Evaluation of ChatGPT-Generated Differential Diagnosis for Common Diseases With Atypical Presentation: Descriptive Research.

Background: The persistence of diagnostic errors, despite advances in medical knowledge and diagnostics, highlights the importance of understanding atypical disease presentations and their contribution to mortality and morbidity. Artificial intelligence (AI), particularly generative pre-trained transformers like GPT-4, holds promise for improving diagnostic accuracy, but requires further exploration in handling atypical presentations. Objective: This study aimed to assess the diagnostic accuracy of ChatGPT in generating differential diagnoses for atypical presentations of common diseases, with a focus on the model's reliance on patient history during the diagnostic process. Methods: We used 25 clinical vignettes from the Journal of Generalist Medicine characterizing atypical manifestations of common diseases. Two general medicine physicians categorized the cases based on atypicality. ChatGPT was then used to generate differential diagnoses based on the clinical information provided. The concordance between AI-generated and final diagnoses was measured, with a focus on the top-ranked disease (top 1) and the top 5 differential diagnoses (top 5). Results: ChatGPT's diagnostic accuracy decreased with an increase in atypical presentation. For category 1 (C1) cases, the concordance rates were 17% (n=1) for the top 1 and 67% (n=4) for the top 5. Categories 3 (C3) and 4 (C4) showed a 0% concordance for top 1 and markedly lower rates for the top 5, indicating difficulties in handling highly atypical cases. The χ2 test revealed no significant difference in the top 1 differential diagnosis accuracy between less atypical (C1+C2) and more atypical (C3+C4) groups (χ²1=2.07; n=25; P=.13). However, a significant difference was found in the top 5 analyses, with less atypical cases showing higher accuracy (χ²1=4.01; n=25; P=.048). Conclusions: ChatGPT-4 demonstrates potential as an auxiliary tool for diagnosing typical and mildly atypical presentations of common diseases. However, its performance declines with greater atypicality. The study findings underscore the need for AI systems to encompass a broader range of linguistic capabilities, cultural understanding, and diverse clinical scenarios to improve diagnostic utility in real-world settings.

Bochdalek Hernia Causing Sigmoid Colon Strangulation in an Elderly Patient.

Diaphragmatic hernia with bowel strangulation is a fatal condition requiring a prompt diagnosis. Bochdalek hernia is a common type of diaphragmatic hernia that rarely but occasionally occurs in adults. We herein report a case of Bochdalek hernia causing sigmoid colon strangulation in an elderly patient whose condition was initially misdiagnosed as empyema. The early diagnosis of strangulated bowel stemming from diaphragmatic hernia can be challenging because of its rarity and the nonspecificity of its symptoms. However, tracing the mesenteric arteries on computed tomography can enable a quick diagnosis.

Morbidity and Mortality Conferences in Internal Medicine Specialty Training in Japan: A Nationwide Cross-sectional Study.

Objective This study evaluated the implementation status of morbidity and mortality conferences in internal medicine specialty training programs in Japan. Methods This cross-sectional study surveyed hospitals in Japan with certified internal medicine specialty training programs. Program directors or equivalently responsible physicians managing certified internal medicine training programs were invited to participate in this study (n=619). Materials Data were collected using an online questionnaire that included questions about the number of morbidity and mortality conferences, types of cases covered, collaboration of the patient safety section and other health professions, and whether or not the conferences were conducted by a subspecialty department-led or program-based. Results Responses were received from 123 hospitals (19.8% response rate), of which 59 (48%) had some form of internal medicine morbidity and mortality conference in place. The average number per year was 9.63 (standard deviation: 18.12). Hospitals with morbidity and mortality conferences in subspecialty departments held significantly more conferences than X (please define X). Furthermore, the involvement of the patient safety department tended to be associated with holding more conferences. Autopsy rates were significantly higher in hospitals with program-based internal medicine morbidity and mortality conferences than subspecialty-led. Conclusion Internal medicine specialty training hospitals had more morbidity and mortality conferences than previously reported. Program-based morbidity and mortality conferences in internal medicine are associated with higher autopsy rates and may lead to an organizational reporting culture and lifelong learning attitudes that support patient safety. Collaboration with organizational management sections, such as patient safety, would be effective in implementing these conferences in internal medicine training programs.

