Abstract
Positive attitudes toward caring for dying patients are core competencies in medical education and particularly relevant to cancer education, where physicians must be prepared for communication, symptom management, and end-of-life decision-making. The Frommelt Attitude Toward the Care of the Dying (FATCOD) scale is widely used to assess these attitudes, yet the original 30-item version has shown psychometric limitations across cultures. This study examined the factorial structure and measurement properties of the FATCOD in the U.S. academic setting and tested the performance of an 11-item version previously validated in Italy. A cross-sectional survey was conducted among first- and second-year medical students at Loma Linda University School of Medicine (California, USA) during the 2024–2025 academic year. Participants completed the 30-item FATCOD online. Psychometric evaluation combined Principal Component Analysis, Confirmatory Factor Analysis, and Rasch modeling using Partial Credit and Rating Scale Models. Among 295 respondents, the 30-item FATCOD showed modest internal consistency (α = 0.667) and poor structural fit, with neither PCA nor CFA supporting its hypothesized multidimensional structure. By contrast, the 11-item version (FATCOD-11) demonstrated satisfactory fit (CFI = 0.932; RMSEA = 0.056; SRMR = 0.043) and unidimensionality under Rasch analysis, with ordered thresholds and a Person Separation Index of 2.04, indicating discrimination across three levels of attitude. In this sample of U.S. medical students, the FATCOD-11 outperformed the original scale and appears to be a robust instrument for assessing attitudes toward end-of-life care and evaluating educational interventions in palliative and oncology care. Future research may expand this study to other countries or other health professionals dealing with end-of-life care.