Objective: To evaluate the prognostic impact of radiologic and pathologic features in patients with interstitial lung disease (ILD) other than idiopathic pulmonary fibrosis (IPF), and to identify the factors associated with the development of progressive pulmonary fibrosis (PPF) and survival. Materials and Methods: This study retrospectively enrolled 75 patients diagnosed with ILD other than IPF who underwent surgical lung biopsy between January 2004 and December 2020. Three chest radiologists independently reviewed the CT features and extent of fibrosis on preoperative and follow-up CT scans. Two pathologists reviewed the histopathological features, including the presence of interstitial pneumonia. The time to PPF and overall survival were estimated using the Kaplan-Meier method. The associations of CT and pathological features with PPF and all-cause mortality were examined using standard Cox regression and time-dependent Cox models, respectively. Results: A total of 75 non-IPF ILD patients (mean age ± standard deviations 56.4 ± 13.2 years; range, 40–88 years) were enrolled. The median follow-up duration was 75.3 months (range, 7.8–189.8 months). Traction bronchiectasis on CT (adjusted hazard ratio [HR], 6.40; P = 0.003) and body mass index (adjusted HR per 1-kg/m2 increase, 0.82; P = 0.002) were found to be significantly associated with PPF in multivariable analysis. Radiological progression (adjusted HR, 18.44;P < 0.001), symptomatic progression (adjusted HR, 4.19; P = 0.011), and age (adjusted HR for 1-year increase, 1.12; P < 0.001) were significantly associated with death. Conclusion Traction bronchiectasis on CT was a significant predictor of PPF, while radiologic and symptomatic progression and older age were associated with poorer survival in patients with ILD other than IPF. These findings indicate that careful radiological evaluation and symptom monitoring may help to predict disease progression and outcomes in patients with nonIPF ILD.

Background: The management of sexually transmitted infections (STIs) at the national level is necessary for the health and well-being of individuals in society. The purpose of this study was to propose a redesign of the STI surveillance system by collecting opinions from experts. Methods: After the major STIs were reviewed, 1) gonorrhea, 2) chancroid, 3) Mycoplasma genitalium infection, and 4) trichomoniasis were selected as diseases requiring a redesign of the surveillance system through the Delphi method, a technique for solving problems by collecting opinions from experts. The expert panel was selected from experts on STIs (n = 17) and included members of the Korean Association of Urogenital Tract Infection and Inflammation, the Korean Society of Infectious Diseases, and the 3rd advisory committee on STIs. Results: The experts agreed upon the exclusion of surveillance for chancroid on the basis of the low incidence, with an agreement rate of 83%. The experts agreed upon the inclusion of M. genitalium in the surveillance system, with a high agreement rate of 88%. However, there was disagreement among the experts on whether to include trichomoniasis in the surveillance system and whether to switch to a mandatory surveillance system for gonorrhea. Conclusion Although a consensus was not reached in some of the surveys in this study, it involved the collection of various expert opinions. Therefore, policies should be promoted with a prioritized focus on items with high agreement, but items for which no consensus was reached need to be reviewed in greater depth through additional research and investigation.

This study explored how Generation Z characteristics—specifically work–life balance, growth needs, and individualism—along with work environment, emotional labor, and interpersonal skills, affect job embeddedness among newly graduated nurses. It also examined whether job embeddedness mediates the relationship between these factors and turnover intention. Methods: A correlational descriptive study was conducted by analyzing survey data collected from 131 clinical nurses born between 1995 and 2000, who had less than three years of nursing work experience in South Korea. Hierarchical multiple regression analysis was employed to identify the independent effects of Generation Z characteristics on job embeddedness, over and above the effects of work environment, emotional labor, and interpersonal skills. Path analysis using bootstrapping method was conducted to examine the mediating role of job embeddedness in turnover intention. Results: Needs for growth and work environment were significant predictors of job embeddedness among newly graduated nurses. Furthermore, the effects of needs for growth and work environment on turnover intention were significantly mediated by job embeddedness. Conclusion: Understanding the characteristics of Generation Z, particularly their needs for growth, along with the work environment, is essential for enhancing job embeddedness among newly graduated nurses. Strengthening job embeddedness can effectively reduce their turnover intention.

