Impact of Baby-Friendly Practices on Breastfeeding: A Multilevel Analysis

母嬰親善步驟與母乳哺育意願多層次邏輯斯迴歸分析

楊美雪(Mei-Hsueh Yang) ； 康峻銘(Chun-Min Kang) ； 郭宗正(Tsung-Cheng Kuo) ； 林為森(Wei-Sen Lin)

《南臺灣醫學雜誌》 6卷1期 (2010/06) Pp. 1-13

https://doi.org/10.6726/MJST.201006_6(1).0001

摘要

目的：以多層次分析探討哺餵母乳意願有關的重要母嬰親善步驟。方法：橫斷面自填式問卷調查中南部母嬰親善醫療院所之產婦母嬰親善感受與產後母乳哺餵意願之關聯，有效問卷648份來自26家母嬰親善機構。採多層次邏輯斯廻歸分析探究母嬰親善感受與醫院彙整的脈絡效果對於母乳哺育率之影響。結果：當醫院間哺餵母乳率有顯著的差異存在，兩醫院脈絡效果，哺餵時間依需求（步驟8）（OR=32.63，CI=2.71-52.84）與不給母奶之外的食物（步驟6）（OR=0.12，CI=0.02-0.57），在母乳哺育率變異的解釋上是重要的。在控制醫院變項、產婦年齡以及社會經濟等因素後，產婦層次的持續哺餵支持（步驟10）（OR=1.40，CI=1.04-1.88）、哺餵時間依需求（步驟8）（OR=1.44，CI=1.06-1.96）以及不給母奶之外的食物（步驟6）（OR=1.67，CI=1.25-2.25)等3個步驟對母乳哺育率有顯著的正向關聯。合併產婦層次與醫院層次變項的完整模式解釋了醫院間母乳哺育率變異的40.09%。結論：本研究發現不同醫院間母乳哺育率有顯著變異，哺餵時間依需求（步驟8）在產婦層次與醫院層次皆對母乳哺餵率有顯著的正向關聯，建議醫院應多關注產婦哺餵時間依需求等執行的感受。

關鍵字

多層次分析 ； 母嬰親善 ； 母嬰親善感受 ； 哺餵母乳 ； 哺餵時間依需求

並列摘要

Objective: The aim of this study was to adopt a multi-level modeling approach in order to identify the major steps in the Baby Friendly practices that lead to initiation of breastfeeding.Methods: A cross-sectional study using self-administered questionnaires was undertaken in Baby-Friendly institutions throughout central and southern Taiwan. The questionnaires focused on the association between the Baby-Friendly practices experienced by mothers and their impact on the initiation of breastfeeding during the immediate postpartum period. We attempted to investigate a possible contextual (aggregate level) effect of institutions on breastfeeding and disentangle contextual from individual influences. Multilevel logistic regression modeling was used to analyze data on 648 mothers who had given birth to healthy infants in 26 institutions. Statistical analysis including descriptive statistics and factor analysis was performed with SPSS for Windows 14.00. Multi-level regression models were per- formed with HLM 6.06 statistical software.Results: There was significant variation in the rates of breastfeeding initiation between institutions. Multi-level models showed that two contextual effects of the Baby-Friendly practices, breast- feeding on demand (step 8) (OR=32.63, CI=2.71-52.84) and breast milk only (step 6) (OR=0.12, CI = 0.02-0.57) was important in explaining the variance in the rates of breastfeeding initiation. After controlling for variables in institutions as well as the mothers' age and socio-economic status, three of the Baby-Friendly practices at individual-level such as continuous breast feeding support (step 10) (OR=1.40, CI =1.04-1.88), breastfeeding on demand (step 8) (OR=1.44, CI=1.06-1.96), and breast milk only (step 6) (OR=1.67, CI=1.25-2.25) was positively associated with the likelihood of initiation of breastfeeding. The full model combining variables at both individual and aggregate levels accounted for 40.09% of the variation in the rates of breastfeeding initiation between institutions.Conclusions: There is the considerable variation between institutions in the rates of breastfeeding initiation. The multilevel findings from the perspectives of post- partum mothers and institutional contextual effects highlights that breastfeeding on demand (step 8) has an important effect on the mother's willingness to breastfeed.

並列關鍵字

multilevel analysis ； baby friendly ； baby friendly experiences ； breastfeeding ； breastfeeding on demand