Diagnostic error rates and associated factors for lower gastrointestinal perforation.

Lower gastrointestinal perforation is rare and challenging to diagnose in patients presenting with an acute abdomen. However, no study has examined the frequency and associated factors of diagnostic errors related to lower gastrointestinal perforation. This large-scale multicenter retrospective study investigated the frequency of diagnostic errors and identified the associated factors. Factors at the level of the patient, symptoms, situation, and physician were included in the analysis. Data were collected from nine institutions, between January 1, 2015 and December 31, 2019. Timely diagnosis was defined as diagnosis at the first visit in computed tomography (CT)-capable facilities or referral to an appropriate medical institution immediately following the first visit to a non-CT-capable facility. Cases not meeting this definition were defined as diagnostic errors that resulted in delayed diagnosis. Of the 439 cases of lower gastrointestinal perforation identified, delayed diagnosis occurred in 138 cases (31.4%). Multivariate logistic regression analysis revealed a significant association between examination by a non-generalist and delayed diagnosis. Other factors showing a tendency with delayed diagnosis included presence of fever, absence of abdominal tenderness, and unavailability of urgent radiology reports. Initial misdiagnoses were mainly gastroenteritis, constipation, and small bowel obstruction. In conclusion, diagnostic errors occurred in about one-third of patients with a lower gastrointestinal perforation.

Pharmacists can improve diagnosis and help prevent diagnostic errors.

We present two cases that highlight the role of pharmacists in the diagnostic process and illustrate how a culture of safety and teamwork between pharmacists and physicians can help prevent diagnostic errors.

COVID Blindness: Delayed Diagnosis of Aseptic Meningitis in the COVID-19 Era.

Diagnostic errors are a serious problem in healthcare. The diagnostic process is highly susceptible to cognitive bias and the current COVID-19 pandemic may cause normally accurate healthcare workers to make incorrect decisions. We report a case of aseptic meningitis that required five healthcare visits before it was correctly diagnosed. This case highlights the risk of anchoring bias and the importance of carefully assessing diagnostic processes during the COVID-19 pandemic. LEARNING POINTS: COVID-19 can disrupt the healthcare system and clinical environment and affect diagnosis due to anchoring bias.Healthcare providers should carefully check the COVID-19 illness script to reduce diagnostic errors.Healthcare providers should prepare and practice a diagnostic debiasing strategy during the COVID-19 pandemic.

Multimorbidity and patient-reported diagnostic errors in the primary care setting: multicentre cross-sectional study in Japan.

OBJECTIVES: There is lack of evidence for the association between multimorbidity and diagnostic errors. Information on diagnostic errors from patients' perspectives is crucial to improve the diagnostic process. In this study, we aimed to investigate patient-reported diagnostic errors and to examine the relationship between multimorbidity and patient-reported diagnostic errors in the primary care setting. DESIGN: Multicentre cross-sectional study. SETTING: A primary care practice-based research network in Japan (25 primary care facilities). PARTICIPANTS: Adult outpatients filled out a standardised questionnaire. PRIMARY OUTCOME MEASURE: Patient-reported diagnostic errors. RESULTS: Data collected from 1474 primary care outpatients were analysed. The number of participants who reported diagnostic errors was 57 (3.9%). Most of the missed diagnoses were common conditions in primary care, such as cancer, dermatitis and hypertension. After adjustment for possible confounders and clustering within facilities, multimorbidity was positively associated with patient-reported diagnostic errors (adjusted OR=1.83, 95% CI 1.01 to 3.31). The results of the sensitivity analysis were consistent with those of the primary analysis. CONCLUSIONS: The present study showed a lower proportion of patients reporting experiences of diagnostic errors in primary care than those reported in previous studies in other countries. However, patients with multimorbidity are more likely to report diagnostic errors in primary care; thus, further research is necessary to improve the diagnostic process for patients with multimorbidity.

