Changes in gait can predict conversion to Parkinson’s

Original article: Gait analysis with wearables predicts conversion to Parkinson’s disease, Ann Neurol: July 11, 2019. 

The takeaway

A large longitudinal study has found gait characteristics which predict the conversion to Parkinson’s up to 4 years before the appearance of typical symptoms prompting a diagnosis.

Why is it important?

Subtle changes in gait, which may otherwise go unnoticed, can be tracked by wearables. Understanding and identifying when someone is in the prodromal phase is important – the earlier these changes are detected, the earlier it may be possible to intervene and halt Parkinson’s in its tracks.


Parkinson’s has a prodromal phase, that is, a period during which the disease has started but the definitive symptoms on which a concrete diagnosis is based have not yet emerged. The current study focused on changes in walking characteristics, or gait, and asked whether in a sufficiently large group of healthy participants monitored using wearables over a long enough period of time, subtle changes in gait could predict the development of Parkinson’s. 

The details

The study based in Tuebingen, Germany, recruited 696 healthy individuals, between the ages of 50-80 years, who underwent an evaluation of their gait characteristics at 2-year intervals between 2009 and 2016. The participants also underwent a number of tests and assessments which included the Unified Parkinson’s Disease Rating Scale (UPDRS), sense of smell, mood, digestive and urinary changes, and sleep among others.

On assessment days, participants wore a device strapped on their back roughly at the level of their beltline, and were asked to walk in a straight line over a 20 m walkway at their usual speed, at fast speed, at fast speed while checking boxes on a sheet of paper on a clipboard (dual task 1), and another form of dual tasking which involved walking fast while counting backwards in sevens (dual task 2).

Analysing this huge amount of data revealed a combination of these gait characteristics which predicted Parkinson’s up to 4 years before the emergence of symptoms that permitted a diagnosis. Walking at usual speed, greater gait variability in step and swing time as well as asymmetry (step, swing and stance time and step length) were predictive. The group also found that these changes occurred in a progressive manner.

Next steps

The authors acknowledge that a range of different devices could be used to assess gait. However, these findings strongly suggest that changes in gait characteristics could be included in assessments for prodromal Parkinson’s.

Where can I learn more?

Original article: Del Din S, Elshehabi M, Galna B, Hobert M, Warmerdam E, Suenkel U, Maetzler W. July 11, 2019. Gait analysis with wearables predicts conversion to Parkinson’s disease.