Background: Hospitalization in older adults with cognitive impairment is a seminal event that often leads to iatrogenic complications and exacerbation of behavioral disturbances that require a higher level of observation to ensure the patient’s safety. Our study aimed to determine whether an innovative model, using Patient Engagement Specialists (PES) improves care for hospitalized older adults with cognitive impairment who require a higher level of observation.
Methods: A one-year retrospective study was conducted of all patients 65 and older admitted to the medicine service requiring a higher level of observation (constant or enhanced observation) due to dementia and/or delirium at a large tertiary medical center. Patients were located either on the intervention unit or usual care unit. The intervention unit was constructed of: PES, geographic cohorting, multidisciplinary approach, and education implementation. The primary outcomes included: in-hospital mortality, length of stay (LOS), and readmission. Secondary measures included: the presence and duration of constant/enhanced observation, delirium preventative strategies, and management of behavioral disturbances.
Results: Of the 1,270 patient admissions that required constant and enhanced observation, 712 were in the intervention unit and 558 on the usual care units. In the intervention group, the average age was 81.0 years, 53.9% were White, 34.1% were married and 38.6% had dementia on admission. There was no difference in patient demographics between the intervention and usual care units. Patients admitted to the intervention unit had lower in-hospital mortality (1.1% vs. 3.8%, p=.03) and decreased LOS (4.9 vs. 5.9 days, p=.002). There was no difference in discharge home (p=0.44) or 30-day readmission rates (p=0.78). Patients were significantly less likely to be placed under either constant observation or enhanced observation in the intervention unit as compared to the usual care units (12.1% vs. 46.4%, p=0.001 and 20.1% vs. 79.9%, p<0.001, respectively). Patients on the intervention unit were less likely to have an order for benzodiazepines (25.8% vs. 37.1%, p<.001), NPO (29.1% vs. 39.4%, p=.001), bedrest (16.7% vs. 25.4%, p<.001), antipsychotics (42.0% vs. 57.0%, p<.001), and restraints (2.9% vs. 7.9%, p<.001), respectively.
Conclusions: Admission to the PES unit was associated with improved management of behavioral disturbances and overall clinical outcomes. The need to develop interventions for cost-effective, sustainable hospital “sitter” models is imperative to prevent harmful outcomes and to ensure quality care, dignity and respect for this vulnerable population and their caregivers. The PES model may offer a new paradigm in the management of hospitalized older patients with cognitive impairment and behavioral disturbance.