Physician and patient interactions : the role of beliefs and values in directing clinical conversations.




Franzen, Aaron B.

Journal Title

Journal ISSN

Volume Title



This dissertation examines how physicians’ beliefs and values influence the content of their conversations with patients. After an introductory chapter, chapter two primarily provides an overview of the data and the religious beliefs and practices of physicians in the United States. Physicians in the United States tend to be more spiritual and less religious, and practice and affiliate more than the general population but believe less. These trends, in turn, relate to how they converse with patients. Chapter three focuses on who discusses religion and spirituality with patients and why this might be the case. A physician’s ability to connect with patients depends at least in part on his or her ability to empathize with them, but some physicians will be disadvantaged in their ability to connect due to a lack of shared experiences. Being able to connect with a religious patient will depend on the physician’s own religious/spiritual orientation and whether they see a connection between religion and medicine. Using a mediated bi-factor structural equation model, I find that physicians who are religious and spiritual are most likely to have made this religion-medicine link and talk to patients about it.
Instead of asking who talks to patients about religion, chapter four analyzes how physicians react when it does come up with patients. By analyzing a series of mediated path models, I again find that those physicians who have connected their beliefs and the work they do are least likely to avoid religion in the clinical context.
Chapter five examines whether there is a relationship between physicians’ religious characteristics and the religious characteristics of their county when predicting religious clinical interactions. After proposing competing hypotheses stating that the population characteristics will be important and that structural constraints are more important, I find that the population characteristics do not influence clinical conversations.
Chapter six briefly reviews the theoretical implications of the empirical chapters and considers the importance of the findings for future research. In doing do, I suggest a number of potentially helpful future developments.