Decoding machine learning in nursing research: A scoping review of effective algorithms.
Jeeyae Choi, Hanjoo Lee, Yeounsoo Kim-Godwin
PMID 39294553WHAT IT FOUND
In nursing machine learning studies, random forest was the most common algorithm and was reported as the most effective in six of the eleven studies that named a best algorithm.
No study tested whether these models changed patient outcomes.
Key findings
01Random forest was used in 24 of the 26 included studies.
02Among the 11 studies that reported a most effective algorithm, six reported random forest as most effective.
03None of the 26 included studies used an experimental design or measured patient or health outcomes.
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 review included 26 studies, but none used an experimental design or measured patient or health outcomes, so it cannot show that machine learning models improved care. Only English-language studies published from January 2019 to December 2023 were included, so relevant non-English or earlier work may be missing. Study quality was judged with MERSQI, a tool developed for medical education research, and other tools might have scored the studies differently. The included studies used datasets rather than traditional participant samples, and dataset sizes varied widely from 102 to 727,676, so the review does not describe a single patient population. Eleven studies reported which algorithm was most effective, and the review did not pool results because performance measures and nursing areas differed.
Declared interests
The authors declare no conflicts of interest.
The easy way to misread this
Do not read random forest as a proven clinical tool for nurses. It was reported as most effective in six of the eleven studies that named a best algorithm, but none of the 26 studies used experimental design or measured patient or health outcomes.