Psychometric properties of an Arabic translation of the multidimensional assessment of interoceptive awareness (MAIA-2) questionnaire in a non-clinical sample of Arabic-speaking adults

BACKGROUND: Interoception refers to processes through which the nervous system identifies, analyzes, and integrates the information generated by the physiological state of the body (e.g., from internal organs such as the stomach, heart, or lungs). Despite its potential interest for clinical research and its wide use globally, no Arabic adaptation and validation of the Multidimensional Assessment of Interoceptive Awareness (MAIA-2) questionnaire exists to date. The goal of this study was to examine the psychometric properties of an Arabic translation of the MAIA-2 in a sample of Arabic-speaking community adults from Lebanon. We hypothesized that the Arabic version of the MAIA-2 would yield adequate internal consistency coefficients; the 8-factor structure model would show a good fit to our data, with measurement invariance and good convergent validity. METHOD: The Arabic adaptation of the MAIA-2 was developed using the forward–backward translation method. A non-clinical sample of Arabic-speaking adults (n = 359, 59.9% females, mean age = 22.75 years (SD = 7.04)) took part of this validation study. To check if the model was adequate, several fit indices were calculated: the normed model chi-square (χ²/df), the Steiger-Lind root mean square error of approximation (RMSEA), the Tucker-Lewis Index (TLI) and the comparative fit index (CFI). Values ≤ 3 for χ²/df, and ≤ 0.08 for RMSEA, and 0.90 for CFI and TLI indicate good fit of the model to the data. RESULTS: Confirmatory Factor Analyses corroborated the validity of the original 8-factor structure of the MAIA-2 [χ(2)/df = 1603.86/601 = 2.67, RMSEA = 0.068 (90% CI 0.064, 0.072), SRMR = 0.058, CFI = 0.903, TLI = 0.892]. Reliability estimates in our sample revealed good internal consistency, with McDonald’s ω coefficients for the subscales ranging from 0.86 to 0.93. Our analyses also revealed measurement invariance of the Arabic MAIA-2 for gender. No statistically significant difference between men and women in all dimensions, except for the not worrying and attention regulation subscales where men scored significantly higher than women. Finally, the Arabic MAIA-2 dimensions showed positive correlations with the intuitive eating dimension “Reliance on Hunger and Satiety Cues”, thus providing support for convergent validity. CONCLUSION: We contribute the literature by providing the first Arabic adaptation and validation of a measure assessing the multidimensional construct of self-reported interoception. The Arabic MAIA-2 demonstrated good psychometric properties. We thus preliminarily recommend its use to measure the interoceptive awareness construct among Arabic-speaking communities worldwide.