Mortality Risk in Homebound Older Adults Predicted From Routinely Collected Nursing Data.
Suzanne S Sullivan, Sharon Hewner, Varun Chandola and 1 others
PMID 30531348WHAT IT FOUND
Routinely collected home health nursing data identified 73% of older adults who died within a year.
A high-risk path combined likely to remain fragile or die, an ADL score below 21.5, cancer, and oxygen use.
Key findings
01The best decision tree identified 73% of people who died within 12 months and missed 27%.
02A high-risk path combining likely to remain fragile or die, ADL score below 21.5, cancer, and oxygen use had an 87.9% probability of death.
03The logistic regression model correctly identified 64% of deaths and missed 36%.
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 data were retrospective and not collected for research, so routine home health records may be incomplete or inconsistent. The models were cross-validated but not externally validated, so performance in a new home health population is uncertain. The analysis over-sampled deaths to create balanced datasets, which may not reflect the real distribution of risk. False positive rates were substantial, so many people predicted to be at risk may not die within 12 months.
Declared interests
The authors report no conflicts of interest. The publication types indicate NIH extramural research support.
The easy way to misread this
Do not treat the model as a validated tool for deciding who will die or when to start advance care planning. It was developed from retrospective home health records, not externally validated, and the models had false positive rates ranging between 30 and 42%.