Computerised balance and posture analysis

A force platform records the centre of pressure while standing, quantifying sway area, sway velocity and weight distribution between the legs, under systematically varied conditions — eyes open and closed, firm and compliant surfaces.

Varying the conditions is what makes it diagnostic. Sway that worsens sharply with eyes closed points to a sensory or vestibular problem rather than a strength one, and that finding changes the programme.

Falls assessment in older adults, balance disorder after stroke and brain injury, Parkinson's disease, multiple sclerosis, peripheral neuropathy, vestibular disorders, and postural assessment in scoliosis and chronic back pain.

It is also used as the outcome measure for balance programmes, since it repeats identically and is not subject to observer optimism.

This system tells you what is wrong and how much. It does not fix anything, and a balance assessment that is not followed by a properly dosed balance programme has helped no one.

We report the numbers before and after. If they did not move, that is what we report.

Force plate measurement quantifies sway, weight distribution and the limits of stability — the things a clinical balance score approximates. It shows whether a person relies on vision, on the inner ear or on position sense.

That distinction changes the treatment: someone dependent on vision needs different training from someone with a vestibular deficit, and clinical tests alone often cannot separate them.

People can achieve the same balance score by entirely different means: one using ankle movement, another using hip movement, a third relying almost wholly on vision. The platform shows which.

That distinction is what makes the training specific. Someone compensating with vision needs eyes-closed and unstable-surface work; someone with an ankle strategy deficit needs something else entirely.

Sway measurements repeat well only when the test is done carefully. A systematic review of repeat-testing studies found acceptable reliability in most of them, but concluded that most sway measures need recordings of at least 90 seconds, averaged over three to five repetitions 1. A single short trial varies too much from one attempt to the next to compare admission with discharge 1.

On falls, a meta-analysis of 44 studies involving 7,176 older adults found that several sway measures, including sway area per unit of time and the speed of sway, distinguish people who fall from those who do not 2. That helps to identify who is at higher risk; it cannot say whether or when a particular person will fall, because falls also depend on medicines, vision, strength and the home.