Background: Owing to the rapid growth in the popularity of Large Language Models (LLMs), various performance evaluation studies have been conducted to confirm their applicability in the medical field. However, there is still no clear framework for evaluating LLMs.Objective: This study reviews studies on LLM evaluations in the medical field and analyzes the research methods used in these studies. It aims to provide a reference for future researchers designing LLM studies.Methods & materials: We conducted a scoping review of three databases (PubMed, Embase, and MEDLINE) to identify LLM-related articles published between January 1, 2023, and September 30, 2023. We analyzed the types of methods, number of questions (queries), evaluators, repeat measurements, additional analysis methods, use of prompt engineering, and metrics other than accuracy.Results: A total of 142 articles met the inclusion criteria. LLM evaluation was primarily categorized as either providing test examinations (n = 53, 37.3%) or being evaluated by a medical professional (n = 80, 56.3%), with some hybrid cases (n = 5, 3.5%) or a combination of the two (n = 4, 2.8%). Most studies had 100 or fewer questions (n = 18, 29.0%), 15 (24.2%) performed repeated measurements, 18 (29.0%) performed additional analyses, and 8 (12.9%) used prompt engineering. For medical assessment, most studies used 50 or fewer queries (n = 54, 64.3%), had two evaluators (n = 43, 48.3%), and 14 (14.7%) used prompt engineering.Conclusions: More research is required regarding the application of LLMs in healthcare. Although previous studies have evaluated performance, future studies will likely focus on improving performance. A well-structured methodology is required for these studies to be conducted systematically.

To discover the association between eating alone and diet quality among Korean adults who eat alone measured by the mean adequacy ratio (MAR), The cross-sectional study in diet quality which was measured by nutrient intakes, indicated as MAR and nutrient adequacy ratio (NAR) with the Korean National Health and Nutrition Examination Survey (KNHANES) VI 2013–2015 data. Study population was 8523 Korean adults. Multiple linear regression was performed to identify the association between eating behaviour and MAR and further study analysed how socioeconomic factors influence the diet quality of those who eat alone. We found that the diet quality of people who eat alone was lower than that of people who eat together in both male (β: − 0.110, p = 0.002) and female participants (β: − 0.069, p = 0.005). Among who eats alone, the socioeconomic factors that negatively influenced MAR with the living arrangement, education level, income levels, and various occupation classifications. People who eat alone have nutrition intake below the recommended amount. This could lead to serious health problems not only to those who are socially disadvantaged but also those who are in a higher social stratum. Policy-makers should develop strategies to enhance diet quality to prevent potential risk factors.

The LACE index accounts for: Length of stay (L), Acuity of admission (A), Comorbidities (C), and recent Emergency department use (E). This study aimed to explore the LACE index to predict the high risk of 30-day readmission in patients with diverse disease conditions by an updated systematic review. A systematic review carried out by electronic databases from 2011–2021. The studies included a LACE index score for 30-day of readmission and patients with all types of diseases and were published in the English language. The meta-analysis was performed by using a random-effects model with a 95% confidence interval. Of 3300 records, a total of 16 studies met the inclusion criteria. The country of publication was primarily the USA (n = 7) and study designs were retrospective and perspective cohorts. The average mean age was 64 years. The C-statistics was 0.55 to 0.81. The pooled random effects of relative risk readmission were overall (RR, 0.20; 95% CI, 0.12–0.34) and it was favorable. The subgroup analysis of the opted disease-based relative risk of readmissions of all causes, cardiovascular and pulmonary diseases, and neurological diseases were consistent and statistically significant at p < 0.001 level. Current evidence of this review suggested that incorporating a high-risk LACE index showed favorable to risk prediction and could be applied to predict 30-day readmission with chronic conditions. Future study would be planned to predict the high risk of 30-day readmission in acute clinical care for utility, and applicability of promising LACE index in South Korean hospitals.

Aging is characterized by deterioration of physiologic function and a decline in intrinsic capacity to tolerate stressors, thereby increasing vulnerability to disease, disability, and death. The World Health Organization calls for integrated care for older people on the basis of intrinsic capacity.¹ Measuring intrinsic capacity from a clinical assessment may give opportunities to identify people at risk for accelerated aging and test interventions to slow aging.^2,3 A deficit-accumulation frailty index, which quantifies the burden of age-related, clinically detectable health deficits,⁴ has been proposed as a measure of vulnerability and aging. Previous research^5-10 has shown that a frailty index could estimate death, disability, and health care cost, as well as poor outcomes after acute illnesses and stressful treatments, in older populations with a wide age spectrum. However, it remains uncertain whether a frailty index measured at the same chronologic age at 66 years of age, when most people are free of major illnesses and disability, can identify those at risk for development of a range of age-related chronic diseases and disability. Answering this question may lend support to the frailty index as an integrative measure of aging. Additionally, it will help early identification of people who may benefit from clinical and public health interventions to promote healthy aging.^11-15