Objective: To investigate whether reader training improves the performance and agreement of radiologists in interpreting unenhanced breast magnetic resonance imaging (MRI) scans using diffusion-weighted imaging (DWI). Materials and Methods: A study of 96 breasts (35 cancers, 24 benign, and 37 negative) in 48 asymptomatic women was performed between June 2019 and October 2020. High-resolution DWI with b-values of 0, 800, and 1200 sec/mm 2 was performed using a 3.0-T system. Sixteen breast radiologists independently reviewed the DWI, apparent diffusion coefficient maps, and T1-weighted MRI scans and recorded the Breast Imaging Reporting and Data System (BI-RADS) category for each breast. After a 2-h training session and a 5-month washout period, they re-evaluated the BI-RADS categories. A BI-RADS category of 4 (lesions with at least two suspicious criteria) or 5 (more than two suspicious criteria) was considered positive.The per-breast diagnostic performance of each reader was compared between the first and second reviews. Inter-reader agreement was evaluated using a multi-rater κ analysis and intraclass correlation coefficient (ICC). Results: Before training, the mean sensitivity, specificity, and accuracy of the 16 readers were 70.7% (95% confidence interval [CI]: 59.4–79.9), 90.8% (95% CI: 85.6–94.2), and 83.5% (95% CI: 78.6–87.4), respectively. After training, significant improvements in specificity (95.2%; 95% CI: 90.8–97.5; P = 0.001) and accuracy (85.9%; 95% CI: 80.9–89.8; P = 0.01) were observed, but no difference in sensitivity (69.8%; 95% CI: 58.1–79.4; P = 0.58) was observed. Regarding inter-reader agreement, the κ values were 0.57 (95% CI: 0.52–0.63) before training and 0.68 (95% CI: 0.62–0.74) after training, with a difference of 0.11 (95% CI: 0.02–0.18; P = 0.01). The ICC was 0.73 (95% CI: 0.69–0.74) before training and 0.79 (95% CI: 0.76–0.80) after training (P = 0.002). Conclusion Brief reader training improved the performance and agreement of interpretations by breast radiologists using unenhanced MRI with DWI.

Background/Aims: With the global increase in patients with solid malignancies, it is helpful to understand the outcomes of intensive care unit (ICU) admission for these patients. This study evaluated the risk factors for ICU mortality and the shortand long-term outcomes in patients with solid malignancies who had unplanned ICU admission. Methods: This retrospective cohort study included patients with solid malignancies treated at the medical ICU of a single tertiary center in South Korea between 2016 and 2022. Results: Among the 955 patients, the ICU mortality rate was 23.5%. Lung cancer was the most common cancer type (34.2%) and was significantly associated with increased ICU mortality (odd ratio [OR] 1.58, p = 0.030). Higher Sequential Organ Failure Assessment scores at ICU admission (OR 1.11, p < 0.001), the need for mechanical ventilation (OR 6.74, p < 0.001), or renal replacement therapy during the ICU stay (OR 2.49, p < 0.001) were significantly associated with higher ICU mortality. The 1-year survival rate after ICU admission was 29.3%, with a median survival of 37 days for patients requiring mechanical deviaventilation, and 23 days for patients requiring renal replacement therapy. Conclusions This study showed that critically ill patients with solid malignancies had poor 1-year survival despite relatively low ICU mortality. These findings highlight the need for careful consideration of ICU admission in patients with solid malignancy, and decision-making should be based on an understanding of the expected short- and long-term prognosis of ICU admission after an informed discussion among patients, families, and physicians.

