Mental Status Assessment in the Emergency Department: Ready for Prime Time as a Quality Metric?

Background:

Delirium and other forms of cognitive impairment are clinically important patient charactenistics that have been shown to be strong predictors of outcomes. Although mental status assessment is the subject of national quality indicators, considerable heterogeneity exists in mental status assessment, and little is known about screening effectiveness. One common mental status exam involves assessing whether a patient is oriented to person, place and time (0x3). We estimated 0x3 screening rates among patients presenting to the ED with a chief complaint pertaining to altered mental status and examined the sensitivity of 0x3 with respect to patient self‐reported mental status changes.

Methods:

Data on emergency department (ED) visits come from the National Hospital Ambulatory Medical Care Survey (NHAMCS) for the years 2005–2006. The NHAMCS is a nationally representative probability survey of ED and outpatient visits of participating U.S. hospitals, Mental status impairment was measured using patient self‐reported reason for ED visit based on a keyword search of the following terms: delirium, dementia, disorientation, confusion, altered mental status, delusion, hallucination, hearing voices, and psychosis, We grouped the first 5 terms into a general category and the last 4 terms into a psychosis category. Assessment of mental status was determined on the basis of a variable, indicating whether a patient was oriented to person, place, and time. Patients with blank or unknown values were considered unscreened. We calculated population weighted screening rates among patients with mental status decompensation by mental status subgroup and overall. Among those assessed, we calculated population‐weighted sensitivity rates, assuming that patient self‐reported reason for visit is the gold standard.

Results:

The rates of mental status assessment were 78.25% (SE = 2.66) among patients with general mental status complaints, 75.43% (SE = 4.84) among patients with psychotic symptoms, and 77.12% (SE = 2.52) overall among patients with any mental status complaints. Of those who had documented 0x3 screening, sensitivity rates were 50.15% (SE = 4.27) among patients with general mental status complaints, 17.32% (SE = 4.28) among patients with psychotic symptoms, and 40.73% (SE = 3.88) overall.

Conclusions:

A quarter of patients presenting to the ED for alterations of mental status did not have documentation of mental status assessment by 0x3. The sensitivity of 0x3 was only 50% in patients with general mental status complaints and much lower in patients with psychotic features. It is plausible that some patients may have returned to their cognitive baseline on assessment, but it may be that using 0x3 as the lone documentation for mental status is inadequate. Qualitative studies are needed to define a standardized way for documentation before mental status is included as a quality metric.

Author Disclosure:

E. Marlow, none; J. Chung, none.