Abstract: Objective To introduce Nursing Child Interaction (PCI) Feeding Scale into China, and to assess the reliability and validity of PCI Feeding Scale in Chinese parent-child dyads. Methods Under the recommended steps by the International Testing Committee, PCI Feeding Scale was translated, then Chinese Version of Parent-Child Interaction Feeding Scale (PCI-FS-C) was developed. From June 2018 to January 2019, 310 parent-child dyads from four hospitals in Beijing, Yinchuan, Lanzhou and Maanshan were selected by convenience sampling, and PCI-FS-C was adopted to code their feeding interaction. Reliability was evaluated by internal consistency (Cronbach's α coefficient) and test-retest reliability (inter-item consistency) conducted with 10% of the samples randomly selected. Six samples were selected to evaluate rater reliability. Chinese Version of Infant-Toddler Home Observation Measurement of the Environment (IT-HOME-C) was used to exam the concurrent validity by Spearman correlation in 41 samples. Multivariate linear regression was used to evaluate discriminant validity. Results Totally 273valid mother-infant dyads were included in this study. The Cronbach's α coefficients of total score, caregiver and infant subscale scores were 0.846, 0.805 and 0.690, respectively. Both caregiver subscale (r=0.778) and PCI-FS-C (r=0.847) had excellent test-retest reliability, while the infant subcale showed acceptable test-retest reliability (r=0.480) (P< 0.01). For rater reliability, the Kendall coefficient of PCI-FS-C was 0.860. PCI-FS-C, caregiver subscale and infant subscale were moderately associated with IT-HOME-C (r=0.526, 0.506, 0.427, P＜0.05). After controlling for mixed factors, there were significant differences among different age groups in PCI-FS-C, caregiver subscale and infant subscale (b=3.24, 2.31, 0.92). Conclusions PCI-FS-C is a valid assessment for Chinese mother-infants, which fills the gap for assessment using an Objective measurement of caregiver/parent-child interaction in China. Enlarged sample size is suggested for culture-adapted study in future.

Key words: parent-child interaction scale, infant feeding, reliability, validity

摘要： 目的 引入并修订美国华盛顿大学的亲子互动-喂养量表,并在中国人群中进行信效度研究。方法 按照国际测试委员会指南,对原亲子互动-喂养量表进行翻译,形成中文版亲子互动-喂养量表(PCI-FS-C)。2018年6月—2019年1月方便抽样北京、银川、兰州和马鞍山市4所妇幼保健院,招募310例0~12个月母子对,采集婴儿一次完整喂养过程的视频,由经过美国华盛顿大学培训合格获得原版量表评估资质人员进行视频编码评分。抽样6例评估评分者信度;于14~21 d后随机抽样10%评估重测信度;随机抽样41例进行中文版婴儿-学步儿家庭环境观察评估表(IT-HOME-C)效标效度检验。使用多元线性回归进行区分效度分析。结果 共获得273例有效样本。总量表、照护者和婴儿量表内部一致性Cronbach's α 系数分别为0.846、0.805和0.690;总量表、照护者和婴儿量表的重测信度分别为0.847、0.778和0.480(P<0.01);总量表评分者信度Kendall相关性系数为0.860;PCI-FS-C的总分、照护者和婴儿量表总分与IT-HOME-C总分呈显著性中度正相关(r=0.526、0.506、0.427,P＜0.05)。控制混杂因素后,婴儿年龄与PCI-FS-C量表总分、照护者量表和婴儿量表总分显著相关(b=3.24、2.31、0.92)。结论 中文版亲子互动-喂养量表在中国人群中具有良好的信度和效度,可以作为我国婴儿喂养亲子互动的评估工具。今后可以考虑开展大样本量分层抽样的本土化常模研究。

关键词: 亲子互动评估, 婴幼儿喂养, 信度, 效度