Objectives: Sepsis is a leading global cause of mortality, and predicting its outcomes is vital for improving patient care. This study explored the capabilities of ChatGPT, a state-of-the-art natural language processing model, in predicting in-hospital mortality for sepsis patients. Methods: This study utilized data from the Korean Sepsis Alliance (KSA) database, collected between 2019 and 2021, focusing on adult intensive care unit (ICU) patients and aiming to determine whether ChatGPT could predict all-cause mortality after ICU admission at 7 and 30 days. Structured prompts enabled ChatGPT to engage in in-context learning, with the number of patient examples varying from zero to six. The predictive capabilities of ChatGPT-3.5-turbo and ChatGPT-4 were then compared against a gradient boosting model (GBM) using various performance metrics. Results: From the KSA database, 4,786 patients formed the 7-day mortality prediction dataset, of whom 718 died, and 4,025 patients formed the 30-day dataset, with 1,368 deaths. Age and clinical markers (e.g., Sequential Organ Failure Assessment score and lactic acid levels) showed significant differences between survivors and non-survivors in both datasets. For 7-day mortality predictions, the area under the receiver operating characteristic curve (AUROC) was 0.70–0.83 for GPT-4, 0.51–0.70 for GPT-3.5, and 0.79 for GBM. The AUROC for 30-day mortality was 0.51–0.59 for GPT-4, 0.47–0.57 for GPT-3.5, and 0.76 for GBM. Zero-shot predictions using GPT-4 for mortality from ICU admission to day 30 showed AUROCs from the mid-0.60s to 0.75 for GPT-4 and mainly from 0.47 to 0.63 for GPT-3.5. Conclusions GPT-4 demonstrated potential in predicting short-term in-hospital mortality, although its performance varied across different evaluation metrics.

Background: Worldwide, sepsis is the leading cause of death in hospitals. If mortality rates in patients with sepsis can be predicted early, medical resources can be allocated efficiently. We constructed machine learning (ML) models to predict the mortality of patients with sepsis in a hospital emergency department. Methods: This study prospectively collected nationwide data from an ongoing multicenter cohort of patients with sepsis identified in the emergency department. Patients were enrolled from 19 hospitals between September 2019 and December 2020. For acquired data from 3,657 survivors and 1,455 deaths, six ML models (logistic regression, support vector machine, random forest, extreme gradient boosting [XGBoost], light gradient boosting machine, and categorical boosting [CatBoost]) were constructed using fivefold cross-validation to predict mortality. Through these models, 44 clinical variables measured on the day of admission were compared with six sequential organ failure assessment (SOFA) components (PaO 2 /FIO 2 [PF], platelets (PLT), bilirubin, cardiovascular, Glasgow Coma Scale score, and creatinine).The confidence interval (CI) was obtained by performing 10,000 repeated measurements via random sampling of the test dataset. All results were explained and interpreted using Shapley’s additive explanations (SHAP). Results: Of the 5,112 participants, CatBoost exhibited the highest area under the curve (AUC) of 0.800 (95% CI, 0.756–0.840) using clinical variables. Using the SOFA components for the same patient, XGBoost exhibited the highest AUC of 0.678 (95% CI, 0.626–0.730). As interpreted by SHAP, albumin, lactate, blood urea nitrogen, and international normalization ratio were determined to significantly affect the results. Additionally, PF and PLTs in the SOFA component significantly influenced the prediction results. Conclusion Newly established ML-based models achieved good prediction of mortality in patients with sepsis. Using several clinical variables acquired at the baseline can provide more accurate results for early predictions than using SOFA components. Additionally, the impact of each variable was identified.

