Language has a critical role as a social determinant of health and a source of healthcare disparities. Rhetorical devices are ubiquitous in medicine and are often used to persuade or inform care team members. Rhetorical devices help a healthcare team acknowledge and interpret narratives. For example, metaphors are frequently used as rhetorical devices by patients to describe cancer, including winning or losing a battle, surviving a fight, war, potentially implying that the patient feels helpless like a pawn fighting in a struggle directed by the physician, thus reducing patient autonomy and agency. However, this occidental approach is flawed because it excessively focuses on the individual's agency and marginalizes external factors, such as cultural beliefs and social support (Sontag, 1989). Although there is a large body of research about how the rhetoric of medicine affects patients in the United States, there is a lack of such research about how patient experiences' rhetoric can help increase the understanding of Latino populations' unique social determinants. This creative project aims to analyze the rhetorical differences in the description of disease amongst Latino and American communities, translating to creating an educational module for a Spanish for biomedical sciences class. The objective is to increase future healthcare professionals' ability to understand how the composition of descriptions and medical rhetoric in different mediums of humanities can serve as critical tools to analyze social determinants in Latino healthcare delivery.

Reducing the amount of error and introduced data variability increases the accuracy of Western blot results. In this study, different methods of normalization for loading differences and data alignment were explored with respect to their impact on Western blot results. GAPDH was compared to the LI-COR Revert total protein stain as a loading control. The impact of normalizing data to a control condition, which is commonly done to align Western blot data distributed over several immunoblots, was also investigated. Specifically, this study addressed whether normalization to a small subset of distinct controls on each immunoblot increases pooled data variability compared to a larger set of controls. Protein expression data for NOX-2 and SOD-2 from a study investigating the protective role of the bradykinin type 1 receptor in angiotensin-II induced left ventricle remodeling were used to address these questions but are also discussed in the context of the original study. The comparison of GAPDH and Revert total protein stain as a loading control was done by assessing their correlation and comparing how they affected protein expression results. Additionally, the impact of treatment on GAPDH was investigated. To assess how normalization to different combinations of controls influences data variability, protein data were normalized to the average of 5 controls, the average of 2 controls, or an average vehicle and the results by treatment were compared. The results of this study demonstrated that GAPDH expression is not affected by angiotensin-II or bradykinin type 1 receptor antagonist R-954 and is a less sensitive loading control compared to Revert total protein stain. Normalization to the average of 5 controls tended to reduce pooled data variability compared to 2 controls. Lastly, the results of this study provided preliminary evidence that R-954 does not alter the expression of NOX-2 or SOD-2 to an expression profile that would be expected to explain the protection it confers against Ang-II induced left ventricle remodeling.