Psychometric Evaluation of the Postoperative Recovery Profile.

Abstract

To further evaluate the postoperative recovery profile regarding its psychometric properties. The postoperative recovery profile is an instrument for the self-assessment of general postoperative recovery that has received increased attention within nursing research. However, psychometric evaluation during development was sparse. Psychometric evaluation was done using classical test theory. Data quality, targeting, reliability, and scaling assumptions were measured. In addition, confirmatory factor analysis was used to evaluate construct validity. Data collection was made during 2011-2013. Data derived from this study showed acceptable quality; however, item distribution was skewed, with ceiling effects in the majority of items. Cronbach's alpha showed high internal consistency. Item-total correlations indicated unidimensionality, whereas six items demonstrated high correlations pointing at redundancy. The confirmatory factor analysis confirmed problems related to dimensionality as the five proposed dimensions were highly correlated with each other. Furthermore, items were largely uncorrelated with the designated dimensions. This study shows that the postoperative recovery profile needs to be further developed to serve as a robust instrument within nursing as well as medical research. Arguably, values from the instrument should not be calculated at a dimensional level for the time being because of discriminant validity issues.