In Reply to Li et al

In Reply to Li et al: We thank Li and colleagues for their insightful comments on our recent survey validation effort. We understand the concern about using the 10-item Center for Epidemiological Studies Depression Scale (CES-D-10)—which has not been specifically validated in medical students—for what Li and colleagues refer to as “convergent validity.” It is important to note, however, that our approach to validation was much broader than simply correlating our mean shame frequency score with the CES-D-10. The correlational analysis was one small part of our validity argument, the sources of which were guided by Messick’s validity framework.1 Our systematic approach encompassed multisource evidence to support the validity of inferences that might be made using scores from our shame frequency questionnaire, beyond just convergent validity with CES-D-10 scores.1,2 We used the findings of several qualitative studies on shame in medical trainees to establish the content of the initial survey draft. We conducted cognitive interviews to ensure item clarity and respondent comprehension. To confirm factorial structure and internal consistency reliability, we performed factor and reliability analyses. We examined the relationship between the mean shame frequency score and the CES-D-10 score, and we examined correlations with other conceptually related variables, including anxiety, emotional exhaustion, emotional recovery, and emotional thriving. According to the Standards for Educational and Psychological Testing,2 when validating a new assessment tool, the correlational patterns should align with theoretical expectations, which our findings supported. Regarding the authors’ suggestion to use the 20-item CES-D or the Patient Health Questionnaire-9 (PHQ-9), we considered these options but found them less suitable for our study context. The CES-D-20 is longer and could have increased respondent burden, potentially impacting response rates and data quality. Instead, we specifically chose the CES-D-10 because it is the short form of the CES-D. The PHQ-9 includes a question about suicidality, and we elected to avoid the potential for respondents to voice suicidality on an anonymous survey that would preclude follow-on psychological support. Ultimately, we agree that more validity evidence is needed for our new shame frequency questionnaire and that our efforts are a first step. Additionally, we welcome further research to validate commonly used scales for well-being constructs like shame and depression in medical students, and we appreciate the authors’ recommendations in this regard. Anthony R. Artino Jr, PhDProfessor of health, human function, and rehabilitation sciences, and associate dean for educational research, George Washington University School of Medicine and Health Sciences, Washington, DC; ORCID: http://orcid.org/0000-0003-2661-7853William E. Bynum IV, MD, PhDAssociate professor of family medicine, Department of Family Medicine and Community Health, Duke University School of Medicine, Durham, North Carolina; email: [email protected]; X (formerly Twitter): @WillBynumMD; ORCID: https://orcid.org/0000-0003-3796-9301