Improving patient safety is a priority for health care providers in the United States; however medical errors remain a persistent problem. A strong patient-safety culture is associated with decreased medical errors. In hospitals, registered nurses are in a position to identify medical errors and communicate their suggestions to improve patient safety. Despite the importance of communication by registered nurses and the potential impact on improved patient-safety culture, these constructs are largely unexamined in the literature. This quantitative, correlational study was designed to investigate the direct and moderating relationships between registered nurses' perception of prosocial voice, self-monitoring, and patient-safety culture in a hospital setting. Structure-process-outcome (SPO) theory guided the study. Registered nurses were recruited from one hospital located in the Midwest United States. Participants completed their choice of a paper-pencil or an online survey consisting of demographic questions and 36 items from three published scales. Three hundred thirty seven completed surveys were returned. Bivariate correlations and multiple linear regressions were computed to determine the relationships among the variables. Results showed a weak positive correlation between registered nurses' prosocial voice and four dimensions of hospital patient-safety culture: (a) continuous improvement, r(337) = .156; p = .004; (b) communication about errors, r(337) = .177; p = .001; (c) nonpunitive response to errors, r(337) = .131; p = .016; and (d) hospital-management support, r(337) = .117; p = .032. Regression analyses did not support self-monitoring as a moderator of the relationship between prosocial voice and hospital patient-safety culture. The study was the first known to examine the relationships between registered nurses' prosocial voice, self monitoring, and hospital patient-safety culture using SPO theory. Findings may be of benefit to health care leaders seeking to strengthen hospital patient-safety culture and reduce medical errors. Recommendations for further research included studying larger subject populations in diverse hospital settings. Longitudinal and qualitative investigations are also recommended.