Appsolutely secure? Psychometric properties of the German version of an app information privacy concerns measure during COVID-19.

Introduction: Privacy concerns are an important barrier to adoption and continued use of digital technologies, particularly in the health sector. With the introduction of mobile health applications (mHealth apps), the construct of app information privacy concerns has received increased attention. However, few validated measures exist to capture said concerns in population samples, although they can help to improve public health efforts. Methods: Using a cross-sectional survey of German adults (mean age = 35.62; 63.5% female), this study examined psychometric properties of the app information privacy concerns scale (AIPC). Analyses comprised confirmatory factor analysis, factorial validity (exploratory factor analysis), internal consistency, convergent validity (i.e., correlations with privacy victimhood, and app privacy concerns), and discriminant validity (i.e., daily app use, adoption intentions, and attitudes toward COVID-19 contact tracing app use). Results: The analysis did not support the proposed three-factor structure of the AIPC (i.e., anxiety, personal attitude, and requirements). Instead, a four-factor model was preferable that differentiated requirements regarding disclosure policies, and personal control. In addition, factors mirroring anxiety and personal attitude were extracted, but shared a significant overlap. However, these factors showed good reliability, convergent and discriminant validity. Discussion: The findings underline the role of app information privacy concerns as a significant barrier to mHealth app use. In this context, anxiety and personal attitudes seemed particularly relevant, which has implications for health communication. Moreover, the observed differentiation of external (disclosure) and internal (control) requirements aligns with health behavior change models and thus is a promising area for future research.