Challenging the Chatbot: An Assessment of ChatGPT's Diagnoses and Recommendations for DBP Case Studies.

OBJECTIVE: Chat Generative Pretrained Transformer-3.5 (ChatGPT) is a publicly available and free artificial intelligence chatbot that logs billions of visits per day; parents may rely on such tools for developmental and behavioral medical consultations. The objective of this study was to determine how ChatGPT evaluates developmental and behavioral pediatrics (DBP) case studies and makes recommendations and diagnoses. METHODS: ChatGPT was asked to list treatment recommendations and a diagnosis for each of 97 DBP case studies. A panel of 3 DBP physicians evaluated ChatGPT's diagnostic accuracy and scored treatment recommendations on accuracy (5-point Likert scale) and completeness (3-point Likert scale). Physicians also assessed whether ChatGPT's treatment plan correctly addressed cultural and ethical issues for relevant cases. Scores were analyzed using Python, and descriptive statistics were computed. RESULTS: The DBP panel agreed with ChatGPT's diagnosis for 66.2% of the case reports. The mean accuracy score of ChatGPT's treatment plan was deemed by physicians to be 4.6 (between entirely correct and more correct than incorrect), and the mean completeness was 2.6 (between complete and adequate). Physicians agreed that ChatGPT addressed relevant cultural issues in 10 out of the 11 appropriate cases and the ethical issues in the single ethical case. CONCLUSION: While ChatGPT can generate a comprehensive and adequate list of recommendations, the diagnosis accuracy rate is still low. Physicians must advise caution to patients when using such online sources.

Pediatric Traumatic Brain Injury Outcomes and Disparities During the COVID-19 Pandemic.

BACKGROUND: To study the impact of the COVID-19 pandemic on traumatic brain injury (TBI) patient demographic, clinical and trauma related characteristics, and outcomes. METHODS: Retrospective chart review was conducted on pediatric TBI patients admitted to a Level I Pediatric Trauma Center between January 2015 and June 2022. The pre-COVID era was defined as January 1, 2015, through March 12, 2020. The COVID-19 era was defined as March 13, 2020, through June 30, 2022. Bivariate analysis and logistic regression were performed. RESULTS: Four hundred-thirty patients were treated for pediatric TBI in the pre-COVID-19 period, and 166 patients during COVID-19. In bivariate analyses, the racial/ethnic makeup, age, and sex varied significantly across the two time periods (p < 0.05). Unwitnessed TBI events increased during the COVID-19 era. Logistic regression analyses also demonstrated significantly increased odds of death, severe disability, or vegetative state during COVID-19 (AOR 7.23; 95 % CI 1.43, 36.41). CONCLUSION: During the COVID-19 pandemic, patients admitted with pediatric TBI had significantly different demographics with regards to age, sex, and race/ethnicity when compared to patients prior to the pandemic. There was an increase in unwitnessed events. In the COVID period, patients had a higher odds ratio of severe morbidity and mortality despite adjustment for confounding factors. LEVEL OF EVIDENCE AND STUDY TYPE: Level II, Prognosis.

Donor Designation Among Mature Latinas and Lay Health Educators (Promotoras): A Mixed-Methods Study.

BACKGROUND: Despite positive public attitudes toward solid organ donation in the United States, some of the lowest rates of donor designation persist among older adults and Latinx populations. AIMS: To identify barriers and facilitators to organ donation and donor designation among lay health educators (promotoras) and mature Latina (50+ years). METHODS: An explanatory sequential mixed-method design was employed, with telephone surveys followed by focus group interviews, to assess and understand the nuances of organ donation and donor designation knowledge, attitudes, and practices among promotoras and mature Latinas in Chicago (IL), Philadelphia (PA), and San Antonio (TX). Descriptive statistics summarized quantitative survey data; thematic content analysis was performed on qualitative data. RESULTS: Twenty-nine promotoras and 45 mature Latina participated in both the surveys and focus groups (N = 74). Most participants (90%) had limited knowledge of organ donation but reported being "somewhat" or "strongly" in favor of donation (70%); 40.5% were registered donors. Participants lacked knowledge about the registration process and its legal standing and upheld concerns that registered donors would be vulnerable to organ traffickers or targets for murder. Themes emerging from the group interviews revealed additional barriers to designation including distrust of the medical establishment, perceptions of inequities in organ allocation, and family resistance to discussing death. DISCUSSION: Low donor designation rates are primarily driven by concerns about organ trafficking and the fairness of the allocation system, particularly for undocumented immigrants. CONCLUSIONS: The results informed development of a culturally targeted educational and communication skills intervention to increase donor designation in Latinx communities. TRIAL REGISTRATION: ClinicalTrials.gov NCT04007419.

Failure mode and effects analysis drastically reduced potential risks in clinical trial conduct.

BACKGROUND: Failure mode and effects analysis (FMEA) is a risk management tool to proactively identify and assess the causes and effects of potential failures in a system, thereby preventing them from happening. The objective of this study was to evaluate effectiveness of FMEA applied to an academic clinical trial center in a tertiary care setting. METHODS: A multidisciplinary FMEA focus group at the Seoul National University Hospital Clinical Trials Center selected 6 core clinical trial processes, for which potential failure modes were identified and their risk priority number (RPN) was assessed. Remedial action plans for high-risk failure modes (RPN >160) were devised and a follow-up RPN scoring was conducted a year later. RESULTS: A total of 114 failure modes were identified with an RPN score ranging 3-378, which was mainly driven by the severity score. Fourteen failure modes were of high risk, 11 of which were addressed by remedial actions. Rescoring showed a dramatic improvement attributed to reduction in the occurrence and detection scores by >3 and >2 points, respectively. CONCLUSIONS: FMEA is a powerful tool to improve quality in clinical trials. The Seoul National University Hospital Clinical Trials Center is expanding its FMEA capability to other core clinical trial processes.