Detecting Agitated Behaviors in Nursing Home Residents With Dementia Using Electronic Medical Records.
Hyunkyung Yun, Vincent Mor, Christopher Santostefano and 1 others
PMID 39349066WHAT IT FOUND
Adding electronic transfer forms and physician orders to routine assessments raised documented agitated behaviors in residents with dementia from 14.8% to 18.6% of person-months, but the sources overlapped little.
Key findings
01Combining INTERACT forms and physician orders with MDS assessments raised documented agitated behaviors from 14.8% to 18.6% of person-months.
02In July 2021, 12.2% had behaviors only on MDS, 2.0% only in physician orders, and 1.5% only in INTERACT forms, with less than 1% overlap between sources.
03Manual review of 302 keyword-flagged physician orders found 21 false positives, an error rate of 6.9%.
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 collected for pandemic infection management, not specifically for behavior detection. The MDS looked back 7 days, while EMR records covered any time in the assessment month. Nursing and physician progress notes were not included, so behavior capture may be incomplete. Facilities varied widely in use of INTERACT forms and orders, and some had no records from one or both sources. The orders keyword search produced false positives, with a 6.9% error rate on manual review. The study period overlapped with the COVID-19 pandemic, which may limit generalizability.
Declared interests
The authors declared no conflicts of interest. The listed research support includes NIH extramural and non-U.S. government support.
The easy way to misread this
Do not conclude that electronic records prove more residents actually had agitated behaviors. They show more behaviors were documented in some sources, and the sources captured different behaviors with little overlap.