read.
- How reliable self-reported falls are: falls were reported by participants in the 2021 wave, and the abstract does not give the recall window or say how falls were defined.
- Whether measuring calf circumference helps beyond the usual fall-risk tools: the paper reports odds ratios and does not, in its abstract, test whether adding the measurement improves prediction.
- Whether anything that slows calf loss prevents falls: no intervention was tested, so the study cannot say whether changing the measurement changes the outcome.
The Gale read
The strength of this paper is that it looks at change over time in a measurement that costs almost nothing to take, in a large nationally representative cohort, and that it states its method for handling people who could not be re-measured. The effects point in a consistent direction across absolute and relative change, and the confidence intervals at the extreme end do not touch 1. The care a reader owes it is of a familiar kind: odds ratios from a common outcome look larger than the difference in rates, a tape measure cannot tell muscle from other tissue, and a single wave of self-reported falls is a thin outcome. The reasonable reading is that a falling calf measurement travelled with higher odds of a reported fall in this sample, and that the case for using it in a clinic rests on evidence this paper does not yet provide.
Common questions
does losing calf muscle increase the risk of falling in older people
In this cohort, older adults whose calf circumference shrank over three years had higher odds of reporting a fall. A loss of 5 cm or more was associated with 32.5% higher odds than no decrease (odds ratio 1.33), and the study shows an association rather than a cause. Calf circumference also reflects fat and fluid as well as muscle, so it is an indirect stand-in for muscle.
what does calf circumference have to do with fall risk
The authors treat calf circumference as a simple, noninvasive proxy for lower-limb muscle mass. In 7,068 adults aged 60 and over, each 1 cm decrease over three years was associated with 3.2% higher odds of a fall (odds ratio 1.03, 95% confidence interval 1.02 to 1.05), and each 10% relative decrease with 10.3% higher odds (odds ratio 1.10).
how many older adults fall in three years
In this Chinese cohort, 1,671 of 7,068 participants (23.64%) reported an incident fall over a median follow-up of 3.07 years. That figure comes from self-reports in one survey wave in one country, so it is not a rate for every older population.
Sources
- 1.Zhang C, Shen J, Zhang Y, Huo Z, Zeng Y, Zhang J, Yao Y. Longitudinal Change in Calf Circumference and Incident Falls in Older Adults. JAMA Network Open. 2026;9(10):e2640127. doi:10.1001/jamanetworkopen.2026.40127 link
1 sources, numbered by first appearance. General health information, not medical advice. AI-assisted editorial content — every citation independently verified. Editorial policy
More from Gale Staff
Research
Top-Third Vitamin D Came With 19 More Minutes of Sleep a NightResearch
A Stroke Drug Beat Placebo, 69.7% to 56.3%; An Earlier Trial Did Not
Published October 24, 2026 · Gale Staff