Background: Prolonged length of hospital stay (LOS) is associated with an increased risk of hospital-acquired conditions and worse outcomes. We conducted a nationwide, multicenter, retrospective cohort study to determine whether prolonged hospitalization before developing sepsis has a negative impact on its prognosis. Methods: We analyzed data from 19 tertiary referral or university-affiliated hospitals between September 2019 and December 2020. Adult patients with confirmed sepsis during hospitalization were included. In-hospital mortality was the primary outcome. The patients were divided into two groups according to their LOS before the diagnosis of sepsis: early- (< 5 days) and late-onset groups (≥ 5 days). Conditional multivariable logistic regression for propensity score matched-pair analysis was employed to assess the association between lateonset sepsis and the primary outcome. Results: A total of 1,395 patients were included (median age, 68.0 years; women, 36.3%).The early- and late-onset sepsis groups comprised 668 (47.9%) and 727 (52.1%) patients.Propensity score-matched analysis showed an increased risk of in-hospital mortality in the late-onset group (adjusted odds ratio [aOR], 3.00; 95% confidence interval [CI], 1.69–5.34).The same trend was observed in the entire study population (aOR, 1.85; 95% CI, 1.37–2.50).When patients were divided into LOS quartile groups, an increasing trend of mortality risk was observed in the higher quartiles (Pfor trend < 0.001). Conclusion Extended LOS before developing sepsis is associated with higher in-hospital mortality. More careful management is required when sepsis occurs in patients hospitalized for ≥ 5 days.

Mechanical power (MP) has been reported to be associated with clinical outcomes. Because the original MP equation is derived from paralyzed patients under volume-controlled ventilation, its application in practice could be limited in patients receiving pressure-controlled ventilation (PCV). Recently, a simplified equation for patients under PCV was developed. We investigated the association between MP and intensive care unit (ICU) mortality. Methods: We conducted a retrospective analysis of Korean data from the Fourth International Study of Mechanical Ventilation. We extracted data of patients under PCV on day 1 and calculated MP using the following simplified equation: MPPCV = 0.098 ∙ respiratory rate ∙ tidal volume ∙ (ΔPinsp + positive end-expiratory pressure), where ΔPinsp is the change in airway pressure during inspiration. Patients were divided into survivors and non-survivors and then compared. Multivariable logistic regression was performed to determine association between MPPCV and ICU mortality. The interaction of MPPCV and use of neuromuscular blocking agent (NMBA) was also analyzed. Results: A total of 125 patients was eligible for final analysis, of whom 38 died in the ICU. MPPCV was higher in non-survivors (17.6 vs. 26.3 J/min, P<0.001). In logistic regression analysis, only MPPCV was significantly associated with ICU mortality (odds ratio, 1.090; 95% confidence interval, 1.029–1.155; P=0.003). There was no significant effect of the interaction between MPPCV and use of NMBA on ICU mortality (P=0.579). Conclusions: MPPCV is associated with ICU mortality in patients mechanically ventilated with PCV mode, regardless of NMBA use.

Mechanical power (MP) has been reported to be associated with clinical outcomes. Because the original MP equation is derived from paralyzed patients under volume-controlled ventilation, its application in practice could be limited in patients receiving pressure-controlled ventilation (PCV). Recently, a simplified equation for patients under PCV was developed. We investigated the association between MP and intensive care unit (ICU) mortality. Methods: We conducted a retrospective analysis of Korean data from the Fourth International Study of Mechanical Ventilation. We extracted data of patients under PCV on day 1 and calculated MP using the following simplified equation: MPPCV = 0.098 ∙ respiratory rate ∙ tidal volume ∙ (ΔPinsp + positive end-expiratory pressure), where ΔPinsp is the change in airway pressure during inspiration. Patients were divided into survivors and non-survivors and then compared. Multivariable logistic regression was performed to determine association between MPPCV and ICU mortality. The interaction of MPPCV and use of neuromuscular blocking agent (NMBA) was also analyzed. Results: A total of 125 patients was eligible for final analysis, of whom 38 died in the ICU. MPPCV was higher in non-survivors (17.6 vs. 26.3 J/min, P<0.001). In logistic regression analysis, only MPPCV was significantly associated with ICU mortality (odds ratio, 1.090; 95% confidence interval, 1.029–1.155; P=0.003). There was no significant effect of the interaction between MPPCV and use of NMBA on ICU mortality (P=0.579). Conclusions: MPPCV is associated with ICU mortality in patients mechanically ventilated with PCV mode, regardless of NMBA use.