Limb accelerations during sleep are related to measures of strength, sensation, and spasticity among individuals with spinal cord injury.
Stephanie K Rigot, Michael L Boninger, Dan Ding and 3 others
PMID 36329467WHAT IT FOUND
Sleep movement data from ankle accelerometers predicted strength and sensation scores with high accuracy.
This suggests passive overnight monitoring could offer a new, unobtrusive way to assess impairment in spinal cord injury without requiring active patient participation.
Key findings
01A model using 16 features from sleep accelerometry explained 68.7% of the variance in lower limb strength.
02A model using 15 features from sleep accelerometry explained 73.3% of the variance in lower limb sensation.
03Sleep movement data correctly classified spasticity categories with 81.6% overall accuracy.
STILL TO COME
How it was doneWhat they foundWhat it means for PTsWhat it means for OTs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The sample was heavily skewed toward male veterans, which may limit how well the findings apply to other groups with spinal cord injury. Participants with severe spasticity were underrepresented, so the models may not accurately predict spasticity in patients with high muscle tone. The study used a cross-sectional design, meaning it showed a relationship between sleep movements and clinical scores but did not prove that sleep movements cause or directly reflect changes in impairment over time. The models were not validated on a separate, unseen test set, which means their real-world predictive accuracy might be lower than reported.
Declared interests
The study was funded by the National Institutes of Health and the U.S. Department of Defense.
The easy way to misread this
Do not interpret these accuracy rates as evidence that sleep monitoring can replace clinical assessment. The study used a small, specific sample and did not validate the models on independent data, so the high predictive power may not hold in a broader clinical population.