The introduction of large language models (LLMs) into clinical practice promises to improve patient education and empowerment, thereby personalizing medical care and broadening access to medical knowledge. Despite the popularity of LLMs, there is a significant gap in systematized information on their use in patient care. Therefore, this systematic review aims to synthesize current applications and limitations of LLMs in patient care using a data-driven convergent synthesis approach. We searched 5 databases for qualitative, quantitative, and mixed methods articles on LLMs in patient care published between 2022 and 2023. From 4,349 initial records, 89 studies across 29 medical specialties were included, primarily examining models based on the GPT-3.5 (53.2%, n=66 of 124 different LLMs examined per study) and GPT-4 (26.6%, n=33/124) architectures in medical question answering, followed by patient information generation, including medical text summarization or translation, and clinical documentation. Our analysis delineates two primary domains of LLM limitations: design and output. Design limitations included 6 second-order and 12 third-order codes, such as lack of medical domain optimization, data transparency, and accessibility issues, while output limitations included 9 second-order and 32 third-order codes, for example, non-reproducibility, non-comprehensiveness, incorrectness, unsafety, and bias. In conclusion, this study is the first review to systematically map LLM applications and limitations in patient care, providing a foundational framework and taxonomy for their implementation and evaluation in healthcare settings.

Background/Objectives: The aim of this study was to investigate the prognostic value of established response assessment tools for hepatocellular carcinoma (HCC) treated with high-dose-rate interstitial brachytherapy (iBT) alone or with transarterial chemoembolization (cTACE). Methods: (Non-)responders were categorized using size-based RECIST 1.1 and WHO criteria, enhancement-based mRECIST and EASL criteria, and the LI-RADS Treatment Response Algorithm (LR-TRA). The outcomes were the overall survival (OS), progression-free survival (PFS), and time to progression (TTP). The statistics used included Fisher’s exact test, a t-test, the Mann–Whitney-U test, and a Kaplan–Meier analysis. The median OS, PFS, and TTP were higher in patients following iBT (26.3, 9.1, and 13.0 months) than following cTACE/iBT (23.3, 7.6, and 9.2 months). Results: The enhancement-based criteria identified more responders and predicted PFS and TTP better compared to the size-based criteria. At two months, the cTACE/iBT responders showed improved PFS (mRECIST and EASL: 11.3 vs. 2.3 and 11.0 vs. 2.3, p < 0.01) and TTP (mRECIST and EASL: 11.9 vs. 2.4 months, p < 0.01) by the enhancement-based criteria. An EASL assessment at five months predicted improved survival following both cTACE/iBT (PFS: 11.9 vs. 5.1 months, p = 0.03; TTP: 12.4 vs. 5.0, p < 0.01) and iBT (11.1 vs. 5.1 months, p = 0.04; 13.0 vs. 5.3, p < 0.01). The LR-TRA showed OS benefits at five months for cTACE/iBT responders. Size-based criteria were not prognostic. Conclusions: Extending follow-up post-iBT or post-iBT/cTACE may improve responder stratification and prognostication.

Climate change mitigation depends on actions that affect the public interest and lead to widespread changes in public attitudes and behavior. With the global outbreak of the COVID-19 pandemic, humanity faced a more imminent threat to its well-being and viability. This retrospective cross-sectional study examines how public interest in climate change was attenuated by the severity of the COVID-19 pandemic using Google Trends Search Volume Index (SVI), weather, and climate data on a United States state-level basis during the first two years of the pandemic from 2020 to 2022. To identify channels through which the COVID-19 pandemic affected information demand on climate change, a novel fixed effect regression model of public climate change interest was developed. The measure captures changes in the climate change SVI independent of weather and climate conditions, comprising pandemic-related changes in living circumstances such as COVID-19-related cases and deaths, mask mandates, and the proportion of the vaccinated population. Our results indicate that public interest in climate change was systematically attenuated by the severity of the COVID-19 pandemic. In addition, this study provides an approach for identifying drivers of public interest in climate change.