哺餵策略與兒童異位性疾病發生趨勢之相關性研究—臺灣出生世代研究

背景：異位性疾病包括異位性皮膚炎、氣喘和過敏性鼻炎，兒童期常依特定順序出現，呈現「過敏進行曲（atopic march）」現象，對兒童健康造成長期影響。早期哺餵策略被視為影響免疫系統發育的重要因素，然而國內多為橫斷性研究，缺乏以出生世代為基礎的長期追蹤。目的：運用台灣出生世代研究資料，探討異位性皮膚炎、氣喘與過敏性鼻炎的盛行趨勢及其與哺餵策略、父母及小孩背景與環境因子的關聯，並評估早期發病對後續疾病發生風險的預測。方法：本研究採世代研究設計，資料來源為臺灣2005年出生世代研究（Taiwan Birth Cohort Study, TBCS），以18,443名2005年出生兒童為樣本，分析至2011年間共四波(出生6個月、18個月、3歲及5.5歲)調查資料。進行描述性統計分析各項背景變項與異位性疾病之分布狀況，並運用邏輯斯迴歸與廣義估計方程式模型，控制性別、社經地位、父母過敏史、出生條件與居住環境等干擾因子，探討哺餵策略（包括母乳哺育、副食品與營養補充品）與異位性疾病發生趨勢之相關性。結果：本研究共納入18,443名臺灣出生世代兒童。四波追蹤顯示，異位性皮膚炎在嬰兒期盛行率較高，隨年齡增長而下降；氣喘與過敏性鼻炎則呈逐年上升趨勢。三歲前罹患異位性皮膚炎者，五歲時有2.26倍罹患氣喘的風險（p<.001）及2.11倍罹患過敏性鼻炎的風險（p<.001）；而三歲前同時患有異位性皮膚炎與氣喘者，則五歲時有2.85倍罹患氣喘的風險（p=.014）及2.71倍罹患過敏性鼻炎的風險（p=.002）；男嬰兒發生此三種疾病的風險分別為女嬰的1.25倍(p<0.001）、1.51倍(p<.001）與1.36倍（p<0.001）。哺餵策略分析結果顯示，哺乳時間達 6 個月以上的嬰兒，其罹患異位性皮膚炎的風險為未哺乳者的 1.27 倍（p < .001）。此外，相較於哺乳少於4個月的嬰兒，哺乳4–6個月者在3歲以前罹患異位性皮膚炎的風險亦顯著增加28%–51%（AOR=1.28–1.51）；氣喘則在1.5歲時風險顯著增加66%（AOR=1.66），過敏性鼻炎則無顯著差異，與傳統保護假說相左，可能與高敏家庭選擇性哺乳及逆因果關係有關。副食品分析顯示，早期添加糊化米類與麥類產品可分別降低16%（AOR=0.84）與29%（AOR=0.71）的異位性皮膚炎風險；惟糊化米類添加會增加55%氣喘風險(AOR=1.55）。早期添加稀飯可降低氣喘14%（AOR=0.86）及過敏性鼻炎9%（AOR=0.91）的風險。果汁及蔬菜汁攝取則分別增加過敏性鼻炎15%（AOR=1.15）及18%（AOR=1.18）風險；水果泥與蔬菜泥則無顯著風險存在。營養補充品分析顯示，鈣粉補充對異位性皮膚炎具短期保護作用，較未補充者可減少18%風險（AOR=0.82）；而益生菌、綜合維生素及魚肝油的補充在調整干擾因子後則無顯著效果存在。多變項分析顯示，父母過敏史與部分教育程度為主要風險因子。父母皆具過敏病史者，其子女罹患異位性皮膚炎、氣喘與過敏性鼻炎的風險分別增加 1.86 倍（p<.001）、1.84 倍（p<.001）與 2.69 倍（p<.001）。父親教育程度為大專者，其子女過敏性鼻炎風險為教育程度較低者的 1.08 倍（p=.022），氣喘風險為 1.14 倍但未達顯著（p=.057）；母親教育程度為研究所者，其子女過敏性鼻炎風險為教育程度較低者的 1.20 倍（p=.005），氣喘為 1.22 倍但未達顯著（p=.082）。家庭收入中高（10–15 萬）者，過敏性鼻炎風險為 5–7 萬者的 1.15 倍（p=.011）。家中子女人數達 4 人以上時氣喘與過敏性鼻炎風險則分別降至 0.44 倍（p<.001）與 0.44 倍（p<.001）。居住於農村/鄉村地區者其異位性皮膚炎較居住於都市者風險增加至 1.20 倍（p=.004），但過敏性鼻炎風險降低至 0.83 倍（p<.001），氣喘風險則為 0.73 倍（p<.001）。居住環境潮濕發霉則與三種疾病風險均呈正向關聯，發生異位性皮膚炎風險為 1.34 倍（p<.001）、氣喘 1.12 倍（p=.036）、過敏性鼻炎 1.13 倍（p<.001）。綜合結果顯示，多子女家庭、農村居住、年輕或外籍母親及早期多元微生物暴露對呼吸道過敏疾病具保護作用，符合衛生假說；但對皮膚過敏疾病的保護效果有限。結論：本研究運用台灣出生世代長期追蹤資料，揭示異位性疾病在不同年齡與性別間具有差異化的盛行趨勢，結果顯示，2005 年出生之臺灣兒童在嬰幼兒期至學齡前呈現明顯的年齡進展與疾病轉換模式，早期異位性皮膚炎顯著增加後續氣喘與過敏性鼻炎的風險，支持「過敏進行曲」的連續性假設，凸顯早期預防與介入的重要性。哺餵策略、父母教育程度、家庭結構與環境因素對異位性疾病風險具有多層次且方向不一的影響。家庭與環境因素對呼吸道過敏疾病的保護作用符合衛生假說，但對皮膚過敏疾病效果有限。建議未來制定兒童過敏預防策略時，應綜合考量早期發病史、哺育方式、家庭背景與居住環境，並進一步釐清母乳哺餵及營養補充品的作用機制與適用條件。

Background: Atopic diseases, including atopic dermatitis (AD), asthma, and allergic rhinitis (AR), often occur sequentially in childhood, presenting the “atopic march” phenomenon and causing long-term impacts on children’s health. Early feeding strategies are considered critical factors influencing immune system development; however, most studies in Taiwan have been cross-sectional, lacking long-term, birth cohort–based evidence.Objectives:This study utilized data from the Taiwan Birth Cohort Study (TBCS) to examine the prevalence trends of AD, asthma, and AR, and their associations with feeding strategies, parental and child background characteristics, and environmental factors. It further assessed the predictive value of early-onset disease for subsequent allergic outcomes.Methods:This cohort study used data from the TBCS, including 18,443 children born in 2005, with four survey waves conducted at ages 6 months, 18 months, 3 years, and 5.5 years (2005–2011). Descriptive statistics were used to analyze the distribution of background variables and atopic diseases. Logistic regression and generalized estimating equation (GEE) models were applied, adjusting for confounders such as sex, socioeconomic status, parental allergic history, birth conditions, and residential environment, to evaluate the associations between feeding strategies (breastfeeding, complementary foods, and nutritional supplements) and atopic disease trends.Results:A total of 18,443 Taiwanese children were included. Four-wave follow-up data showed that AD prevalence was highest in infancy and declined with age, while asthma and AR prevalence increased steadily over time. Children with AD before age 3 had a 2.26-fold higher risk of developing asthma (p < .001) and a 2.11-fold higher risk of AR (p < .001) at age 5.5. Those with both AD and asthma before age 3 had a 