Development And Evaluation Of Parents' Feeding Practice Scale For Infant And Young Child

ObjectiveTo meet the need of a standard scale with high validity and reliability for assessing the feeding practice of infant and young child in China,we developed a Parent's Feeding Practice Scale for Infant and Young Child at age of 6-24 months based on the Chinese situation.The reliability,validity and measurement invariance across the groups of the scale were evaluated.The scale,as a practicable tool,could be used to early identify the unhealthy feeding practice of infant and young child's parents in China.Methods1.Formation of primary questionnaire.After wide literature review on parents'feeding practice,concepts and influencing factors of parent feeding practice were analyzed.Considering the characters of infant and young child's eating behavior,the domains and item pool of parents'feeding practice were predefined.These items were primarily screened by expert recommendations and two rounds of Delphi expert enquiry.Then the primary questionnaire was established.2.Pre-test of primary questionnaire.Caregivers of children aged 6 to 24 months,were recruited randomly from four community health service centers in Beijing and they were asked to answer questions in the primary questionnaire.Correlation Coefficient analysis,Differentiation analysis and Cronbach's alpha coefficient were used for screening of the items.Parallel analysis and exploratory factor analysis were used to extract the factor structure.3.Finalization and evaluation of the questionnaire.Totally 928 caregivers for children aged 6 to 24 months were recruited as a convenience sample from 5community health service centers in Harbin,3 community health service centers in Xiamen and 3 community health service centers in Xi'an respectively.Correlation Coefficient analysis,Differentiation analysis and Cronbach's alpha coefficient were used for screening of the items.Factor analysis was used to confirm the factor structure of the questionnaire.Each factor was defined and the final scale was constructed.Cronbach's alpha coefficient and split-half reliability was calculated to evaluate the reliability of the scale.Delphi method was used to evaluate content validity of the scale,and construct validity was across validated by exploratory factor analysis combining confirmatory factor analysis.4.Measurement invariance,discriminating ability and correlation analysis.Item parceling technique and multi-group confirmatory factor analysis(MGCFA)was used to test measurement invariance on sex,city,age and residency registry.5.Situation and the influencing factors of parents'feeding practice for infant and young child.Means and standard deviation were used to describe the current situation of parents'feeding practice for infant and young child.Nonparametric test analysis was used to compare the score of each dimension on sex,age and of caregivers,caregivers'residency,education,work,type of family income,and area to explore the possible social demographic factors that influencing the parents'feeding practice.Result1.Preliminary questionnaire of parents'feeding practice for infant and young child was derived from both literature review and Delphi,which consisted of 8 factors and 42 items.2.After survey of 378 parents for children aged 6 to 24 months and items screening,31 items were selected for the formal test questionnaire.3.Final questionnaire consisted of six dimensions and 27 items after survey of928 parents for children aged 6 to 24 months who were recruited from 11 community health service centers in 3 cities.The six dimensions included:pressuring to eat,concern about infant under-eating,emotional and instrumental feeding,encouragement and healthy eating,parent-led eating,concern about infant overeating.And each dimension included three to six items.The results of reliability analysis showed that Cronbach's alpha coefficient was 0.783 with range of 0.737 to 0.843across six domains.Guttman split-half reliability was 0.750 with range of 0.637 to0.831 for six domains.The results of validity analysis showed that the content validity index was 0.910.The fit indexes of CFA were?~2/df=2.000,CFI=0.915,NNFI=0.904,RMSEA=0.046,which indicated good structural validity.4.MGCFA showed that fit measures satisfied strict measurement invariance,and most models satisfied structure invariance.5.General scores for parents in the three cities:pressuring to eat score 2.59±0.81;concern about infant under-eating score 2.59±0.81;emotional and instrumental feeding score 2.80±0.72;encouragement and healthy eating score 3.73±0.65;parent-led eating score 3.58±0.83;concern about infant overeating score 2.52±0.83.Nonparametric test analysis showed that the dimension scores between different gender had no statistical significance(P>0.05).Parents whose children were 12-24months old had higher scores on pressuring to eat,concern about infant under-eating,emotional and instrumental feeding than those whose children were 6-12 months old(P<0.05).The migrant population had higher score on concern about infant under-eating,emotional and instrumental feeding than permanent population,whereas the permanent population had higher scores on encouragement and healthy eating,parent-led eating,concern about infant overeating than the migrant population;mothers had higher scores on parent-led eating than grandparents;Parents who had college degrees had higher scores on encouragement and healthy eating than those with high school education.Parents in Xiamen had higher scores on concern about infant under-eating than Xi'an.There is no difference between other groups in each dimension.ConclusionsIt was reasonable to evaluate feeding practice of infant and young child parent through the six domains:pressuring to eat,concern about infant under-eating,emotion-al and instrumental feeding,encouragement and healthy eating,parent-led eating,concern about infant overeating.The scale had good reliability and validity.The fitting indexes for the measurement invariance models on sex,city,age and residency satisfied strict measurement invariance.The scale could be used as a practicable tool to evaluate eating practice of infant and young child's parents in China.