台灣第2型糖尿病人飲食頻率問卷之效度評估

Validity of a food frequency questionnaire in Taiwan adults with type 2 diabetes mellitus

陳虹均(Hung-Jiun Chen)

指導教授 ： 黃孟娟

高雄醫學大學/醫學院/醫學研究所碩士班/碩士(2017年)

摘要

飲食評估工具可以用來評估個人的攝取量及飲食習慣，其中飲食頻率問卷是一種成本低且容易執行的飲食評估工具，目前已有許多西方國家研究利用有效度的飲食頻率問卷蒐集病人的飲食資料，探討某些飲食生物標誌或營養素與飲食的相關性，但是因飲食文化的不同，這些飲頻率問卷並不完全適用於台灣人。在先前我們發表之兩篇報[1, 2]告中已有使用一份飲食頻率問卷探討第2型糖尿病人飲食模式與體內飲食相關生物標記與臨床生化值之關聯性，但其效度未經評估。故本研究的目的是為評估此份飲食頻率問卷之效度，並以抗氧化營養素作為飲食生物標誌，探討特定食物種類與飲食生物標誌之相關性。
本研究以横斷面探討研究飲食頻率問卷估算之熱量與三大營養素（醣類、脂肪、蛋白質），及三天24小時飲食回憶紀錄之熱量與三大營養素相關性評估。進一步，分析飲食頻率問卷與飲食相關飲食生物標誌之相關性。研究結果顯示飲食頻率問卷與三天24小時飲食回憶記錄之熱量、蛋白質、脂質與碳水化合物呈顯著相關性（r=0.438-0.574，p值皆為＜0.001），並且經Paired t test檢定後皆沒有達顯著差異性（p值=0.068-0.431）。進一步校正熱量之後，三大營養素之相關性（ r=0.587-0.651，p值皆為＜0.001）皆有增加。深海魚攝取頻率及深/淡色蔬菜攝取份數越高者與α-carotene、β-carotene之濃度呈現正相關趨勢（all p for trend＜0.05）。而油炸類食物攝取頻率與α-carotene (p for trend=0.022)呈現負相關趨勢。進一步以向後選取法（Backward selection）多變量線性回歸分析，將年齡、性別、FFQ估算熱量攝取、糖尿病罹病年及教育程度和各種食物攝取放入模式，發現每週深海魚攝取頻率(beta =2.372，p =0.010)、每週淡色蔬菜攝取份數(beta =1.845，p = 0.018)及深色蔬菜攝取份數(beta=2.099，p = 0.009)低中高三等份與α-carotene仍具獨立顯著正相關性。同樣也發現，每週深海魚攝取頻率(beta =7.902，p =0.041)、每週淡色蔬菜 (beta =6.704，p =0.039)及深色蔬菜攝取份數(beta =6.923，p =0.041)低中高三等份與β-carotene具獨立顯著正相關性。而在每週油炸類食物攝取頻率(beta =-1.937，p =0.022) 低中高三等份則與α-carotene具獨立顯著負相關。綜合上述結論，此份問卷是可以有效的用於評估台灣地區第2型糖尿病人之熱量與三大營養素，並且觀察到多攝取深海魚類、淡色蔬菜及深色蔬菜者其血漿α-carotene、β-carotene濃度顯著越高，而與多攝取油炸類食物者則血漿α-carotene濃度顯著越低。故本研究飲食頻率問卷可有效估算第2型糖尿病人之熱量與三大營養素以及特定健康食物攝取，包括深海魚類、淡色及深色蔬菜因此未來也可適用於健康食物攝取型態之評估。

關鍵字

第2型糖尿病 ； 飲食頻率問卷 ； 24小時飲食回憶記錄 ； 效度 ； 類胡蘿蔔素

並列摘要

Dietary assessment tools can be used to assess individual intake and dietary habits. A food frequency questionnaire (FFQ) is a diet assessment tool with low cost and is easy to implement. There are many studies of Western Countries using the valid FFQ to collect information about patient's dietary patterns and explore the relationship between biomarkers and nutrients. Due to different food cultures, the FFQ cannot be fully applicable to Taiwanese. In the previous two studies reported in Taiwan [1, 2], a developed FFQ was used to investigate the association between dietary patterns, dietary biomarkers and metabolic indicators in type 2 diabetes. In this study, a cross-sectional design was conducted to validate this FFQ. We aimed to examine and assess the correlation of energy and the three macronutrients (carbohydrate, fat and protein) between FFQ and those assessed using the three 24-hour dietary recalls. Furthermore, the association between food groups of the FFQ and dietary biomarkers was analyzed. Our results showed that the correlations of energy, protein, lipid and carbohydrate estimated using FFQ were significantly correlated with those by three 24-hour dietary recall (r = 0.438-0.574, p <0.001), and energy, and the three macronutrient levels assessed by the two methods were similarly based on paired t-test (p= 0.068-0.431). After adjustment for total energy intakes, the correlation of the three macronutrients (r = 0.587-0.651, p <0.001) were all increased. The results demonstrated that the consumption of weekly frequencies of marine fish and the weekly servings of light-coloured and dark-coloured vegetable intake were significantly and positively correlated with plasma α-carotene and β-carotene（all p for trend＜0.05）. The frequency of fried food intake was negatively correlated with α-carotene(p for trend=0.022). Furthermore, the backward selection multivariate linear regression analysis adjusting for age, sex, FFQ-estimated energy intake, diabetes duration and education years and various food items revealed that tertile weekly frequencies of marine fish (beta=2.372, p =0.010), weekly servings of light-coloured (beta =1.845, p = 0.018) as well as dark-coloured vegetable (beta=2.099, p = 0.009) intake remained independently correlated with α-carotene. Tertile weekly frequencies of marine fish (beta =7.902, p=0.041), weekly servings of light-coloured (beta =6.704, p=0.039) and dark-coloured vegetable (beta =6.923, p=0.041) intake also positively correlated with β-carotene, while the tertile frequencies of fried food intake were negatively correlated with α-carotene (beta = -1.937, p = 0.022). Based on the above findings, this FFQ could be used to evaluate the energy and macronutrient intakes of type 2 diabetes in Taiwan. In conclusion, Food frequency questionnaire composing forty-five food items employed in this study appears to effectively estimate energy and the three major nutrients and specific health food intake patterns in type 2 diabetes patients.

並列關鍵字

Type 2 diabetes mellitus ； Food frequency questionnaire ； 24-Hour dietary recall ； Validation ； Carotenoids

參考文獻

1.Yu, F.J., et al., Increased ferritin concentrations correlate with insulin resistance in female type 2 diabetic patients. Ann Nutr Metab, 2012. 61(1): p. 32-40.

2.Dietary patterns, dietary biomarkers, and kidney disease in patients with type 2 diabetes: a repeated-measure study in Taiwan. Published online: April 2017.

3.Jiang, Y.D., et al., Incidence and prevalence rates of diabetes mellitus in Taiwan: analysis of the 2000-2009 Nationwide Health Insurance database. J Formos Med Assoc, 2012. 111(11): p. 599-604.

4.Hu, F.B., R.M. van Dam, and S. Liu, Diet and risk of Type II diabetes: the role of types of fat and carbohydrate. Diabetologia, 2001. 44(7): p. 805-17.

5.Knowler, W.C., et al., Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med, 2002. 346(6): p. 393-403.