評估營養師對於手寫與拍照飲食紀錄之一致性與準確性

Assessment of the Accuracy and Consistency on Paper-and-Pen and Photographic Dietary Records Among Dietitians

林佳璇(Chia-Hsuan Lin) ； 張毓泓(Yu-Hung Chang) ； 徐慧君(Hui-Chun Hsu) ； 李洮俊(Yau-Jiunn Lee)

《臺灣膳食營養學雜誌》 12卷1期 (2020/06) Pp. 25-35

https://doi.org/10.6709/TJD.202006_12(1).0003

摘要

食物秤重是記錄食物最精確的方式，但較不便且費時，臨床多以拍照及手寫飲食紀錄方式取代。但評估者是否能一致且準確的估算飲食紀錄中的份數，目前仍較少研究探討。本研究利用拍照及手寫飲食紀錄，探討營養師對於飲食紀錄的食物估算份數之一致性及準確性。手寫為使用標準工具的方式記錄，拍照是以45度角拍攝，並擺放比例尺入鏡。並根據衛生福利部國民健康署公告之食物代換表將秤量食物的熟重（標準重量）轉換成實際份數（標準份數）。研究對象為通過專門職業及技術人員高等考試之營養師，研究者以電子郵件取得營養師同意參與此研究後，以電子郵件請營養師估算份數後回收成果，部分結果再以執業性質進行比較。評估準確性是以估算份數與實際份數相差的絕對值除以實際份數之百分比，當誤差百分比小於10%表示準確性高；份數的一致性是以intraclass correlation coefficient（ICC）進行分析，當ICC大於0.75時，則有良好的一致性；以independent t-test分析不同工作領域營養師對飲食紀錄估算份數誤差之差異。共回收103份飲食紀錄，結果指出全部營養師對於拍照及手寫飲食紀錄之一致性分別為ICC：0.858及ICC：0.812，但分組後僅有非醫院營養師的手寫飲食紀錄一致性低於0.75（ICC：0.737）。對於2種飲食紀錄中的主食類只有手寫飲食紀錄的準確性較好，醫院營養師及非醫院營養師之估算平均誤差皆少於5%，至於豆魚蛋肉類與蔬菜類無論是使用那一種記錄方式，醫院營養師及非醫院營養師之平均估算誤差都大於10%。進一步分析估算準確的營養師人數，發現除了手寫紀錄的主食類之外，其餘豆魚蛋肉類及蔬菜類的拍照紀錄估算不準確的營養師人數比例均超過50%。本研究結果指出無論拍照或手寫飲食紀錄方式，雖然營養師估算飲食份數具有不錯的一致性，但準確性都有待改善。故探討透過飲食紀錄與份數估算的持續訓練以提升份數估算的準確度，仍為重要之培訓及研究的議題。

關鍵字

手寫飲食紀錄 ； 拍照飲食紀錄 ； 飲食評估 ； 營養師 ； 準確性 ； 一致性

並列摘要

It is recognized that the most accurate method of the dietary record is to weigh food compounds; however, it is inconvenient and time-consuming. Paper-and-pen and digital photography dietary records are the other two wildly used methods in clinical assessments; nonetheless, consistency and accuracy are not clear. This study aimed to investigate the accuracy and consistency in two of paper-and-pen and digital photography dietary records among dietitians. We used standard household measures (e.g., bowls and spoon) to build portion size in the paper-and-pen dietary record. The diet image was taken by a digital camera in 45°-angle along with a ruler for reference. The investigated diet was cooked and weighted and then turned into actual food exchanges with the reference of the Taiwan Food Exchange List. We sent these diet records to registered dietitians who worked in different places (e.g., hospitals or non hospital) by e-mail, and asked them to estimate the food exchanges. We firstly have an absolute value of difference between the estimated food exchanges and the actual food exchanges. Then we calculated the estimation error with the absolute value of difference divided by the actual food exchanges. An estimation error of ± 10% was recognized as an acceptable value. The intraclass correlation coefficient (ICC) was used to analyze the consistency of dietary records among dietitians. An ICC value above 0.75 was recognized as good consistency. Independent t-tests were used to analyze the difference between hospital and non-hospital dietitians. 103 dietitians responded to our study. Our results showed all dietitians had good consistency with the ICC value of 0.858 and 0.812 in digital photography and paper recording, respectively. However, when we further looked into the impact of the workplace, the ICC value decreased to 0.737 in non-hospital dietitians by paper-and-pen dietary records. Among all estimations, paper-and-pen dietary record had the best accuracy in grains estimation with less of 5% of difference; however, regardless of either dietary record method, the estimation difference in the proteins and vegetables could be increased more than 10%. Importantly, except the grains, there were more than 50% of dietitians who estimated inaccurately in the paper-and-pen dietary record. Although our study showed that most of the dietitians had good consistency in the dietary estimation, the accuracy needs to be improved. In the future, it would be needed to design a practical program to improve the estimation accuracy of dietary record among dietitians.

並列關鍵字

paper-and-pen dietary record ； digital photographic dietary record ； dietary assessment ； dietitians ； accuracy ； consistency