Implementation of a multiprofessional, multicomponent delirium management guideline in two intensive care units, and its effect on patient outcomes and nurse workload: a pre-post design retrospective cohort study

Aim of the study: Delirium is a frequent intensive care unit (ICU) complication, affecting 26% to 80% of ICU patients, often with serious consequences. This study aimed to evaluate the effectiveness, costs and benefits of following a standardised multiprofessional, multicomponent delirium guideline on eight outcomes: delirium prevalence and duration, lengths of stay in ICU and hospital, in-hospital mortality, duration of mechanical ventilation, and cost and nursing hours per case. It also aimed to explore the associations of delirium with length of ICU stay, length of hospital stay and duration of mechanical ventilation. Methods: This retrospective cohort study used a pre-post design. ICU patients in an historical control group (n = 1608) who received standard ICU care were compared with a postintervention group (n = 1684) who received standardised delirium management – delirium risk identification, preventive measures, screening and treatment – with regard to eight outcomes. The delirium management guideline was developed and implemented in 2012 by a group of experts from the study hospital. As appropriate, descriptive statistics and multivariate, multilevel models were used to compare the two groups and to explore the association between delirium occurrence and the selected outcomes. Results: Twelve percent of the 1608 historical controls and 20% of the 1684 postintervention patients were diagnosed with delirium according to the ICD-10 delirium diagnosis codes. Patients being treated for heart disease, and those with septic shock, ARDS, renal insufficiency (acute or chronic), older age and higher numbers of comorbidities were significantly more likely to develop delirium during their stay. Multivariate models comparing the historical controls with the post intervention group indicated significant differences in delirium period prevalence (odds ratio 1.68, 95% confidence interval [CI] 1.38–2.06; p <0.001), length of stay in the ICU (time ratio [TR] 0.94, CI 0.89–1.00; p = 0.048), cost per case (median difference 3.83, CI 0.54–7.11; p = 0.023) and duration of mechanical ventilation (TR 0.84, CI 0.77–0.92; p <0.001). The observed differences in the other four outcomes – in-hospital mortality, delirium duration, length of stay in the hospital, and nursing hours per case – were not significant. Delirium was a significant predictor for prolonged duration of mechanical ventilation and for both ICU and hospital stay. Conclusion: Standardised delirium management, specifically delirium screening, supports timely detection of delirium in ICU patients. Increased awareness of delirium after the implementation of standardised multiprofessional, multicomponent management leads to increased therapeutic attention, a prolongation of ICU stay and increased costs, but with no influence on mortality