Significance Statement: Early reports have indicated that AKI and other kidney abnormalities are associated with coronavirus disease 2019 (COVID-19). Of 3993 hospitalized patients with COVID-19 in a New York City health system, AKI occurred in 1835 (46%) patients; among patients with AKI, 19% required dialysis, and half of them died in the hospital. Among patients who were discharged, 35% had not recovered to baseline kidney function at the time of discharge. AKI is common among patients with COVID-19 and is associated with higher mortality than in patients without AKI; among those who survive, only about a third are discharged with renal recovery. These findings may help centers with resource planning and preparing for the increased load resulting from survivors of COVID-19–associated AKI who do not experience recovery of kidney function. Background: Early reports indicate that AKI is common among patients with coronavirus disease 2019 (COVID-19) and associated with worse outcomes. However, AKI among hospitalized patients with COVID-19 in the United States is not well described. Methods: This retrospective, observational study involved a review of data from electronic health records of patients aged ≥18 years with laboratory-confirmed COVID-19 admitted to the Mount Sinai Health System from February 27 to May 30, 2020. We describe the frequency of AKI and dialysis requirement, AKI recovery, and adjusted odds ratios (aORs) with mortality. Results: Of 3993 hospitalized patients with COVID-19, AKI occurred in 1835 (46%) patients; 347 (19%) of the patients with AKI required dialysis. The proportions with stages 1, 2, or 3 AKI were 39%, 19%, and 42%, respectively. A total of 976 (24%) patients were admitted to intensive care, and 745 (76%) experienced AKI. Of the 435 patients with AKI and urine studies, 84% had proteinuria, 81% had hematuria, and 60% had leukocyturia. Independent predictors of severe AKI were CKD, men, and higher serum potassium at admission. In-hospital mortality was 50% among patients with AKI versus 8% among those without AKI (aOR, 9.2; 95% confidence interval, 7.5 to 11.3). Of survivors with AKI who were discharged, 35% had not recovered to baseline kidney function by the time of discharge. An additional 28 of 77 (36%) patients who had not recovered kidney function at discharge did so on posthospital follow-up. Conclusions: AKI is common among patients hospitalized with COVID-19 and is associated with high mortality. Of all patients with AKI, only 30% survived with recovery of kidney function by the time of discharge.

Polycystic ovary syndrome (PCOS) is a common, complex genetic disorder affecting up to 15% of reproductive-age women worldwide, depending on the diagnostic criteria applied. These diagnostic criteria are based on expert opinion and have been the subject of considerable controversy. The phenotypic variation observed in PCOS is suggestive of an underlying genetic heterogeneity, but a recent meta-analysis of European ancestry PCOS cases found that the genetic architecture of PCOS defined by different diagnostic criteria was generally similar, suggesting that the criteria do not identify biologically distinct disease subtypes. We performed this study to test the hypothesis that there are biologically relevant subtypes of PCOS. Using biochemical and genotype data from a previously published PCOS genome-wide association study (GWAS), we investigated whether there were reproducible phenotypic subtypes of PCOS with subtype-specific genetic associations. Unsupervised hierarchical cluster analysis was performed on quantitative anthropometric, reproductive, and metabolic traits in a genotyped cohort of 893 PCOS cases (median and interquartile range [IQR]: age = 28 [25–32], body mass index [BMI] = 35.4 [28.2–41.5]). The clusters were replicated in an independent, ungenotyped cohort of 263 PCOS cases (median and IQR: age = 28 [24–33], BMI = 35.7 [28.4–42.3]). The clustering revealed 2 distinct PCOS subtypes: a “reproductive” group (21%–23%), characterized by higher luteinizing hormone (LH) and sex hormone binding globulin (SHBG) levels with relatively low BMI and insulin levels, and a “metabolic” group (37%–39%), characterized by higher BMI, glucose, and insulin levels with lower SHBG and LH levels. We performed a GWAS on the genotyped cohort, limiting the cases to either the reproductive or metabolic subtypes. We identified alleles in 4 loci that were associated with the reproductive subtype at genome-wide significance (PRDM2/KAZN, P = 2.2 × 10⁻¹⁰; IQCA1, P = 2.8 × 10⁻⁹; BMPR1B/UNC5C, P = 9.7 × 10⁻⁹; CDH10, P = 1.2 × 10⁻⁸) and one locus that was significantly associated with the metabolic subtype (KCNH7/FIGN, P = 1.0 × 10⁻⁸). We developed a predictive model to classify a separate, family-based cohort of 73 women with PCOS (median and IQR: age = 28 [25–33], BMI = 34.3 [27.8–42.3]) and found that the subtypes tended to cluster in families and that carriers of previously reported rare variants in DENND1A, a gene that regulates androgen biosynthesis, were significantly more likely to have the reproductive subtype of PCOS. Limitations of our study were that only PCOS cases of European ancestry diagnosed by National Institutes of Health (NIH) criteria were included, the sample sizes for the subtype GWAS were small, and the GWAS findings were not replicated. In conclusion, we have found reproducible reproductive and metabolic subtypes of PCOS. Furthermore, these subtypes were associated with novel, to our knowledge, susceptibility loci. Our results suggest that these subtypes are biologically relevant because they appear to have distinct genetic architecture. This study demonstrates how phenotypic subtyping can be used to gain additional insights from GWAS data.

