"Alert and Oriented × 3?" Correlates of Mini-Cog Performance in a Post/Nondelirious Intensive Care Unit Sample.

Background: Cognitive impairment has been found to be a predictor of adverse medical outcomes, including nonadherence, recurrent medical crises resulting in early readmissions, and death.

Objective: The Mini-Cog has been proposed for bedside/clinic cognitive testing. Its validity as a measure of central nervous system (CNS) impairment has never been tested against measures of CNS-medical history, CNS scans, selected laboratory findings, observed in-hospital nondelirious memory impairment, or collateral history from family.

Methods: We observed Mini-Cog performance in 107 post/nondelirious medical intensive care unit patients and tested its association with age, CNS-medical history, CNS scans, selected laboratory findings, and behavioral history (in-hospital observation of memory problems and collateral history from family or significant others).

Results: The overall Mini-Cog covaried with age, various measures of CNS impairment, abnormal laboratory findings, and measures of preadmission "forgetfulness" per family and by in-hospital staff observation. Unique variance in predicting overall Mini-Cog scores included age, positive CNS scan, and behavioral history. Of 91 patients found to be "alert and oriented × 3," 76% were impaired in immediate memory, short-term memory, or clock drawing.

Conclusions: The Mini-Cog appears to be a brief, yet valid, measure of CNS dysfunction that significantly enhances sensitivity of evaluation at the bedside. Failure to evaluate patients with a formal examination like the Mini-Cog appears to miss up to 76% of patients with moderate cognitive impairment.