以飲食頻率問卷評估婦女飲食型態、昇糖指數及昇糖負荷對懷孕結果之研究

Estimating maternal dietary patterns, dietary glycemic index and glycemic load by food frequency questionnaire for pregnancy outcomes

黃鉅凱(Huang Chu Kai)

指導教授 ： 盧立卿

國立臺灣師範大學/教育學院/人類發展與家庭學系/碩士(2007年)

摘要

本研究為國內首次探討婦女飲食昇糖指數(glycemic index, GI)及昇糖負荷(glycemic load, GL)對懷孕結果(新生兒出生體型及產後體重保留)之影響。研究以飲食頻率問卷收集婦女懷孕前一年、懷孕全期及產後六個月之飲食相關資料，並比較飲食頻率問卷與24小時飲食回憶之相對效度。
研究對象為於民國91年10月至11月於台北市立婦幼醫院所招募婦女151人 (民91世代)，及民國94年2月至95年7月於台北市立婦幼醫院及台灣大學醫學院附設醫院所招募婦女150人(民93世代)。排除基本資料不全的婦女，最後本研究共分析283人，平均年齡30.9±4.0歲，平均身高159.8±5.0公分，未懷孕時平均體重53.6±8.0公斤，平均BMI 21.0±2.8公斤/公尺2，平均孕期體重增加量14.6±4.2公斤。新生兒平均出生週數38.9±1.2週, 平均出生體重3214.8±486.3公克，平均出生身長49.2±2.1公分，平均出生頭圍34.2±2.0公分，平均出生胸圍32.7±2.2公分。婦女產後六個月體重保留平均2.4±3.5公斤。
研究結果發現，以飲食頻率問卷評估婦女三個不同時期飲食GI分別為47.9、47.1及47.5，GL為231.5、218.6及158.6，GL密度為61.9、59.5及58.2，皆近似常態分佈。飲食頻率問卷可呈現個人長期飲食狀況，但易高估熱量與營養素攝取，故常以營養素密度呈現數據。比較婦女懷孕期與未懷孕時期(孕前一年及產後六個月)之營養素密度結果發現，懷孕期間顯著攝取較高鉀、鈣、鎂、磷、膳食纖維，攝取較低維生素K (p<0.01)。在影響新生兒出生體型結果發現，懷孕週數、婦女未懷孕體重、未懷孕BMI值、孕期總體重增加量與新生兒體型呈顯著正相關，新生兒性別亦影響出生體型。新生兒體重與子女人數呈正相關，而與教育程度、家庭社經地位呈負相關。新生兒身長與配偶身高呈正相關。分析影響婦女產後體重保留結果發現，婦女產後六個月的體重保留量，與孕期體重增加量及坐月子體重保留量呈顯著正相關，而與已有的子女人數呈負相關。飲食GI與產後六個月體重及BMI值呈正相關，GL密度則與產後六個月體重保留呈正相關。FFQ與24小時飲食回憶相對效度結果發現，營養素方面，有57.6%的營養素Kappa值達顯著相關，87.9%的營養素相關係數呈顯著相關，Kappa值由-0.008(維生素B6)至0.182(鈣)；營養素密度方面，共有64.5 %的營養素密度Kappa值達顯著相關，77.4 %的營養素密度相關係數呈顯著相關，Kappa值由0.006(維生素A)至0.250(鈣)。營養素與營養素密度之平均相關係數均為0.17，以24小時飲食回憶為指標，相對效度最佳的營養素皆為鈣質。
多元迴歸分析結果顯示，婦女飲食GI、GL及GL密度對新生兒出生體型的影響力，在懷孕期間影響力大於懷孕前一年。婦女懷孕期間之飲食GL及GL密度，與新生兒出生體型呈正相關，對新生兒體重、身長及頭圍的解釋力約為1 %，對胸圍的解釋力約為1.5 %；在婦女產後體重保留方面，此時期婦女飲食GL及GL密度，對婦女產後六個月體重保留量的解釋力約達2 %。
綜合上述，本研究以飲食頻率問卷評估婦女飲食昇糖指數、昇糖負荷與昇糖負荷密度，對懷孕結果之研究，發現以飲食頻率問卷評估婦女昇糖負荷及昇糖負荷密度，的確會影響新生兒出生體型及婦女產後體重保留，而昇糖指數則較不具有相關性存在。

