The Internal Reliability and Construct Validity of the Evidence-Based Practice Questionnaire (EBPQ): Evidence from Healthcare Professionals in the Eastern Mediterranean Region

Abstract

Background: Fostering a culture of clinical effectiveness among healthcare professionals (HCPs) is crucial to achieving optimal patient health outcomes. To our knowledge, there is a lack of robust evidence-based practice (EBP) tools to assess the competence of HCPs in EBP in the Eastern Mediterranean Region (EMR). Aim: This study aims to comprehensively investigate the construct validity and internal reliability of the evidence-based practice questionnaire (EBPQ) among HCPs in the EMR. Methods: This multinational and multi-disciplinary cross-sectional study was conducted between 27 April and 11 May 2023. Convenience and snowball sampling methods were used to recruit a sample of HCPs (physicians, nurses, physiotherapists, dentists, and pharmacists) using an electronic survey questionnaire for data capture. To assess the reliability of the instrument, Cronbach’s alpha, inter-item reliability, and split-half reliability analyses were conducted. Furthermore, the convergent and discriminant validity of the questionnaire was ensured by calculating the average variance extracted (AVE) and the correlation coefficient between the different constructs, respectively. Factor loadings and cross-loadings of different indicators within each construct were calculated by performing both exploratory and confirmatory factor analyses. Results: A total of 1536 HCPs from 18 countries in the EMR (response rate = 96.786%) with a median age of 28 years participated; 47% were female, and 55% had Arabic as their first language. English was the most common language for a bachelor’s degree in science (54%). The construct validity of the EBPQ was investigated using exploratory factor analysis (EFA), which yielded four loaded factors. The confirmatory factor analysis (CFA) confirmed the four loaded factors. The CFA model showed that the root mean square error of approximation = 0.066, comparative fit index = 0.95, Tucker–Lewis’s index = 0.94, standardized root mean square residual = 0.033, normal fit index = 0.94, goodness of fit = 0.91, and χ2 test statistic= 22,553, with p < 0.001. The AVE values of the four factors were close to 1 (knowledge = 0.6, practice = 0.6, attitude = 0.5, and sharing = 0.7), thus supporting the convergent validity of the EBPQ. The four domains had Cronbach’s alpha coefficients and Omega ≥ 0.7 (knowledge = 0.9, practice = 0.9, attitude = 0.7, and sharing = 0.8), suggesting that the items within each domain had good internal consistency. These results support the discriminant validity of the EBPQ. Conclusions: The EBPQ is a robust questionnaire that can be completed in less than 10 min by EMR HCPs and can be used as a gold-standard questionnaire to collect valid data on the attitudes, knowledge, and proficiency of HCPs in making clinical decisions based on evidence. Future studies are recommended to investigate the retest reliability.