Search

BackgroundChatGPT, a conversational artificial intelligence developed by OpenAI, has rapidly become an invaluable tool for researchers. With the recent integration of Python code interpretation into the ChatGPT environment, there has been a significant increase in the potential utility of ChatGPT as a research tool, particularly in terms of data analysis applications. ObjectiveThis study aimed to assess ChatGPT as a data analysis tool and provide researchers with a framework for applying ChatGPT to data management tasks, descriptive statistics, and inferential statistics. MethodsA subset of the National Inpatient Sample was extracted. Data analysis trials were divided into data processing, categorization, and tabulation, as well as descriptive and inferential statistics. For data processing, categorization, and tabulation assessments, ChatGPT was prompted to reclassify variables, subset variables, and present data, respectively. Descriptive statistics assessments included mean, SD, median, and IQR calculations. Inferential statistics assessments were conducted at varying levels of prompt specificity (“Basic,” “Intermediate,” and “Advanced”). Specific tests included chi-square, Pearson correlation, independent 2-sample t test, 1-way ANOVA, Fisher exact, Spearman correlation, Mann-Whitney U test, and Kruskal-Wallis H test. Outcomes from consecutive prompt-based trials were assessed against expected statistical values calculated in Python (Python Software Foundation), SAS (SAS Institute), and RStudio (Posit PBC). ResultsChatGPT accurately performed data processing, categorization, and tabulation across all trials. For descriptive statistics, it provided accurate means, SDs, medians, and IQRs across all trials. Inferential statistics accuracy against expected statistical values varied with prompt specificity: 32.5% accuracy for “Basic” prompts, 81.3% for “Intermediate” prompts, and 92.5% for “Advanced” prompts. ConclusionsChatGPT shows promise as a tool for exploratory data analysis, particularly for researchers with some statistical knowledge and limited programming expertise. However, its application requires careful prompt construction and human oversight to ensure accuracy. As a supplementary tool, ChatGPT can enhance data analysis efficiency and broaden research accessibility.

Abstract Background Various studies have revealed that there is a disproportionately low representation of skin of colour (SOC) in medical school dermatologic curriculum and board study resources. Methods First‐year and second‐year medical students were emailed an 18‐question survey regarding (1) identifying correct diagnoses of dermatologic conditions on either White skin or SOC and (2) their confidence in identifying dermatologic conditions on SOC. Results 15% of the images of dermatologic conditions included in the institutional preclinical curriculum show images of patients with SOC. Regarding overall scores for diagnosing dermatologic diseases, students performed similarly on both the White image survey (61.73%) and SOC image survey (66.20%) with no statistically significant differences between surveys (p = 0.14). Second‐year medical students performed better than first‐year medical students overall (p = 0.01) and on White skin image survey scores (p = 0.02) but not on people of colour image survey scores (p = 0.09). Students largely agreed that they were more comfortable identifying dermatologic diagnoses on White skin and that their school could benefit from increased SOC dermatological resources. Conclusion The overall low scores for the diagnosis of common skin conditions on both the White image and SOC image survey by first‐ and second‐year students are not surprising given the results of a prior study and support the need for re‐exposure to dermatology presentations in all skin types during the preclinical curriculum. The low scores support the need for changes in the pre‐clinical dermatology curriculum to improve diagnostic ability. Second‐year students performed better at diagnosing dermatologic conditions overall and on White skin compared to SOC at the end of their didactic years, possibly due to an underrepresentation of SOC images in institutional and outside educational resources. Attitudes towards school‐specific dermatologic SOC education demonstrated a clear desire amongst students for more exposure to dermatologic conditions in various skin colours throughout the curriculum.

Abstract INTRODUCTION Identifying mild cognitive impairment (MCI) patients at risk for dementia could facilitate early interventions. Using electronic health records (EHRs), we developed a model to predict MCI to all‐cause dementia (ACD) conversion at 5 years. METHODS Cox proportional hazards model was used to identify predictors of ACD conversion from EHR data in veterans with MCI. Model performance (area under the receiver operating characteristic curve [AUC] and Brier score) was evaluated on a held‐out data subset. RESULTS Of 59,782 MCI patients, 15,420 (25.8%) converted to ACD. The model had good discriminative performance (AUC 0.73 [95% confidence interval (CI) 0.72–0.74]), and calibration (Brier score 0.18 [95% CI 0.17–0.18]). Age, stroke, cerebrovascular disease, myocardial infarction, hypertension, and diabetes were risk factors, while body mass index, alcohol abuse, and sleep apnea were protective factors. DISCUSSION EHR‐based prediction model had good performance in identifying 5‐year MCI to ACD conversion and has potential to assist triaging of at‐risk patients. Highlights Of 59,782 veterans with mild cognitive impairment (MCI), 15,420 (25.8%) converted to all‐cause dementia within 5 years. Electronic health record prediction models demonstrated good performance (area under the receiver operating characteristic curve 0.73; Brier 0.18). Age and vascular‐related morbidities were predictors of dementia conversion. Synthetic data was comparable to real data in modeling MCI to dementia conversion. Key Points An electronic health record–based model using demographic and co‐morbidity data had good performance in identifying veterans who convert from mild cognitive impairment (MCI) to all‐cause dementia (ACD) within 5 years. Increased age, stroke, cerebrovascular disease, myocardial infarction, hypertension, and diabetes were risk factors for 5‐year conversion from MCI to ACD. High body mass index, alcohol abuse, and sleep apnea were protective factors for 5‐year conversion from MCI to ACD. Models using synthetic data, analogs of real patient data that retain the distribution, density, and covariance between variables of real patient data but are not attributable to any specific patient, performed just as well as models using real patient data. This could have significant implications in facilitating widely distributed computing of health‐care data with minimized patient privacy concern that could accelerate scientific discoveries.

Healthcare workers have a unique opportunity to recognize and provide support to HT/DV victims. The objective of this study is to determine if students are more prepared to identify and work with these patient populations after being trained on the topic by the International Rescue Committee (IRC). The IRC came to the University of Arizona College of Medicine- Phoenix (UACOMP) to do an optional one hour training of HT/DV in healthcare for current first and second year medical students. After IRB approval from the UACOMP Research Office for Medical Education (ROME), pre and post surveys were given to a total of 26 students who attended the training. Wilcoxon Signed-Rank tests adjusted for multiple comparisons were used to evaluate differences between pre- and post-training scores. Students had statistically significant improvement in self-reported comfort level scores for recommending resources, providing care, and calling the national hotline for individuals who have experienced human trafficking and domestic violence. The marked improvement in students’ comfort providing resources and care to HT/DV victims after just a 1 h training from the IRC on these topics demonstrates the impact that early exposure to HT/DV education can have in vulnerable patient populations.