Danish Translation and Adaptation of the Context Assessment Index With Implications for Evidence‐Based Practice. Issue 3 (28th January 2019)

• Record Type: Journal Article
• Authors: Hølge‐Hazelton, Bibi; Bruun, Line Zacho; Slater, Paul; McCormack, Brendan; Thomsen, Thora Grothe; Klausen, Susanne Hwiid; Bucknall, Tracey
• Is Part Of: Worldviews on evidence-based nursing. Volume 16:Issue 3(2019)
• Page Start: 221
• Page End: 229
• Publication Date: 2019-01-28
• Subjects: Context Assessment Index -- translation -- nursing -- psychometric evaluation -- PARIHS framework -- instrument development -- context -- person centred care -- clinical decision making -- capacity building; Evidence-based nursing -- Periodicals; 610.7
• DOI: 10.1111/wvn.12347
• Languages: English
• ISSNs: 1545-102X

Abstract

Abstract: Background: Healthcare contexts are rapidly changing, with growing demand for health services to accommodate an aging population and financial pressures. Assessment of context in healthcare settings has been the subject of increasing debate. The Context Assessment Index (CAI) examines three interconnected contextual elements derived from the Promoting Action on Research Implementation in Health Services (PARIHS) framework to provide practitioners with an understanding of the context in which they work. Aims: (a) To describe the translation of the CAI into Danish and adapt the instrument for use in Danish hospitals and (b) to evaluate the psychometric properties of the Danish version of the CAI. Methods: Translation and adaptation included an expert panel and a translation and back‐translation process. The CAI was then sent to 4, 416 nurses in the Region Zealand, Denmark. There are two alternative measurement models to explain the factor structure of the CAI, the five‐factor model and the three‐factor model. To provide the best explanation for the data, both measurement models were examined using confirmatory factor analysis. Results: The CAI was translated and modified based on expert review and usability testing. Completing the CAI were 2, 261 nurses. For both models, factor loadings and fit statistics were acceptable, appropriate, and statistically significant, and the measurement models were confirmed (five‐factor model: root mean square error of approximation [RMSEA] 0.07, comparative fit index [CFI] = 0.923; three‐factor model: RMSEA 0.07, CFI = 0.924). Cronbach's alpha scores showed the models to have broadly acceptable scores (five‐factor model 0.64–0.89; three‐factor model 0.72–0.89). Linking Evidence to Action: The three‐factor model can advantageously be used when the PARIHS framework is part of the project. In a translation process, differences in cultural specificity, language, and working environment have to be considered. By understanding the context of practice, nurses may enable person‐centered care and improve patient outcomes.