Rubin D H, Leventhal J M, Krasilnikoff P A, Kuo H S, Jekel J F, Weile B, Levee A, Kurzon M, Berget A
Division of Pediatric, Albert Einstein College of Medicine-Montefiore Medical Center, Bronx, New York.
Pediatrics. 1990 Apr;85(4):464-71.
Prior studies investigating the relationship between infant feeding and infectious illnesses in developed countries have provided conflicting data about whether breast-feeding protects against common infectious illnesses early in life. These conflicts may in part be due to the failure to consider the following methodologic issues: (1) collecting data prospectively at frequent intervals for active surveillance of the detection of infections and of feeding practices, (2) specifying what is meant by infectious illnesses and breast-feeding, (3) controlling for confounding variables such as social class or presence of siblings in the household, and (4) applying appropriate analytical strategies to a population in which both feeding and exposure to illness change over time. A total of 500 infants born consecutively in a university-affiliated community hospital in Copenhagen, Denmark, were studied prospectively for the first 12 months of life by means of a detailed, monthly, mailed questionnaire that focused on feeding practices and illnesses (overall response rate, 73%). The percentage of infants who were completely or mostly breast-fed decreased from 88% at 1 month to 20% at 12 months of age. After adjustment for major covariates, no statistically significant relationship was found between the type of infant feeding and the incidence of four categories of infectious illnesses: gastroenteritis, upper respiratory illness, otitis media, and lower respiratory illness. The adjusted incidence density ratio for gastroenteritis was 1.067 (95% confidence interval = 0.982, 1.226) and for upper respiratory illnesses 0.984 (95% confidence interval = 0.883, 1.096).(ABSTRACT TRUNCATED AT 250 WORDS)
先前在发达国家进行的关于婴儿喂养与传染病之间关系的研究，对于母乳喂养是否能预防生命早期常见传染病提供了相互矛盾的数据。这些矛盾部分可能是由于未能考虑以下方法学问题：（1）通过频繁前瞻性收集数据，对感染检测和喂养方式进行主动监测；（2）明确传染病和母乳喂养的定义；（3）控制混杂变量，如社会阶层或家庭中兄弟姐妹的存在；（4）对喂养和患病暴露随时间变化的人群应用适当的分析策略。在丹麦哥本哈根一家大学附属医院连续出生的500名婴儿，在其生命的前12个月通过详细的月度邮寄问卷进行前瞻性研究，问卷重点关注喂养方式和疾病（总体回复率为73%）。完全或主要母乳喂养的婴儿比例从1个月时的88%降至12个月时的20%。在对主要协变量进行调整后，未发现婴儿喂养类型与四类传染病（肠胃炎、上呼吸道疾病、中耳炎和下呼吸道疾病）的发病率之间存在统计学上的显著关系。肠胃炎的调整发病密度比为1.067（95%置信区间 = 0.982，1.226），上呼吸道疾病为0.984（95%置信区间 = 0.883，1.096）。（摘要截断于250字）
