RNOtherJournal of the American Medical Directors Association2021

Identifying Urinary Tract Infection-Related Information in Home Care Nursing Notes.

Kyungmi Woo, Victoria Adams, Paula Wilson and 6 others

PMID 33434568

WHAT IT FOUND

Home care nurses' notes mention UTI-related terms more often before UTI-related hospital or ED admission, and software can find them.

This is not a tested way to improve care.

Key findings

01The text-searching system identified UTI-related information in home care notes with very good performance (average F measure 0.9, 95% CI 0.87 to 0.93).

02UTI-related terms were significantly more frequent in episodes with UTI-related hospital or ED admission: 81% had at least one UTI-related category, compared with 21.6% of episodes without such events.

03UTI-related documentation increased before UTI-related hospital or ED admission, peaking within a few days before the event, and the increase was most evident in care coordination notes.

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.

Already have one?

What it does not show

The data came from one nonprofit home care agency in New York, so the notes and results may not match other home care systems. The software was tested on 300 notes from patients who had UTI-related ED or hospital admission, so its accuracy on patients without those events is not known. The language models did not include temporal terms. The reason for hospitalization was taken from nurse-documented OASIS fields, which the authors say can be erroneous or incomplete. The proposed alert workflow was hypothetical and not tested in a clinical trial. The authors did not explore the statistical significance of the association between UTI-related information in notes and negative outcomes.

Declared interests

No author disclosure statement is in the supplied text. The article is tagged as NIH and U.S. government research support.

The easy way to misread this

Do not treat the software as a tested UTI alert. It was not tested to improve care, and UTI misdiagnosis is common, so clinicians still need to review the patient before diagnosis or antibiotics.

Read it on PubMed →