Validación de la versión mexicana del cuestionario pro children project en diferentes muestras mexicanas

Abstract

Abstract The evaluation of psychosocial factors related to the dietary behavior of fruit and vegetable intake (F/V I) is fundamental for designing effective intervention programs for the prevention of chronic diseases and obesity in schoolchildren. Despite the importance of these factors in Mexico, research on theoretical and predictive models of F/V I in schoolchildren in Mexico is scarce. Additionally, to this situation, there are no valid and reliable instruments to evaluate individual and socio-environmental factors associated with low F/V I in schoolchildren. Therefore, in order to overcome these limitations related to evaluation and research, and to contribute to the applied knowledge about the evaluation of psychosocial factors based on research evidence, this Doctoral Thesis is presented with three general objectives. The main objectives of this Doctoral Thesis were: 1) to adapt and validate the Pro Children Project Questionnaire (PCHPQ; De Bourdeaudhuij et al., 2005) in Mexican schoolchildren population; 2) to validate of a theoretical model in explaining the preferences for eating fruit and actual fruit intake in samples of Mexican schoolchildren; and, 3) to validate a predictive model with psychosocial variables of vegetable intake in samples of Mexican schoolchildren. The first objective was to determine the content validity, construct validity, the test-retest reliability, the internal consistency, and the predictive validity of the PCHPQ in several samples of 10 to 12 year-old Mexican schoolchildren. Afterward, for the second objective we tested the goodness of fit of the Motivation-Ability-Environmental Opportunity model (MAEO model), with the purpose of evaluating the observed variance in Mexican schoolchildren’s preferences to eat fruit and daily fruit intake; we also evaluated the factorial invariance across the gender and type of population (urban and semi-urban) of schoolchildren. Finally, in order to achieve the third objective, we explored the predictive power of motivation, ability and environmental opportunity factors associated with vegetable intake among Mexican schoolchildren from six cities of the State of Chihuahua, Mexico. In this study we explored to evaluate the differences in psychosocial potential factors of motivation, ability and opportunity of vegetable intake between urban and semi-urban population schoolchildren 10 to 12-year-old. According to the plan to achieve the objectives and steps of this Doctoral Thesis, four studies associated with three main objectives were performed, that are summarized below: Objective 1: To evaluate the reliability and validity of the Pro Children Project Questionnaire in Mexican population Study 1: Validity of Pro Children Project questionnaire for assessing psychosocial factors of fruit and vegetable intake in Mexican population The aim of the first study was to determine content validity and construct validity for the Mexican version of Pro Children Project questionnaire for assessing psychosocial factors, related to fruit and vegetable intake in samples of 10 - 12 year-old schoolchildren. The questionnaire consisted of 87 items. It was administrated to 2084 children in an instrumental study conducted in Ciudad Juarez, Chihuahua, Mexico. Results showed that the Kappa statistic resulted in good agreement between experts (kappa=0.69), very good agreement in children (kappa=0.93). Seven factors were obtained that explained 37.87% of the variance in fruit and 48.18% of the variance in vegetable. The Cronbachs alpha values were low to moderate (range 0.55 to 0.83). An internal structure of seven factors of first order was confirmed in fruit and six linked to vegetable. It was concluded that the questionnaire provides sufficient validity for assessing psychosocial factors related to fruit and vegetable intake in 10 – 12 - year old schoolchildren. Study 2: Reliability and validity of a Mexican version of the Pro Children Project questionnaire The aim of the second study was to determine the test-retest reliability, the internal consistency, and the predictive validity of the constructs of the Mexican version of the Pro Children Project questionnaire for assessing personal and environmental factors related to fruit and vegetable intake in 10 - 12 - year-old schoolchildren. Test-retest design with a 14 days’ interval. A sample of 957 children completed the questionnaire with 82 items. The study was conducted at eight primary schools in Ciudad Juarez, Chihuahua, Mexico. Results presented that for all fruit constructs and vegetable constructs, the test-retest reliability was moderate (intra-class correlation coefficient (ICC) > 0.60). Cronbach’s alpha values were from moderate to high (range of 0.54 to 0.92) like those in the original study. Values for predictive validity ranged from moderate to good with Spearman correlations between 0.23 and 0.60 for personal factors and between 0.14 and 0.40 for environmental factors. It was concluded that the Mexican version of the Pro Children Project questionnaire provide a sufficient reliability and validity for assessing personal and environmental factors of fruit and vegetable intake in 10-12-year old schoolchildren. Objective 2: To test the goodness of fit and equivalence of structural model in explaining the preferences to eat fruit and fruit intake in schoolchildren samples Study 3: Validity and factorial invariance of the social ecological model in explaining fruit intake in Mexican schoolchildren The aim of the third study was to test the goodness of fit of a Motivation-Ability-Opportunity model (MAO-model) to evaluate the observed variance in Mexican schoolchildren’s preferences to eat fruit and daily fruit intake; also, to evaluate the factorial invariance across the gender and type of population (urban and semi-urban) in which children reside. A model with seven constructs was designed from a validated questionnaire to assess preferences, cognitive abilities, attitude, modeling, perceived barriers, accessibility at school, accessibility at home, and fruit intake frequency. The instrument was administered in a representative sample of 1434 schoolchildren of 5th and 6th grade of primary school in a cross-sectional and ex post fact study conducted in six cities of the State of Chihuahua, Mexico. Results showed the goodness of fit indexes was adequate for the MAO-model and explained 39% of the variance in preference to eat fruit. The structure of the model showed very good factor structure stability and the dimensions of the scale were equivalent in the different samples analyzed. It was concluded that the model analyzed with structural equation modeling showed a parsimonious model that can be used to explain the variation in fruit intake of 10 to 12-year old Mexican schoolchildren. The structure of the model was strictly invariant in the different samples analyzed and showed evidence of cross validation. Objective 3: To explore the predictive power of factors associated with vegetable intake in samples of schoolchildren residing in urban and semi - urban populations Study 4: Psychosocial factors of vegetable intake in urban and semi-urban Mexican schoolchildren The aim of the fourth study was to explore a predictive capacity of psychosocial factors of motivation, ability and environmental opportunity factors of vegetable intake between urban and semi-urban schoolchildren population 10 to 12 years old of six cities of the State of Chihuahua (Mexico). A validated questionnaire was administered in a representative sample of 1434 schoolchildren of elementary school. Frequency of daily consumption of vegetables was higher in semi-urban schoolchildren (vegetables 2 portion/day). The regression analysis explained 33% (p<.01) to 45% (p<.01) of the variance in vegetable intake by potential psychosocial factors. Preferences, cognitive abilities, perceived barriers and accessibility at home were a significant related to vegetable intake in urban (B<0.18, p<.01) and semi-urban schoolchildren (B<0.11, p<.01). It is concluded that in designing effective intervention to increment vegetable intake between urban and semi-urban children should focus on increase the preferences and cognitive abilities. Also, decrease perceived barriers and facilitate the availability at home. Finally, implication of the findings in psychosocial factors and for more efficient future research interventions need to be applied on mediation analysis to determine the predictive capacity and strength of the factors to obtain more effective intervention.