Introduction: Enhancing diabetes self-efficacy (SE) level can improve the self-management behaviour in patients living with diabetes mellitus (DM). This study aimed to translate and assess the psychometric properties of Malaysian version of diabetes Medication Understanding and Use Self-Efficacy Scale (M-MUSE). Methods: Following the translation of English version of MUSE to Malay language using established international standard translation guidelines, 252 adult diabetics (≥ 18 years old; DM type 1 or 2) attending the Endocrine Clinic at Hospital Kuala Lumpur, Malaysia, were recruited in this cross-sectional study. After testing the face and content validity, the psychometric properties of the final M-MUSE were evaluated using the Classical Test Theory (CTT) for reliability (Cronbach’s alpha () and intra-class correlation coefficient (ICC)) and construct validity (factor analysis (FA)). Results: The semantic and conceptual problems in M-MUSE were identified and modified by a qualified professional translation committee. The final version showed good reliability values for internal consistency (Cronbach’s = 0.89) and one month test-retest reliability (ICC=0.72). The Bartlett’s test of sphericity and the Kaiser-Meyer-Olkin tests proved the suitability of M-MUSE for factor analysis. The extracted single component M-MUSE (eigenvalue > 1) explained a total variance of 57.58% with an eigenvalue of 4.60. The two factor structures; namely taking medication (item # 1, 6, 7 and 8) and learning about medication (item # 2, 3, 4 and 5) explained a total variance of 59.25% with good factor loading values (ranged from 0.63 to 0.89 for taking medication, and 0.66 to 0.83 for learning about medication). Conclusion: The M-MUSE appears to be a linguistically reliable and valid measure that is conceptually equivalent to the original version. The M-MUSE can be used in Malaysian healthcare settings to evaluate the SE in understanding and using prescribed diabetes medications.