目標：本研究旨在探討三歲幼兒居家環境安全與事故傷害的相關性。方法：分析2008年嬰幼兒健康照護需求調查（台灣出生世代研究第三波調查）資料檔，先檢視幼兒過去一年各種事故傷害之盛行率，及事故傷害與幼兒及其照顧者特性、幼兒居家安全簡要檢核表各個項目及居家安全檢核分數（分為低、中、高風險組）的相關性。再使用多變項邏輯迴歸模式，控制幼兒及其照顧者特性，以檢視居家安全檢核分數與事故傷害之相關性。結果：在完訪的19,910位三歲幼兒家長當中，約13.08%（2,653人）自述在過去一年其幼兒曾因事故傷害而就醫；前五項事故傷害盛行率高低排序為：跌傷(8.03%)、物體或人撞擊(1.33%)、尖利物刺割傷(1.06%)、燒燙傷(1.00%)及器物夾壓(0.77%)。19,251人進入後續分析，雙變項分析發現：事故傷害盛行率男童顯著高於女童；白天為幼托園或保母且晚上為父母親照顧者，高於全天皆由父母親照顧者，以及白天為（外）祖父母，晚上為父母親照顧者。事故傷害盛行率隨居家安全檢核分數之增加而升高。以多變項邏輯迴歸控制幼兒性別及現住地；母親年齡、教育程度、過去一年工作情形及競爭照護需求；幼兒主要照顧者，發現：幼兒居家安全檢核分數高風險組之幼兒事故傷害及跌傷風險均為低風險組之1.5倍。結論：三歲幼兒過去一年事故傷害盛行率為13.08%，幼兒居家安全檢核分數與事故傷害風險均顯著相關，其因果關係值得後續研究。

Objectives: This study examined the association between household safety and unintentional injury among three-year-old children in Taiwan. Methods: Data from the third wave TBCS (Taiwan Birth Cohort Study) were analyzed to examine the prevalence of unintentional injuries (e.g. falls, being hit by objects or persons, being stabbed or cut by sharp instruments), and the association between injury and each item on a household safety checklist and its summary household safety score (categorized as low, moderate, and high risk groups). The association was further investigated with multiple logistic regression after controlling for the characteristics of children and their caregivers. Results: Of the 19,910 three-year-old children studied, 13.08% received medical care due to an unintentional injury. The top five categories of injuries were falls (8.03%), being hit by objects or persons (1.33%), being stabbed or cut by sharp instruments (1.06%), burns (1.00%) and being clamped by an instrument (0.77%). Data about 19,251 of these children were analyzed further. On bivariate analysis, the prevalence of unintentional injury was higher for boys than for girls, and higher for those at kindergarten in the daytime and cared for by a parent at night than for those cared for by a parent for the whole day and those cared for by a grandparent in the daytime and by a parent at night. The prevalence of unintentional injury increased as the household safety score did. The high risk group had 1.5 times the risk of unintentional injury as did the low risk group in the multiple logistic regression model when controlled for children's gender, current address, mothers' age, education, and job in the previous year, competitive demand of caring, and type of principal caregiver. Conclusions: Over thirteen percent of three-year-old children received medical care due to unintentional injury in the past year. The household safety score was significantly associated with the risk of unintentional injury. A causal relationship is worth further study.

幼兒；居家安全；事故傷害；跌傷；台灣出生世代研究

children; household safety; injury; fall injury; TBCS Taiwan Birth Cohort Study