A Perspective from a Case Conference on Comparing the Diagnostic Process: Human Diagnostic Thinking vs. Artificial Intelligence (AI) Decision Support Tools.

Artificial intelligence (AI) has made great contributions to the healthcare industry. However, its effect on medical diagnosis has not been well explored. Here, we examined a trial comparing the thinking process between a computer and a master in diagnosis at a clinical conference in Japan, with a focus on general diagnosis. Consequently, not only was AI unable to exhibit its thinking process, it also failed to include the final diagnosis. The following issues were highlighted: (1) input information to AI could not be weighted in order of importance for diagnosis; (2) AI could not deal with comorbidities (see Hickam's dictum); (3) AI was unable to consider the timeline of the illness (depending on the tool); (4) AI was unable to consider patient context; (5) AI could not obtain input information by themselves. This comparison of the thinking process uncovered a future perspective on the use of diagnostic support tools.

Pneumococcal perihepatitis due to hematogeneous dissemination.

Perihepatitis is mainly caused by a direct extension of pelvic inflammatory disease, in which the causative pathogen is typically Neisseria gonorrhoeae or Chlamydia trachomatis. We herein discuss the case of a 61-year-old female patient who presented with a fever and right upper quadrant pain. Perihepatitis was diagnosed by contrast-enhanced computed tomography. She had no previous history of sexual activity, genital symptoms, remarkable physical findings or examination results indicative of pelvic inflammatory disease or other diseases. A blood culture detected Streptococcus pneumoniae, leading to the suspicion of hematogeneous dissemination. The patient was therefore treated with the appropriate antimicrobials. While invasive pneumococcal disease mainly results in bacteremic pneumonia, meningitis or endocarditis, the present case showed that it can also lead to perihepatitis; a blood culture is therefore useful for clarifying the infection route and pathogens in perihepatitis if the patient has no past history of sexual activity, genital symptoms or physical or other findings indicative of pelvic inflammatory disease.

Disease vocabulary size as a surrogate marker for physicians' disease knowledge volume.

OBJECTIVE: Recognizing what physicians know and do not know about a particular disease is one of the keys to designing clinical decision support systems, since these systems can fulfill complementary role by recognizing this boundary. To our knowledge, however, no study has attempted to quantify how many diseases physicians actually know and thus the boundary is unclear. This study explores a method to solve this problem by investigating whether the vocabulary assessment techniques developed in the linguistics field can be applied to assess physicians' knowledge. METHODS: The test design required us to pay special attention to disease knowledge assessment. First, to avoid imposing unnecessary burdens on the physicians, we chose a self-assessment questionnaire that was straightforward to fill out. Second, to prevent overestimation, we used a "pseudo-word" approach: fictitious diseases were included in the questionnaire, and positive responses to them were penalized. Third, we used paper-based tests, rather than computer-based ones, to further prevent participants from cheating by using a search engine. Fourth, we selectively used borderline diseases, i.e., diseases that physicians might or might not know about, rather than well-known or little-known diseases, in the questionnaire. RESULTS: We collected 102 valid answers from 109 physicians who attended the seminars we conducted. On the basis of these answers, we estimated that the average physician knew of 2008 diseases (95% confidence interval: (1939, 2071)). This preliminary estimation agrees with the guideline for the national license examination in Japan, suggesting that this vocabulary assessment was able to evaluate physicians' knowledge. The survey included physicians with various backgrounds, but there were no significant differences between subgroups. Other implication for researches on clinical decision support and limitation of the sampling method adopted in this study are also discussed, toward more rigorous estimation in future surveys.

The Tokyo GIM Conference: Clinical reasoning conference from real cases.

The view of Tokyo GIM Conference.

Special report of the first Australasian Diagnostic Error in Medicine Conference from 23 to 25 May, 2017.

We attended the first Australasian Diagnostic Error in Medicine Conference (Aus DEM) in Melbourne from 23 to 25 May, 2017. We believe hosting the DEM conference in Japan is vital in promoting diagnostic error prevention initiatives in our region. We hope all the stakeholders in health care will join the DEM conference to be hosted in our country to address issues surrounding diagnostic errors.