Abstract
BACKGROUND: Poor cognitive function is associated with negative consequences across settings of care, but research instruments are arduous for routine clinical implementation. This study examined the association between impaired arousal, as measured using an ultra-brief screen, and risk of 2 adverse clinical outcomes: hospital length of stay and discharge to a skilled nursing facility (SNF). DESIGN, SETTING, AND PARTICIPANTS: A secondary data analysis was conducted using 2 separate groups of medical ward patients: a Veterans Affairs medical center in the northeast (N = 1487, between 2010 and 2012) 60 years and older and a large tertiary care, university-based medical center (N = 669, between 2007 and 2013) 65 years and older in the southeastern United States. MEASUREMENTS: The impact of impaired arousal, defined by the Richmond Agitation Sedation Scale as anything other than "awake and alert," was determined using Cox Proportional Hazard Regression for time to hospital discharge and logistic regression for discharge to a SNF. Hazard ratios (HRs) and odds ratios (OR) with their 95% confidence intervals (CI) are reported, respectively. Both models were adjusted age, sex, and dementia. RESULTS: The 2156 total patients included in these groups had a mean age of 76 years, of whom 16.4% in group 1 and 28.5% in group 2 had impaired arousal. In the first group, patients with normal arousal spent an average of 5.9 days (standard deviation 6.2) in the hospital, while those with impaired arousal spent 8.5 days (9.2). On any given day, patients with impaired arousal had 27% lower chance of being discharged (adjusted HR 0.73 (95% CI 0.63-0.84). In the second group, individuals with normal arousal spent 3.8 (4.1) days in the hospital compared with 4.7 (4.6) for those with impaired arousal; indicating a 21% lower chance of being discharged [adjusted HR 0.79 (95% CI 0.66-0.95). With regard to risk of discharge to SNF, those with impaired arousal in group 1 had a 65% higher risk than those without impaired arousal [adjusted OR 1.65 (95% CI 1.21-2.25)], and those in group 2 had a nonsignificant 27% higher risk [adjusted OR 1.27 (0.80-2.03)]. Because of the quality improvement nature, this analysis did not control for comorbidities, which is a significant limitation. CONCLUSIONS: In this study of over 2000 older hospitalized patients, the simple observation of an abnormal arousal level may be an independent predictor of a longer hospital stay and discharge to SNF.