關鍵字

飲食頻率問卷 ； 昇糖指數 ； 昇糖負荷 ； 新生兒體型 ； 產後體重保留 ； 相對效度

並列摘要

The purpose of this study was to evaluate the effect of maternal dietary glycemic index (GI) and glycemic load (GL) on pregnancy outcomes, including infant birth outcomes and postpartum weight retention. We collected dietary information by food frequency questionnaire (FFQ), and compared the relative validity with 24-hour recalls. Women were recruited from Taipei Municipal Women’s and Children’s Hospital (n=151), and National Taiwan University Hospital (n=150). After excluding the missing data, there were total 283 women in our final analyses. The average age was 30.9 years old, height was 159.8 cm, weight was 53.6 kg, BMI was 21.0 kg/m2, and the gestational weight gain was 14.6 kg. The average gestational age for newborns was 38.9 weeks. The average birth weight, length, head circumference and chest circumference of the newborns were 3214.8 g, 49.2 cm, 34.2 cm, and 32.7 cm, respectively. For pre-pregnancy, pregnancy and 6 month after postpartum, the average dietary GI was 47.9, 47.1 and 47.5, the average dietary GL was 231.5, 218.6, and 158.6, and the average dietary GL density was 61.9, 59.5 and 58.2. All three values appeared to have normal distributions. FFQ can estimate personal long term dietary status, but it often over-estimate total energy and nutrients. Thus, we present data by nutrients density per 1000 kcal. We found that the nutrients density of potassium, calcium, magnesium, phosphorous and dietary fiber during pregnancy were higher than pre-pregnancy and 6 month after postpartum, while the vitamin K were lower during pregnancy. We also found that maternal gestational age, male gender, body weight and BMI before pregnancy, and weight gain during pregnancy were positively correlated with birth outcomes. Birth weight was positively correlated with number of children, but negatively correlated with women’s education and family socioeconomic status. Birth length was positively correlated with father’s height. We found women postpartum 6 month weight retention was positively correlated with weight gain during pregnancy and postpartum 1 month weight retention, but negatively correlated with number of children. Dietary GI was positively correlated with postpartum 6 month body weight and BMI, and GL density was positively correlated with postpartum 6 month weight retention. Compared to the 24 hour recall, the relatively validity of FFQ by correlation coefficients and kappa statistics, we found the nutrients kappa statistics were -0.008 (vitamin B6) to 0.182 (calcium), and nutrients density kappa statistics were -0.006 (vitamin A) to 0.250 (calcium). Average correlation coefficients in nutrients and nutrients density were 0.17, and nutrient and nutrient density show that calcium had the high relative validity between the two methods. By multiple regression analysis, maternal dietary GI, GL and GL density during pregnancy had stronger correlations with newborn outcomes than during pre-pregnancy. Dietary GL and GL density during pregnancy explain 1 % of variances for birth weight, length and head circumference, and 1.5 % of variances for chest circumference. Dietary GL and GL density also explain 2 % of variances for 6 month postpartum weight retention. In conclusion, we found maternal dietary GL and GL density will affect infant birth outcomes and 6 month postpartum weight retention, but dietary GI appears to have weaker correlations than dietary GL and GL density.

參考文獻

行政院衛生署(2002)。國人膳食營養素參考攝取量。

李建南(2002)。妊娠與糖尿病。台灣醫學，6(5)，785-789。

張永生、黃禮偉、江千代、林妙卿、邵文逸(2006)。孕產婦體重管理之效果。北市醫學雜誌，3(4)，373-384。

陳培君、李中一、林瑞雄、宋鴻樟(2002)。婦女抽菸及二手煙暴露和低出生體重嬰兒。台灣衛誌，21(5)，321-328。

陳瓊華、徐南麗、楊燦、鍾蝶起(2007)。不同種族婦女懷孕期間物質使用和身體受虐的情況與其嬰兒出生體重之關係。台灣衛誌，26(1)，58-65。