Reliability of Perceptual Assessment According to Speech Stimuli in Thai Patients with Cleft Lip and/or Palate

Cleft lip and/or palate (CLP) frequently leads to speech disorders, such as hypernasality, hyponasality, nasal air emission (NAE), and compensatory articulation disorders. Perceptual speech assessment is the gold standard for evaluating these features, especially in community-based programs. In Thailand, the Articulation, Resonation, Nasal Emission, and Nasal Turbulence Test is widely used, but its interrater reliability has not been formally assessed. This cross-sectional study included 32 Thai-speaking individuals (ages 3.9–25.8) with repaired cleft palate. Three experienced speech-language pathologists conducted perceptual assessments using standardized speech stimuli at word and sentence levels. Interrater reliability for hypernasality, hyponasality, NAE, and low-pressure consonant production was evaluated using intraclass correlation coefficients (ICCs). Hypernasality assessments showed moderate reliability, especially for voiced plosive phrases with high vowels (e.g., ICC = 0.738 for /bóp bìip bàup bûu bîi/). Hyponasality ratings varied, with better agreement for sentence-level stimuli (ICC = 0.592) than for individual nasal sounds. NAE results ranged from poor to moderate reliability, with the highest for /puu/ (ICC = 0.729). Low-pressure consonant assessments also showed variable reliability, with /ják jàj jók jâa/ achieving the highest (ICC = 0.751). Perceptual reliability varied substantially by speech feature and phonetic context. Voiced plosive phrases with high vowels showed the most consistent hypernasality ratings, whereas hyponasality, NAE, and low oral pressure contexts demonstrated lower and more variable agreement. These findings suggest that stimulus selection and rater calibration are critical for improving reliability, particularly in community-based settings.