[Adaptation and validation of a naming test for Spanish speakers: reliability and discrimination of patients with dementia and unilateral brain lessions].

INTRODUCTION: Psychometric studies with this confrontation- naming test have not been reported in Spanish speakers.
OBJECTIVES: To study the internal consistency (N = 304), test-retest (N = 33) and inter-scorer (N = 37) reliability on healthy subjects. To study the concurrent validity (preliminary data) in 46 patients with unilateral brain lesions (left: 18, right: 28) including healthy subjects (N = 101), and in 28 patients referred for suspected dementia. MATERIAL: the test was based initially on the nouns used by Oldfield and Wingfield in English. Data were obtained from consecutive patients and healthy volunteers. Lateral lesions were confirmed by neuroimages and surgery. The three groups of the validity study with lateral lesions were matched by gender, age, education and manual preference. Both groups of the dementia validity study were matched by gender, age, education and presence of vascular lesion. Patients who scored < 24 points in the mini- mental test and>9 points in the Blessed behavioral rating scale were operationally defined as patients with dementia (N=15). STATISTICAL ANALYSIS: For reliability studies: Cronbach's alpha, Pearson and intraclass correlation coefficients were used. Differences between scorers and between tests were analyzed by dependent sample Student's t tests. For the validity study with lateral lesions, ANOVA was employed. For the dementia validity study: independent sample Student's t test was used.
RESULTS: All reliability coefficients were > 0.82 without differences in the t tests between scorers and tests. Validity was verified by the significant differences obtained between all the pairs of groups analyzed. DISCUSSION: The test demonstrated a suitable validity and reliability.