Development of Prediction Models for Healthy Ageing in Community-Dwelling Middle-Aged and Older Adults: A Longitudinal Study Using Machine Learning.
Daniel E C Leme, Adriane R Costodio, Cesar de Oliveira and 1 others
PMID 40945924WHAT IT FOUND
Older adults predicted to age healthily had higher household wealth, younger age, internet use, normal chair-rise speed, physical activity, no pain, high life satisfaction, and no abdominal obesity.
Key findings
01The random forest model was the best performer, with a ROC area of 0.78 and the highest positive predictive value reported.
02Household wealth, chronological age, internet use, physical performance, life satisfaction, physical activity, balance, pain status, and abdominal obesity were the most significant predictors.
03Higher household wealth, perceived and chronological age 50 to 59 years, normal physical performance, absence of pain, and extreme life satisfaction contributed positively; physical inactivity, lack of internet or email use, and abdominal obesity contributed negatively.
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
Many predictors were self-reported, so measurement error may bias the results. The healthy ageing definition did not cover all WHO criteria and missed some chronic diseases. Follow-up was only four years, which may not capture longer-term healthy ageing changes. Participants lost to follow-up may differ from those retained, introducing bias. The models were not externally validated in another dataset. The outcome was imbalanced, with fewer healthy ageing cases, which can affect some model metrics. The study is predictive and observational, so it does not show that changing a predictor causes healthy ageing.
Declared interests
The authors declared no conflicts of interest. The article is listed as having non-U.S. government research support.
The easy way to misread this
Do not conclude that changing household wealth, internet use, chair-rise performance, physical activity, balance, pain, or abdominal obesity will cause healthy ageing. This was a prediction model in one observational dataset, and the models were tested only internally.