Validation of the Spanish version of the evidence based practice questionnaire in nurses

Background: The lack of adequate instruments prevents the possibility of assessing the competence of health care staff in evidence-based decision making and further, the identification of areas for improvement with tailored strategies. The aim of this study is to report about the validation process in the Spanish context of the Evidence-Based Practice Questionnaire (EBPQ) from Upton y Upton. Methods: A multicentre, cross-sectional, descriptive psychometric validation study was carried out. For cultural adaptation, a bidirectional translation was developed, accordingly to usual standards. The measuring model from the questionnaire was undergone to contrast, reproducing the original structure by Exploratory Factorial Analysis (EFA) and Confirmatory Factorial Analysis (CFA), including the reliability of factors. Results: Both EFA (57.545% of total variance explained) and CFA (chi2=2359,9555; gl=252; p<0.0001; RMSEA=0,1844; SRMR=0,1081), detected problems with items 7, 16, 22, 23 and 24, regarding to the original trifactorial version of EBPQ. After deleting some questions, a reduced version containing 19 items obtained an adequate factorial structure (62.29% of total variance explained), but the CFA did not fit well. Nevertheless, it was significantly better than the original version (chi2=673.1261; gl=149; p<0.0001; RMSEA=0.1196; SRMR=0.0648). Conclusions: The trifactorial model obtained good empiric evidence and could be used in our context, but the results invite to advance with further refinements into the factor “attitude”, testing it in more contexts and with more diverse professional profiles.YesFundamento. La falta de instrumentos adecuados impide conocer la competencia de los profesionales en la toma de decisiones basadas en la evidencia e identificar áreas de mejora en las que intervenir con estrategias “ad hoc”. El objetivo de la investigación es validar en el entorno español del Evidence-Based Practice Questionnaire (EBPQ) (Upton y Upton, 2006). Métodos. Se llevó a cabo un estudio multicéntrico, descriptivo y transversal de validación psicométrica. Se siguió un proceso de traducción-retrotraducción según los estándares habituales. Se contrastó el modelo de medida del cuestionario, replicando la estructura original anglosajona mediante análisis factorial exploratorio (AFE) y confirmatorio (AFC), incluyendo análisis de la fiabilidad de los factores. Resultados. Tanto el AFE (57,545% de la varianza total explicada) como el AFC (chi2=2359,9555; gl=252; p<0,0001; RMSEA=0,1844; SRMR=0,1081) han detectado problemas en los ítems 7, 16, 22, 23 y 24 sobre la versión original trifactorial del EBPQ. Una versión reducida de 19 ítems presenta una estructura factorial exploratoria adecuada (62,29% de varianza total explicada), aunque el AFC no mostró adecuación, fue significativamente mejor que en la versión original (chi2=673.1261; gl=149; p