Development and validation of a prediction model for activities of daily living dysfunction among stroke survivors: insights from the CHARLS cohort.
Hongying Ren, Wen Wang, Li Tao and 4 others
PMID 42324550WHAT IT FOUND
This model flags stroke survivors more likely to have current ADL difficulty using depression, older age, falls, hypertension, arthritis, lung disease, social contact, sleep and drinking.
It was not tested in practice.
Key findings
01The final model used ten variables: CES-D score, sleep time duration, age, alcohol consumption, lung disease, social contact, falls, hypertension, arthritis, and sex.
02The model’s ability to separate those with and without ADL dysfunction was acceptable in the training set (AUC 0.76, 95% CI 0.73-0.79) and validation set (AUC 0.76, 95% CI 0.72-0.81).
03CES-D score was the most influential predictor in the model, followed by age, falls, hypertension, and social contact.
STILL TO COME
How it was doneWhat they foundWhat 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 data were cross-sectional, so the model cannot show whether predictors came before ADL dysfunction or whether ADL dysfunction led to depression. All variables were self-reported, which may misclassify stroke history, ADL difficulty, depression, falls, or medical conditions. Stroke severity, lesion details, and rehabilitation interventions were not available, and stroke severity is a known predictor of post-stroke function. The model was only split-sample internally validated in the same CHARLS dataset, not externally validated in a separate cohort. The sample was middle-aged and older adults in China, so it may not apply to other populations or settings.
The easy way to misread this
Do not read this as proof that depression, falls, sleep or social contact cause ADL dysfunction, or that treating them will improve ADL. The data were cross-sectional and self-reported, and the model was only internally validated.