Transferring nursing home residents to emergency departments by emergency physician-staffed emergency medical services: missed opportunities to avoid inappropriate care?

ABSTRACT

Background

The decision to transfer a nursing home (NH) resident to an emergency department (ED) is multifactorial and challenging but many of the emergency physician-staffed emergency medical service (EP-EMS) interventions and ED transfers are probably inappropriate.

Methods

We conducted a retrospective, cross-sectional study in three EP-EMSs in Belgium over a period of three years. We registered indicators that are potentially associated with inappropriate transfers: patient characteristics, availability of written do not resuscitate (DNR) orders or treatment restrictions, involvement of a general practitioner (GP) and availability of transfer notes. We also explored the association between age, the Charlson Comordity Index (CCI), polypharmacy, dementia, and the availability of DNR documents.

Results

We registered 308 EP-EMS interventions in NH residents. In 98% the caller was a health-care professional. In 75% there was no GP present and 40% had no transfer note. Thirty-two percentage of the patients had dementia, 45% had more than two comorbidities and 68% took five medications or more. In 6% cardiopulmonary resuscitation was performed. DNR orders were available in 25%. Eighty-eight percentage of the NH residents were transferred to the ED. Forty-four percent had a CCI >5. In patients of ≥90 years, with a CCI >5, with dementia and with polypharmacy, DNR orders were not available in 81%, 67%%,and 69%, respectively.

Conclusions

Improved EMS dispatch centre-NH caller interaction, more involvement of GP’s, higher availability of DNR orders and better communication between GPs/NHs and EP-EMS could prevent inappropriate interventions, futile prehospital aactions,and ED transfers.

KEYWORDS:

• Do-not-resuscitate (DNR)
• emergency department
• nursing home
• transfer
• emergency medical services (EMS)

Acknowledgments

We thank Marieke Delvoye, Evelien D’hondt, Annelore Gemmeke, Anke Goyvaerts Nele van Gaver, Evi Van den Kerckhove and Ragna Verlent for helping with data collection, and Joo-Ree Melis for entering the data.

Disclosure statement

No potential conflict of interest was reported by the author(s).