The Effect of an Impaired Arousal on Short- and Long-Term Mortality of Elderly Patients Admitted to an Acute Geriatric Unit

Objective: Impaired arousal is associated with negative outcomes in intensive care units, but studies in acute medical wards are scanty. The study aim was to evaluate the association between impaired arousal, as measured using an ultrabrief screen, and risk of both 1- and 6-month mortality and discharge to nursing home (NH) or hospice. Design: Prospective cohort study with 6-month follow-up. Setting: An acute geriatric unit (AGU) of a university-based hospital in Northern Italy. Participants: All patients aged 65 years or older, admitted to the AGU between September 2012 and February 2015. Measurements: The modified Richmond Agitation Sedation Scale (m-RASS) was used to assess patients' arousal; a score of 0 denotes normal arousal, scores ranging from +1 to +4 denote increased arousal, and scores ranging from -1 to -5 denote decreased levels. The association of m-RASS scores with 6-month mortality was assessed by a Kaplan-Meier analysis. The impact of impaired arousal, defined by the m-RASS as anything other than "awake and alert," was determined using Cox proportional hazard regression for 1- and 6-month mortality after admission and logistic regressions were used for discharge to NH or hospice. The models were adjusted for age, sex, dementia, Sequential Organ Failure Assessment score, and disability. Results: Patients (n = 2477) had a mean age of 84 years, and were predominantly women (59.8%). Impaired arousal on admission was present in 644 (25.9%) patients: 33 (1.3%) were comatose (m-RASS = -5), 56 (2.3%) awakened to pain only (m-RASS = -4), 43 (1.7%) were very drowsy (m-RASS = -3), 93 (3.8%) drowsy (m-RASS = -2), and 212 (8.6%) were slightly drowsy (m-RASS = -1), but there were also 110 (4.4%) patients with restlessness, 75 (3.0%) with agitation, 17 (0.7%) with severe agitation, and 3 (0.1%) with combative behavior. Globally, 337 patients died within 1 month and 689 patients within 6 months. After adjustment for covariates, patients with impaired arousal had a significantly higher chance of having died at 1-month (adjusted hazard ratio [HR] 1.56, 95% confidence interval [CI] 1.22-2.03) and 6-month follow-up (adjusted HR 1.31, 95% CI 1.10-1.57). Those with impaired arousal were more likely to be discharged to a new NH (odds ratio [OR] 1.75, 95% CI 1.19-2.57) or to hospice (OR 1.96, 95% CI 1.18-3.23) than those without impaired arousal. Conclusions: An abnormal arousal level is an independent predictor of increased risk of 1- and 6-month mortality and of discharge to a new NH or hospice. The assessment of arousal with m-RASS should be routinely performed on all older patients on admission to acute hospital wards to screen potentially critical conditions.