BackgroundThe Brief Interview for Mental Status (BIMS) is the Centers for Medicare and Medicaid Services' primary cognitive assessment for multiple health care settings. However, research suggests that the BIMS has low sensitivity in detecting mild cognitive impairment in community samples and nursing home residents.ObjectiveTo determine the sensitivity of the BIMS in identifying patients with mild or moderate cognitive impairments in an inpatient rehabilitation setting and the relationship of BIMS and Montreal Cognitive Assessment (MoCA) scores with rehabilitation outcomes.DesignRetrospective cohort.SettingInpatient rehabilitation facility.PatientsA total of 2252 patients admitted for orthopedic or general rehabilitation.MethodsThe sensitivity and specificity of the BIMS for identification of cognitive impairments were determined using validated cutoff scores on the MoCA as criterion. Multivariable ordinal regression was employed to determine if MoCA and BIMS scores were independent predictors of rehabilitation outcomes.Main Outcome MeasurementsFunctional independence at discharge (GG scores), rehabilitation efficiency (GG change per day), length of stay (LOS), and discharge destination.ResultsThe BIMS had low sensitivity in identifying moderate (0.384) or mild (0.257) cognitive impairment. Patients classified as having moderate cognitive impairment on the MoCA had lower GG scores at discharge (odds ratio [OR] 0.37, 95% confidence interval [CI] 0.30 to 0.46) and rehabilitation efficiency (OR 0.43, CI 0.34 to 0.53) compared to individuals identified as impaired on the BIMS (GG scores: OR 0.63, CI 0.52 to 0.78; efficiency: OR 0.71, CI 0.58 to 0.86). LOS (OR 1.49, CI 1.19 to 1.85) and discharge destination (OR 0.47, CI 0.28 to 0.77) were uniquely predicted in patients identified as moderately impaired on the MoCA but not the BIMS (LOS: OR 0.99, CI 0.80 to 1.21; discharge home: OR 0.93, CI 0.62 to 1.38).ConclusionsThe BIMS had low sensitivity for detection of mild and moderate cognitive impairments. Compared to the MoCA, the BIMS was less effective at predicting rehabilitation outcomes.