Reliability of nurses' neurological assessments in the cardiothoracic surgical intensive care unit.

BACKGROUND

Alterations in mental status are common among patients in the cardiothoracic surgical intensive care unit. Changes in mental status can be caused by metabolic factors, medications, or brain injury. In this setting, reliable, serial neurological evaluations are critical for assessing the effectiveness of treatment and the need for additional studies.

OBJECTIVES

To estimate the reliability of the Rancho Los Amigos Cognitive Scale and the newly developed Neurologic Intensive Care Evaluation as measures of cognitive function in the cardiothoracic surgical intensive care unit.

METHODS

Nurses used 1 of the 2 scales as part of routine neurological assessments of patients in the cardiothoracic surgical intensive care unit. For each test, scores of different observers were correlated and a reliability estimate formed.

RESULTS

Interrater reliability was high for both evaluations (Rancho scale, 0.91; Neurologic Intensive Care Evaluation, 0.94). Correlations between the scores of different pairs of observers were also high (mean rho values, 0.84 for the Rancho scale and 0.77 for the Neurologic Intensive Care Evaluation).

CONCLUSIONS

Both scales are reliable indicators of the neurological state of patients in the cardiothoracic surgical intensive care unit. These scales measure different, although limited, aspects of cognitive function. Each test was simple to administer and did not take more time than the standard nursing neurological examination. Most of the variability in scoring was related to the different degrees of stimulation used by examiners when assessing patients, not to differences in the interpretation of patients' responses.