BMC Oral Health (Oct 2021)

Cross-cultural adaptation and validation of Malay version of Rapid Estimate of Adult Literacy in Dentistry (MREALD-30) among Orang Asli population in Malaysia

  • Avita Rath,
  • Melissa Wong,
  • Claudio Mendes Pannuti,
  • Priyadarshini Hesarghatta Ramamurthy,
  • Bennete Fernandes,
  • Amelia Shelton,
  • Khairiyah Abdul Muttalib

DOI
https://doi.org/10.1186/s12903-021-01866-9
Journal volume & issue
Vol. 21, no. 1
pp. 1 – 11

Abstract

Read online

Abstract Background The aim of this study was to adapt, translate and validate the Rapid Estimate of Adult Literacy in Dentistry (MREALD-30) instrument for the Orang Asli population in Malaysia. Methods After translation and cross-cultural adaptation, interviews were conducted with 326 participants of the Temuan tribe from village Kampung Tering in Johol, Negeri Sembilan, Malaysia. The instrument's validity was assessed using the scores of MREALD-30, which were compared based on occupation, monthly household income, educational attainment, general literacy, use of dental services, and three dental outcomes. A questionnaire containing socio-behavioral information and validated Malay Oral Health Impact Profile (M-OHIP-14) was also administered. The reliability of the MREALD-30 was assessed by re-administering it to 30 subjects after two weeks. Its correlations evaluated convergent and discriminative validity of MREALD-30 with the level of education and dental visiting habits, monthly household income, respectively. Predictive validity was assessed with M-OHIP-14, while construct validity was evaluated by exploratory factor analysis using the Rasch model. Results The internal consistency of the MREALD-30 measured by Cronbach's alpha was 0.89. The test–retest reliability was excellent (ICC 0.95, k = 0.85). MREALD-30 exhibited good construct validity. Rasch analysis showed two factors, and infit mean-square statistics for MREALD-30 were all within the desired range of 0.50–2.0. The discriminant validity and predictive validity were statistically significant (p < 0.05). Conclusions MREALD-30 showed very strong reliability, good construct, discriminant, and predictive validity, but poor convergent validity. Overall, it showed good psychometric properties and can be used in these community settings.

Keywords