Abstract
BACKGROUND AND OBJECTIVES: Broad Autism Phenotype (BAP) refers to specific behaviors or minor deficits in social/communication skills not meeting the diagnostic criteria for autism spectrum disorders (ASDs). The BAP Questionnaire (BAPQ) measures aloof personality (AP), rigid personality (RP), and pragmatic language (PL) deficits of BAP. The study objective is to translate and validate BAPQ into Malayalam, a South Indian language, and to establish new cut-off scores for detecting BAP in the Indian population. METHODS: BAPQ was translated into Malayalam (BAPQ-M), following which it was tested for reliability and internal consistency and factor analysis was conducted in 40 parents of ASD and neurotypical children. The receiver operating characteristic curve analysis generated new cut-off scores. RESULTS: The content validity index was ≥ 0.8 for each item. For the self and informant versions, Cronbach's α was 0.753 and 0.807, respectively, while Spearman's correlation coefficient was 0.953 ( P < 0.001) and 0.918 ( P < 0.001). The items demonstrated good communality (0.532-0.870), and on factor rotation, 12 components had eigenvalues > 1, which explained 72.3% of the total variance. The new cut-off scores for total estimate, AP, RP, and PL subscales were 2.60 (sensitivity 0.825; specificity 0.575; odds ratio [OR] 6.38, 95% confidence interval [CI] 2.28-17.84; P < 0.001), 2.60 (sensitivity 0.60; specificity 0.75; OR 4.50, 95% CI 1.73-11.70; P 0.002), 2.90 (sensitivity 0.8; specificity 0.475; OR 3.62, 95% CI 1.342-9.77; P 0.011), and 2.70 (sensitivity 0.5; specificity 0.825; OR 4.71, 95% CI 1.69-13.13; P 0.003), respectively. CONCLUSIONS: BAPQ-M has good psychometric properties for determining the presence of BAP among the people of Kerala, a South Indian state.