2.85-fold higher risk of asthma (p = .014) and a 2.71-fold higher risk of AR (p = .002) at age 5.5. Male infants had significantly higher risks for all three diseases than females—1.25-fold for AD (p < .001), 1.51-fold for asthma (p < .001), and 1.36-fold for AR (p < .001). Feeding strategy analysis revealed that breastfeeding for ≥6 months increased the risk of AD by 1.27-fold compared with no breastfeeding (p < .001). Compared with breastfeeding for <4 months, breastfeeding for 4–6 months increased the risk of AD before age 3 by 28%–51% (AOR = 1.28–1.51) and increased asthma risk by 66% at 1.5 years (AOR = 1.66), with no significant difference for AR. These findings contradict the traditional protective hypothesis and may be related to selective breastfeeding among high-risk families and reverse causation. For complementary foods, early introduction of gelatinized rice and wheat products reduced AD risk by 16% (AOR = 0.84) and 29% (AOR = 0.71), respectively; however, gelatinized rice increased asthma risk by 55% (AOR = 1.55). Early porridge intake reduced asthma risk by 14% (AOR = 0.86) and AR risk by 9% (AOR = 0.91). Juice and vegetable juice increased AR risk by 15% (AOR = 1.15) and 18% (AOR = 1.18), respectively, while fruit purée and vegetable purée showed no significant associations. Regarding nutritional supplements, calcium powder was associated with an 18% reduced risk of AD (AOR = 0.82), while probiotics, multivitamins, and cod liver oil showed no significant effects after adjusting for confounders. Multivariate analysis identified parental allergic history and certain educational levels as major risk factors. Children with both parents having allergic history had significantly increased risks of AD (OR = 1.86, p < .001), asthma (OR = 1.84, p < .001), and AR (OR = 2.69, p < .001). Paternal education at college level was associated with an increased risk of AR (OR = 1.08, p = .022) and a non-significant increase in asthma (OR = 1.14, p = .057). Maternal education at graduate level was associated with a higher AR risk (OR = 1.20, p = .005) and a non-significant increase in asthma risk (OR = 1.22, p = .082). Children from middle-high income households (annual income NT$ 1.0–1.5 million) had higher AR risk compared to those from middle-income households (OR = 1.15, p = .011). Having ≥4 children in the household was associated with lower asthma (OR = 0.44, p < .001) and AR risk (OR = 0.44, p < .001). Living in rural/township areas increased AD risk (OR = 1.20, p = .004) but reduced AR (OR = 0.83, p < .001) and asthma risk (OR = 0.73, p < .001). Damp and moldy living environments were positively associated with all three diseases—AD (OR = 1.34, p < .001), asthma (OR = 1.12, p = .036), and AR (OR = 1.13, p < .001). Overall, larger family size, rural residence, younger or foreign-born mothers, and early exposure to diverse microorganisms appeared protective against respiratory allergic diseases, supporting the hygiene hypothesis, but these factors had limited protective effects for skin allergies.Conclusions:This long-term Taiwanese birth cohort study reveals distinct age- and sex-specific prevalence patterns of atopic diseases, showing clear progression and disease transition from infancy to preschool years. Early-onset AD significantly increased the risk of subsequent asthma and AR, supporting the continuity of the atopic march and underscoring the importance of early prevention and intervention. Feeding strategies, parental education levels, family structure, and environmental factors exerted complex and sometimes opposing effects on allergic disease risks. Family and environmental protective factors for respiratory allergies were consistent with the hygiene hypothesis but had limited benefit for AD. Future prevention strategies should consider early disease history, feeding practices, family background, and living environment, and further clarify the mechanisms and applicability of breastfeeding and nutritional supplements in allergy prevention.