Reducing emergency presentations from long-term care: A before-and-after study of a multidisciplinary team intervention

• Published in: Maturitas Vol. 117; pp. 45 - 50
• Main Authors: Connolly, M.J., Broad, J.B., Bish, T., Zhang, X., Bramley, D., Kerse, N, Bloomfield, K, Boyd, M.
• Format: Journal Article
• Language: English
• Published: Ireland Elsevier B.V 01-11-2018
• Subjects: Aged; Chronic conditions; Emergency Service, Hospital - statistics & numerical data; Hospitalization - statistics & numerical data; Hospitalizations; Humans; Long-Term Care; Nursing Homes; Patient Care Team; Nursing homes; Chronic conditions; Aged; Hospitalizations; Long-term care
• ISSN: 0378-5122; 1873-4111
• DOI: 10.1016/j.maturitas.2018.08.014

•The study assessed a multidisciplinary team outreach intervention targeted at selected conditions.•The intervention decreased avoidable emergency department presentations to from facilities with previously high admission rates.•This reduction, of approximately 25%, is of high clinical significance.
The complexity of care required by many older people living in long-term care (LTC) facilities poses challenges that can lead to potentially avoidable referrals to a hospital emergency department (ED). The Aged Residential Care Intervention Project (ARCHIP) ran an implementation study to evaluate a multidisciplinary team (MDT) intervention supporting LTC facility staff to decrease potentially avoidable ED presentations by residents.
ARCHIP (conducted in 21 facilities [1,296 beds] with previously noted high ED referral rates) comprised clinical coaching for LTC facility staff by a gerontology nurse specialist (GNS) and an MDT (facility senior nurse, resident’s general practitioner, GNS, geriatrician, pharmacist) review of selected high-risk residents’ care-plans. A before-after repeated measures analysis was conducted for 9 months before and 9 months after intervention commencement (a 29-month period because of staggered facility enrolment). Modelling was adjusted for time trend, seasonality, facility size, and cluster effect.
ED admission rate ratio post- versus pre-intervention was 0.75 (95% C.I. 0.63, 0.89, p-value = 0.0008), a 25% reduction in ED presentations post-intervention. A sensitivity model used a shorter, staggered time period centred on intervention start (9 months pre-intervention and 9 months post-intervention) for each facility, and a four-level categorical intervention variable testing intervention effect over time. The sensitivity test showed a 24% reduction in ED presentations in months 1–3 post-intervention (p-value = 0.07), a 34% reduction in months 4–6 (p-value = 0.01), and a 32% reduction in ED presentations in months 7–9 (p-value = 0.03). However, when the higher ED referral rates for 3 months immediately pre-intervention were modelled, the impact of the intervention on ED presentation rates reverted almost to previous levels.
A GNS-led MDT outreach intervention, targeted at selected conditions, decreases avoidable ED admissions of high-risk residents from selected facilities.