Evidências de validade do questionário para avaliação do conhecimento dos pais sobre alimentação complementar do lactente (QPAC) à luz da teoria da resposta ao item

Abstract

Complementary feeding is characterized by the offer of food or liquids in addition to breast milk from the sixth month of life. Parents are primarily responsible for establishing eating habits. The objective was to analyze the evidence of validity of the Questionnaire for Assessment of Parental Knowledge on Complementary Infant Feeding (QPAC), according to the Item Response Theory (IRT). Methodological study developed between March 2020 and March 2021 with 870 parents/caregivers of children between 6 months and 2 years of age. The semantic analysis took place with 50 parents/caregivers. Then, this version was evaluated by three experts, originating the version of the QPAC that was applied to 820 parents/caregivers by telemonitoring, distributed among 14 Primary Health Care Units (UAPS) in the city of Fortaleza-CE, selected by drawing lots. Descriptive analysis, internal reliability (Cronbach's Alpha and McDonald's Omega), exploratory factor analysis - AFE, calibration/estimation of item parameters (a,b), Item Characteristics Curves and construction of a scale for score distribution were performed. of the questionnaire, with the aid of the statistical program R. The study was submitted to the Ethics and Research Committee of the Federal University of Ceará, respecting the precepts of Resolution 466/12, receiving approval under opinion nº 3,836,768. In the semantic analysis, items 1 and 2, 18 and 20 were condensed, originating each pair 1 single item, totaling 31 items that became interrogative. Respondents were mostly female (90.5%) and 48% said they had never received information about complementary feeding. The AFE showed a statistically significant correlation between the variables and the polychoric correlation matrix pointed to a dominant dimension, explaining 32.3% of the variance in the responses of the items, resulting in a questionnaire with 27 items, which were submitted to the calibration process for the estimation of the IRT parameters, ending up with 19 items that presented good power of discrimination and standard error within the expected. Cronbach's alpha improved to 0.82 and MacDonald's Omega to 0.87. Items 13, 19, 22, 26 and 27 were the ones that best described parents' knowledge about complementary feeding. Items 1, 4, 15, 29 and 30 had lower power of discrimination. As for the difficulty, the items were considered medium level, as they were below the average of the scale, showing that they were easy to answer. To score the items, a mean of 100 and a standard deviation of 10 were used, positioning these between 60 and 110 points, and they are centered between 80 and 90 points. The scale score was given in six increasing levels, being more evident, the moderate (37.2%) and sufficient (43.7%). It is concluded that the QPAC presented evidence of content validity and adequate internal structure for application by telemonitoring, being easy to apply, low cost and that will enable the construction of health indicators on complementary feeding for individuals with low health literacy.