Outcomes Following Implementation of a Hospital-Wide, Multicomponent Delirium Care Pathway

Abstract

BACKGROUND: Delirium is associated with poor clinical outcomes that could be improved with targeted interventions. OBJECTIVE: To determine whether a multicomponent delirium care pathway implemented across seven specialty nonintensive care units is associated with reduced hospital length of stay (LOS). Secondary objectives were reductions in total direct cost, odds of 30-day hospital readmission, and rates of safety attendant and restraint use. METHODS: This retrospective cohort study included 22,708 hospitalized patients (11,018 preintervention) aged ≥50 years encompassing seven nonintensive care units: neurosciences, medicine, cardiology, general and specialty surgery, hematology-oncology, and transplant. The multicomponent delirium care pathway included a nurse-administered delirium risk assessment at admission, nurse-administered delirium screening scale every shift, and a multicomponent delirium intervention. The primary study outcome was LOS for all units combined and the medicine unit separately. Secondary outcomes included total direct cost, odds of 30-day hospital readmission, and rates of safety attendant and restraint use. RESULTS: Adjusted mean LOS for all units combined decreased by 2% post intervention (proportional change, 0.98; 95% CI, 0.96-0.99; P = .0087). Medicine unit adjusted LOS decreased by 9% (proportional change, 0.91; 95% CI, 0.83-0.99; P = .028). For all units combined, adjusted odds of 30-day readmission decreased by 14% (odds ratio [OR], 0.86; 95% CI, 0.80-0.93; P = .0002). Medicine unit adjusted cost decreased by 7% (proportional change, 0.93; 95% CI, 0.89-0.96; P = .0002). CONCLUSION: This multicomponent hospital-wide delirium care pathway intervention is associated with reduced hospital LOS, especially for patients on the medicine unit. Odds of 30-day readmission decreased throughout the entire cohort.

COVID-19 severity and age increase the odds of delirium in hospitalized adults with confirmed SARS-CoV-2 infection: a cohort study

“…Both NuDESC and CAM-ICU results are entered into the EMR by bedside nursing staff each shift. At least 86% of patients underwent on average two in person delirium screens a day during the study period, which increased to 94% of patients when assessing screening compliance only during full calendar days (e.g., excluding day of admission or discharge), which is similar to prior compliance assessments [ 16 ]. The remaining patients underwent on average at least one in person delirium screen a day despite intensive hospital-wide infection control isolation protocols for patients with COVID-19.…”

“…However, the specificity of the delirium screens used in this study is high [ 14 , 19 ], so false positives are unlikely. In addition, presence of delirium has been shown to be associated with longer length of stay, and interventions that target delirium reduce length of stay [ 16 ]. Because we did not measure the date on which delirium was first diagnosed in each of our patients, we are unable to distinguish between delirium that started on hospital day 2 (and thus was more likely due to COVID-19) and delirium that started on hospital day 30 (and thus was more likely due to a hospital acquired complication).…”

“…The presence of delirium was identified using EMR data generated by pre-existing systematic, hospital-wide screening protocols. Our study leveraged the UCSF Delirium Care Pathway [ 16 ], which is an interdisciplinary, multicomponent pathway that includes in person twice-daily bedside delirium screening, as well as standardized, evidence-based non-pharmacologic delirium management recommendations.…”

“…Secondary analyses included measuring the association between age and presence and delirium; the associations between delirium and safety attendant use, restraint use, discharge to home as opposed to an alternative location, and length of stay were also assessed. In addition to being a hospital-wide quality improvement metric [ 16 ], we were particularly interested in restraint use since, due to infection control measures limiting the time staff could safely spend with COVID-19 patients, restraint use may be more common in adults with delirium during the pandemic.…”

KIBRA, MTNR1B, and FKBP5 genotypes are associated with decreased odds of incident delirium in elderly post-surgical patients

“…There is no medication that prevents or treats delirium at this time. Current delirium management focuses on non-pharmacological intervention, primarily focusing on frequent reorientation, mobilization, maintenance of sleep–wake cycles, avoiding constipation and urinary retention, and minimizing polypharmacy 31 , 32 . This multicomponent approach may reduce delirium incidence by 30–40% 31 .…”