RESUMO
Background: Patients are increasingly using artificial intelligence (AI) chatbots to seek answers to medical queries. Methods: Ten frequently asked questions in anaesthesia were posed to three AI chatbots: ChatGPT4 (OpenAI), Bard (Google), and Bing Chat (Microsoft). Each chatbot's answers were evaluated in a randomised, blinded order by five residency programme directors from 15 medical institutions in the USA. Three medical content quality categories (accuracy, comprehensiveness, safety) and three communication quality categories (understandability, empathy/respect, and ethics) were scored between 1 and 5 (1 representing worst, 5 representing best). Results: ChatGPT4 and Bard outperformed Bing Chat (median [inter-quartile range] scores: 4 [3-4], 4 [3-4], and 3 [2-4], respectively; P<0.001 with all metrics combined). All AI chatbots performed poorly in accuracy (score of ≥4 by 58%, 48%, and 36% of experts for ChatGPT4, Bard, and Bing Chat, respectively), comprehensiveness (score ≥4 by 42%, 30%, and 12% of experts for ChatGPT4, Bard, and Bing Chat, respectively), and safety (score ≥4 by 50%, 40%, and 28% of experts for ChatGPT4, Bard, and Bing Chat, respectively). Notably, answers from ChatGPT4, Bard, and Bing Chat differed statistically in comprehensiveness (ChatGPT4, 3 [2-4] vs Bing Chat, 2 [2-3], P<0.001; and Bard 3 [2-4] vs Bing Chat, 2 [2-3], P=0.002). All large language model chatbots performed well with no statistical difference for understandability (P=0.24), empathy (P=0.032), and ethics (P=0.465). Conclusions: In answering anaesthesia patient frequently asked questions, the chatbots perform well on communication metrics but are suboptimal for medical content metrics. Overall, ChatGPT4 and Bard were comparable to each other, both outperforming Bing Chat.
RESUMO
Objectives: This study compared hospital readmission and mortality for patients with sepsis who received ceftaroline or daptomycin as first-line MRSA therapy.Methods: This retrospective comparative-effectiveness study included adults ≥18 years old hospitalized in the United States Veterans Health Care System with sepsis between 10/1/2010-9/30/2014, who received ceftaroline or daptomycin within 14 days of hospital admission as the first antibiotic effective against methicillin resistant Staphylococcus aureus (MRSA). Patients with pneumonia, and those who received both study drugs, were excluded. Baseline characteristics were compared using Chi-square, Fischer's exact, Student's t, and Wilcoxon Rank Sum tests. Patient outcomes were compared with multivariable logistic regression models.Results: 409 patients were included (ceftaroline = 67, daptomycin = 342). Ceftaroline patients were older, less likely to be Black, more likely to have diabetes with complications, and had higher Charlson comorbidity scores. Median (interquartile range) time from admission to drug initiation was 1 (0-1) day for ceftaroline and 1 (1-3) day for daptomycin (p = 0.01). Unadjusted hospital readmission rates for ceftaroline and daptomycin, respectively, were: 30-day (25%/37%, p = 0.06), 60-day (27%/44%, p = 0.008), and 90-day (28%/46%, p = 0.01). Unadjusted mortality rates were: in-hospital (7%/12%, p = 0.4), 30-day (3%/9%, p = 0.1), 60-day (6%/12%, p = 0.2), and 90-day (7%/15%, p = 0.1). In multivariable models with all divergent baseline characteristics included as covariates, patients treated with ceftaroline were less likely to experience (OR, 95% CI): 30/60/90-day hospital readmission (0.54, 0.29-0.98; 0.42, 0.23-0.76; 0.42, 0.23-0.75) and 30/60/90-day mortality (0.23, 0.04-0.82; 0.34, 0.10-0.93; 0.34, 0.11-0.86).Conclusion: In patients with sepsis, ceftaroline was associated with fewer hospital readmissions and lower mortality as compared to daptomycin. Prospective investigations in larger, more generalized cohorts are needed to examine outcomes with specific MRSA therapies.