The Mini Nutrition Assessment (MNA) is a widely used and valid tool for screening and assessment of malnutrition among the elderly population worldwide. However, MNA has not been validated among the Ethiopian elderly population and this study assessed the validity of the tool for the target population. Cross-sectional validation study design employed to validate MNA in Meki town, East Ethiopia. This study included 176 randomly selected elders living in the community, whereas amputated, bedridden, visible deformity, known liver and/or renal disorders were excluded. The original MNA questionnaires were translated to local language and administered to each participant after doing the pretest. The anthropometric, self-perception of nutritional status and serum albumin concentrations were measured. Reliability, validity, sensitivity, specificity, Positive Predictive Value (PPV), and Negative Predictive Value (NPV) were calculated. Receiver-operating characteristic (ROC) curve analysis was plotted to identify the area under the curve (AUC) and optimal cut-off value for the prediction of malnutrition. A total of one hundred and seventy-six elders participated in this study. Of the total participants, 78(44.3%) were males. The mean (SD) age of the participants was 67.6 (±5.8) years and ranged from 60 to 84 years. The prevalence of malnutrition based on the MNA criteria (MNA < 17 points) was 18.2%, and 13.1% based on serum albumin concentration (<3 g/dl).The MNA had an overall Internal consistency of Cronbach's alpha 0.61. The tool also demonstrated significant criterion-related validity (0.75, p<0.001) and concurrent validity (0.51, p<0.001) with serum albumin concentration and self-perception of nutritional status respectively. Using the original cut-off point, the sensitivity, specificity, PPV and NPV of the tool were 93.5%, 44.6%, 65.4% and 86.0%, respectively. By modifying, the cut-off point to a value of <20.5, the sensitivity and specificity of the tool increases to 97.6% and 82.8% respectively. The AUC (95%CI) showed an overall accuracy of 92.7% (88.5, 96.9). The MNA tool can be used as a valid malnutrition screening tool for the Ethiopian elderly population by modifying the original cut-off point.