Ethnic Concordance in Patient–Physician Communication: Experimental Evidence from Germany

Abstract

Minority group members tend to have more negative health outcomes compared to majority group members. As reducing health inequalities is a global imperative, research testing strategies to improve minority patient health outcomes are important. Evidence exists that ethnic concordance in patient–physician communication is statistically associated with positive outcomes for minority patients. Previous research has exclusively relied on non-experimental observational methods. The present study adds to this literature by presenting supplementary experimental evidence, thus increasing confidence in the causal interpretation of the relationships observed in previous studies. Individuals with Turkish migration backgrounds living in Germany (N = 256) were randomly assigned to a hypothetical medical consultation in which a physician, Dr. Thomas Kirsch (“German majority physician”; ethnic-discordance condition) or Dr. Çağdaş Kılıç (“Turkish minority physician”; ethnic-concordance condition) talked about lifestyle factors associated with chronic non-communicable diseases (tobacco smoking, an unhealthy diet, and physical inactivity). The analysis indicates that ethnic concordance improved belief in the physician, reduced reactance-related outcomes, and improved prevention-related knowledge transfer. Notably, the effect of ethnic concordance on knowledge was especially pronounced in low health-literacy participants. We discuss the implications related to the ongoing calls for a more diverse physician workforce.

Notes

¹ To test whether the interaction effect holds under controls, we additionally ran the regression analysis with covariates (age, gender, education, Turkish citizenship, German citizenship). This model produced almost identical results: The interaction effect was significant, B = −0.58, SE = 0.26, p = .028. Of the covariates, only age, B = 0.03, SE = 0.15, p = .001, and gender, B = −0.43, SE = 0.15, p = .004, elicited significant effects: Older and female participants showed higher prevention-related knowledge scores. The full model can be obtained upon request.

² One reviewer noted his/her idea that ethnic concordance may have elicited a higher attention which in turn influenced our health outcomes. This idea represents a mediator model where experimental condition (X) elicited an effect on outcomes such as beliefs toward the physician (Y) via increased levels of attention (Z). Although we did not design the study for a thorough test of this idea, we ran a further analysis: Our software recorded the time participants read the messages of the hypothetical medical consultation. The three messages (see Appendix) were presented on three separate survey pages. The software recorded the time in milliseconds that participants spent on each page. We used this reading time as an attention proxy. Using three t-tests with experimental condition as the independent variable and reading time as the dependent variable, we found no significant differences. This analysis can be obtained upon request.

Additional information

Funding

This research was supported by Deutsche Forschungsgemeinschaft DFG (grant number AR 1045/1-1)