Factor Structure and Construct Validity of a Hospital Survey on Patient Safety Culture Using Exploratory Factor Analysis

Abstract

Objective Nurses' voluntary reporting of adverse events and errors is critical for improving patient safety. The operationalization and application of the concept, patient safety culture, warrant further study. The objectives are to explore the underlying factor structure, the correlational relationship, between items of the Agency for Healthcare Research and Quality Hospital Survey on Patient Safety Culture and examine its construct validity. Methods Exploratory factor analysis was conducted using secondary data from the instrument’s database. Using pattern matching, factors obtained through exploratory factor analysis were compared with the 6-component Patient Safety Culture Theoretical Framework: degree of psychological safety, degree of organizational culture, quality of culture of safety, degree of high reliability organization, degree of deference to expertise, and extent of resilience. Results 6 exploratory factors, explaining 51% of the total variance, were communication lead/speak out/resilience, organizational culture and culture of safety–environment, psychological safety–security/protection, psychological safety-support/trust, patient safety, communication, and reporting for patient safety. All factors had moderate to very strong associations (range, 0.354–0.924). Overall, construct validity was good, but few exploratory factors matched the theoretical components of degree of deference to expertise and extent of resilience. Conclusions Factors essential to creating an environment of transparent, voluntary error reporting are proposed. Items are needed, specifically focusing on deference to expertise, the ability of the person with the most experience to speak up and lead, despite hierarchy or traditional roles, and resilience, which is coping and moving forward after adversity or mistakes. With future studies, a supplemental survey with these items may be proposed.