目標：改良MNA (Mini-Nutritional Assessment)使其更適用於血液透析(hemodialysis, HD) 患者的營養評估。方法：本研究徵求中部某地區醫院門診洗腎中心血液透析患者100名(18-97歲，男43女57名)。經同意後，以短式及長式MNA-台灣二版評估營養狀況並與生化指標比較。MNA台灣二版為採用台灣老人體位指標的改良版。版本之修正以選擇較能反應HD病情的非短式題項(如O題)與較不反應HD病情的短式題項(如E題)互換題位，並檢定其改良成效。以Wilcoxon Signed-Rank Test進行事後檢定，比較短式MNA台灣二版各改良版本與長式MNA台灣二版的評估結果。以Pearson's相關分析長式及各短式MNA台灣二版的總分與血清白蛋白、血清肌酸酐、住院天數、及急診次數的相關性。結果：本研究發現短式MNA台灣二版用於評估血液透析病患時，以量表O題(自評營養狀況)取代(對換)E題(失智)，最能改進評估功能，且其評估成果與長式MNA台灣二版相當一致。結論：本研究的結果顯示MNA量表以O題(自評營養狀況)取代(對換)E題(失智)或D題(精神壓力)最能改進其血液透析患者的評估功能。此簡單的改良，使短式MNA量表可適用於台灣的血液透析患者。

To modify the Mini-Nutritional Assessment (MNA) so that it would be suitable for use in hemodialysis (HD) patients. Methods: Subjects were 100 HD outpatients (18-97 years of age, 43 men and 57 women) recruited from a district hospital in Central Taiwan. With their consent, we rated the nutritional status of each subject with the long-, short-, and modified short-form versions of the MNA and measured their biochemical values. We examined various alternative short-form versions of the MNA based on the dietary characteristics of HD patients. The study used the Wilcoxon Signed-Rank Test, Pearson’s correlation analysis and the cross-tabulation test to evaluate the ability of the short-form versions to predict the long-form MNA (T2). Results: Results showed that a modified short-form version that exchanged item O (self-rated nutritional status) for item E (dementia) improved the predictive ability of the MNA in HD patients. The modified short-form performed well compared to the long-form. Conclusions: The study demonstrated that replacing item O for E or D in the MNA can best improve the functionality of the MNA. This simple revision makes the short-form suitable for assessing the risk of malnutrition in patients on hemodialysis. (Taiwan J Public Health. 2012;31(2):176-183)