A Fast Clinical Screening of Higher Cerebral Functions: From MMSE to BNIS
Jean-Luc Truelle, MD*
Matei Marinescu, MD*
Robert Rusina, MD**
*Department of Neurology, Hôpital Foch, Suresnes, France
**Neurologic Clinic, Thomayerova Hospital, Prague, Czech Republic
The French translation of the BNI Screen for Higher Cerebral Functions (BNIS) was administered to 102 patients with traumatic brain injury and to 100 normal subjects. It was also administered to 50 individuals with known psychiatric disorders. The French translation of the BNIS appears to be both a reliable and valid measure of disturbances of higher cerebral functioning. The mean time for both neurological and psychiatric patients to complete the BNIS was 13.8 minutes compared to a mean of 11.2 minutes for normal controls. The sensitivity and specificity of the BNIS are more promising than those of the Mini Mental State Examination for this application.
Key Words: BNI Screen, higher cerebral functions, Mini Mental State Examination, reliability, validity
Until 1947 mental functions were primarily evaluated using tests such as the Wechsler-Bellevue scale, which originated from psychiatric practice. In 1947, however, Ward Halstead described a series of tests specifically developed to evaluate persons with brain disorders. In the 1950s and 1960s, Benton in the United States and Luria in Russia developed tests devoted to evaluating cortical functions. These tests were intended to localize lesions to account for various disturbances.
In 1975 Folstein and coworkers published the Mini Mental State Examination (MMSE), a screening tool designed to assess the cognitive state of hospitalized patients and outpatients. This tool was quickly embraced by clinicians because it was easy and could be administered in 10 to 15 minutes. Those two qualities, essential in a typical clinical practice, compensated for certain limitations. The sensitivity of the MMSE to mild disturbances of higher functions is mediocre, and it is unsuitable for aphasic patients. However, no global short or simple tests were available for the evaluation of higher cerebral functions until the BNI Screen (BNIS) for Higher Cerebral Functions was introduced in 1991.
The BNIS assesses both quantitative and qualitative information. The Total score on the BNIS is the sum of seven subscores related to seven functions: language, orientation, concentration, visuospatial function, memory, affect, and self-evaluation of performance. The seven subscores are calculated by summing the scores of the 38 items of which they are composed. The BNIS is easy to administer and requires no training for an experienced clinician.
First, a patient’s level of concentration, language, and degree of cooperation are evaluated to determine whether he or she can perform the BNIS in a valid way. Language function is assessed: oral expression through verbal fluency, the search for paraphasia and dysarthria, aural comprehension, and handwriting through copying and dictating. Of the 50-point maximum score, 15 points are allocated to language. Sequential evaluations of orientation (right and left, spatial, and temporal) follow. Concentration is evaluated with digit span forward and backward items. Visuospatial functions are studied through visual exploration, visual sequences, and copy and pattern recognition. Memory is explored through memory recall and the digit-symbol association and through self-evaluation of memory performance. Reasoning is evaluated with tests of arithmetic, continuation of a logical sequence, and identifying similarities within several series. The assessment of affect is the original contribution of the BNIS compared to the MMSE. Affect is assessed by the analysis of affective expression, spontaneous emotion, emotion perception, and mood control.
The English version of the BNIS has been validated in 500 patients with known brain disorders and 200 normal control subjects. Unlike the MMSE, a database is available for the BNIS by which comparisons can be made to various groups. Based on a comparison of patients with known brain disorders and those with no history of brain disorder, a score of 47 of 50 points indicates normal higher cerebral functioning in a young population (15 through 39 years). The BNIS manual provided T scores for three different age ranges using the Total score obtained on the BNIS.
In 1998 J. L. Truelle, P. A. Joseph, and J. M. Mazaux translated the BNIS into French. The first validation of the French translation was based on 102 head-injured patients, 100 normal patients, and 50 patients with a psychiatric diagnosis. The findings on the French translation correlated with those obtained by Prigatano on similar populations.[6,7] The interrater reliability also was satisfactory (mean on English version, 37.0, SD=7.58; mean on French translation, 36.07, SD=8.12). In the study on the French translation of the BNIS, the mean age of the head-injured patients was 53 years and the mean length of time the head injury was present before testing was 75.2 months. The three most common diagnoses of the French patients were stroke, severe head injury, and multiple sclerosis.
The significant difference between the control group (mean, 48.2 points) and the patients (mean, 36 points) supports the specificity (82%), accuracy (95%), and high positive predictive value (91%) of the French translation of the BNIS. There were no significant demographic differences between patients and control subjects.
BNIS Compared to MMSE
The mean time needed to administer the French translation of the BNIS was 13.8 minutes in the neurological and psychiatric populations and 11.2 minutes in the normal population. This is slightly more time than is needed to administer the MMSE (7 to 8 minutes).
Conversely, in these populations, the French translation of BNIS was more accurate (88%) than the MMSE (68%) in terms of both positive and negative predictive value. The relative difficulty of a few items on the BNIS tends to make it slightly less specific than the MMSE. The MMSE is limited in a psychiatric environment while the BNIS can be used to evaluate deficits in patients with schizophrenia or patients with thymic, exogenic, or reactional disorders.
Twenty-five years after the MMSE was published, there was an obvious need for a screening test of higher cerebral functions that could be performed rapidly and that would reflect advances in our understanding of deficits in these functions. The battery of tests on the BNIS is more accurate in detecting changes in higher cerebral functions than the MMSE. It may serve as a guide in further investigations and help to define the medical treatment or neurorehabilitation of patients with brain injuries from a variety of causes.
- Folstein MF, Folstein SE, McHugh PR: “Minimental state.” A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12:189-198, 1975
- Halstead WC: Brain and Intelligence: A Quantitative Study of Frontal Lobes. Chicago: University of Chicago Press, 1947
- Lezak MD: Neuropsychological Assessment. New York: Oxford University Press, 1995
- Prigatano GP: Higher cerebral deficits: History of methods of assessment and approaches to rehabilitation: Part II. BNI Quarterly 2(4):9-17, 1986
- Prigatano GP: BNI Screen for Higher Cerebral Functions: Rationale and initial validation. BNI Quarterly 7(1):2-9, 1991
- Prigatano GP, Amin K, Rosenstein LD: Validity studies on the BNI Screen for Higher Cerebral Functions. BNI Quarterly 9(1):2-9, 1993
- Prigatano GP, Amin K, Rosenstein LD: Administration and Scoring Manual for the BNI Screen for Higher Cerebral Functions. Phoenix, AZ: Barrow Neurological Institute, 1995