Background: Empathy, a fundamental aspect of human interaction, is characterized as the ability to experience another being’s emotions within oneself. In health care, empathy is a fundamental for health care professionals and patients’ interaction. It is a unique quality to humans that large language models (LLMs) are believed to lack. Objective: We aimed to review the literature on the capacity of LLMs in demonstrating empathy. Methods: We conducted a literature search on MEDLINE, Google Scholar, PsyArXiv, medRxiv, and arXiv between December 2022 and February 2024. We included English-language full-length publications that evaluated empathy in LLMs’ outputs. We excluded papers evaluating other topics related to emotional intelligence that were not specifically empathy. The included studies’ results, including the LLMs used, performance in empathy tasks, and limitations of the models, along with studies’ metadata were summarized. Results: A total of 12 studies published in 2023 met the inclusion criteria. ChatGPT-3.5 (OpenAI) was evaluated in all studies, with 6 studies comparing it with other LLMs such GPT-4, LLaMA (Meta), and fine-tuned chatbots. Seven studies focused on empathy within a medical context. The studies reported LLMs to exhibit elements of empathy, including emotions recognition and emotional support in diverse contexts. Evaluation metric included automatic metrics such as Recall-Oriented Understudy for Gisting Evaluation and Bilingual Evaluation Understudy, and human subjective evaluation. Some studies compared performance on empathy with humans, while others compared between different models. In some cases, LLMs were observed to outperform humans in empathy-related tasks. For example, ChatGPT-3.5 was evaluated for its responses to patients’ questions from social media, where ChatGPT’s responses were preferred over those of humans in 78.6% of cases. Other studies used subjective readers’ assigned scores. One study reported a mean empathy score of 1.84-1.9 (scale 0-2) for their fine-tuned LLM, while a different study evaluating ChatGPT-based chatbots reported a mean human rating of 3.43 out of 4 for empathetic responses. Other evaluations were based on the level of the emotional awareness scale, which was reported to be higher for ChatGPT-3.5 than for humans. Another study evaluated ChatGPT and GPT-4 on soft-skills questions in the United States Medical Licensing Examination, where GPT-4 answered 90% of questions correctly. Limitations were noted, including repetitive use of empathic phrases, difficulty following initial instructions, overly lengthy responses, sensitivity to prompts, and overall subjective evaluation metrics influenced by the evaluator’s background. Conclusions: LLMs exhibit elements of cognitive empathy, recognizing emotions and providing emotionally supportive responses in various contexts. Since social skills are an integral part of intelligence, these advancements bring LLMs closer to human-like interactions and expand their potential use in applications requiring emotional intelligence. However, there remains room for improvement in both the performance of these models and the evaluation strategies used for assessing soft skills.

Drug repurposing—identifying new therapeutic uses for approved drugs—is often a serendipitous and opportunistic endeavour to expand the use of drugs for new diseases. The clinical utility of drug-repurposing artificial intelligence (AI) models remains limited because these models focus narrowly on diseases for which some drugs already exist. Here we introduce TxGNN, a graph foundation model for zero-shot drug repurposing, identifying therapeutic candidates even for diseases with limited treatment options or no existing drugs. Trained on a medical knowledge graph, TxGNN uses a graph neural network and metric learning module to rank drugs as potential indications and contraindications for 17,080 diseases. When benchmarked against 8 methods, TxGNN improves prediction accuracy for indications by 49.2% and contraindications by 35.1% under stringent zero-shot evaluation. To facilitate model interpretation, TxGNN’s Explainer module offers transparent insights into multi-hop medical knowledge paths that form TxGNN’s predictive rationales. Human evaluation of TxGNN’s Explainer showed that TxGNN’s predictions and explanations perform encouragingly on multiple axes of performance beyond accuracy. Many of TxGNN’s new predictions align well with off-label prescriptions that clinicians previously made in a large healthcare system. TxGNN’s drug-repurposing predictions are accurate, consistent with off-label drug use, and can be investigated by human experts through multi-hop interpretable rationales.

Background Large language models (LLMs) show promise in clinical contexts but can generate false facts (often referred to as “hallucinations”). One subset of these errors arises from adversarial attacks, in which fabricated details embedded in prompts lead the model to produce or elaborate on the false information. We embedded fabricated content in clinical prompts to elicit adversarial hallucination attacks in multiple large language models. We quantified how often they elaborated on false details and tested whether a specialized mitigation prompt or altered temperature settings reduced errors. Methods We created 300 physician-validated simulated vignettes, each containing one fabricated detail (a laboratory test, a physical or radiological sign, or a medical condition). Each vignette was presented in short and long versions—differing only in word count but identical in medical content. We tested six LLMs under three conditions: default (standard settings), mitigating prompt (designed to reduce hallucinations), and temperature 0 (deterministic output with maximum response certainty), generating 5,400 outputs. If a model elaborated on the fabricated detail, the case was classified as a “hallucination”. Results Hallucination rates range from 50 % to 82 % across models and prompting methods. Prompt-based mitigation lowers the overall hallucination rate (mean across all models) from 66 % to 44 % (p < 0.001). For the best-performing model, GPT-4o, rates decline from 53 % to 23 % (p < 0.001). Temperature adjustments offer no significant improvement. Short vignettes show slightly higher odds of hallucination. Conclusions LLMs are highly susceptible to adversarial hallucination attacks, frequently generating false clinical details that pose risks when used without safeguards. While prompt engineering reduces errors, it does not eliminate them.