The Development of a Model to Predict Cognitive Decline Within 12 Months in Home Care Clients.
Gary Cheung, Ruth Teh, Eamon Merrick and 2 others
PMID 40103181WHAT IT FOUND
A model using routine home-care assessments flagged clients without a dementia diagnosis who were likely to decline cognitively within 12 months.
It was moderately accurate, so it can prompt further review rather than diagnose decline.
Key findings
01The model separated clients who did and did not worsen on the Cognitive Performance Scale, with a c-statistic of 0.70 in the derivation cohort and 0.69 in the validation cohort.
02Baseline Cognitive Performance Scale score 1, difficulty understanding others and age were among the most significant predictors of decline within 12 months.
03In the validation cohort, the model both under- and overestimated the observed rate of Cognitive Performance Scale decline.
STILL TO COME
How it was doneWhat they foundWhat it means for RNs
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 model was developed and internally validated in one New Zealand home-care dataset, and the authors recommend further validation in other cohorts. Discrimination was only moderate (c-statistic 0.69 in validation), and calibration both underestimated and overestimated observed decline. The data were deidentified and could not be linked to mortality or admission to long-term care, so the study did not address people with rapid deterioration leading to those outcomes. The sample included heterogeneous clients, such as younger adults and people with cancer, and some predictors had counterintuitive negative associations. Possible multicollinearity remained between interRAI scales despite checks. The model did not include biomarkers such as neuroimaging, amyloid-beta or apolipoprotein E genotype 4. The study period was 2013 to 2021, and overall rates of cognitive impairment over time in New Zealand are unknown. The authors state the sample size may overestimate the incidence of cognitive decline. No external validation in another country or care system was reported.
Declared interests
The authors declare no conflicts of interest.
The easy way to misread this
Do not treat a high predicted risk as a diagnosis of dementia or proof that an intervention will prevent decline. The model was developed and internally validated on routine home-care assessment data. Its validation accuracy was moderate, and calibration both underestimated and overestimated observed decline.