Patient-Friendly Discharge Summaries in Korea Based on ChatGPT: Software Development and Validation.

BACKGROUND: Although discharge summaries in patient-friendly language can enhance patient comprehension and satisfaction, they can also increase medical staff workload. Using a large language model, we developed and validated software that generates a patient-friendly discharge summary. METHODS: We developed and tested the software using 100 discharge summary documents, 50 for patients with myocardial infarction and 50 for patients treated in the Department of General Surgery. For each document, three new summaries were generated using three different prompting methods (Zero-shot, One-shot, and Few-shot) and graded using a 5-point Likert Scale regarding factuality, comprehensiveness, usability, ease, and fluency. We compared the effects of different prompting methods and assessed the relationship between input length and output quality. RESULTS: The mean overall scores differed across prompting methods (4.19 ± 0.36 in Few-shot, 4.11 ± 0.36 in One-shot, and 3.73 ± 0.44 in Zero-shot; P < 0.001). Post-hoc analysis indicated that the scores were higher with Few-shot and One-shot prompts than in zero-shot prompts, whereas there was no significant difference between Few-shot and One-shot prompts. The overall proportion of outputs that scored ≥ 4 was 77.0% (95% confidence interval: 68.8-85.3%), 70.0% (95% confidence interval [CI], 61.0-79.0%), and 32.0% (95% CI, 22.9-41.1%) with Few-shot, One-shot, and Zero-shot prompts, respectively. The mean factuality score was 4.19 ± 0.60 with Few-shot, 4.20 ± 0.55 with One-shot, and 3.82 ± 0.57 with Zero-shot prompts. Input length and the overall score showed negative correlations in the Zero-shot (r = -0.437, P < 0.001) and One-shot (r = -0.327, P < 0.001) tests but not in the Few-shot (r = -0.050, P = 0.625) tests. CONCLUSION: Large-language models utilizing Few-shot prompts generally produce acceptable discharge summaries without significant misinformation. Our research highlights the potential of such models in creating patient-friendly discharge summaries for Korean patients to support patient-centered care.

The Effect of Daytime Surgical Hospitalists on Reducing Night Shift Physicians' Workload.

BACKGROUND: The shortage of doctors at night makes it difficult to execute complex tasks and deliver accurate decisions. Therefore, reducing the night shift physician's workload is essential for patient safety. This study aimed to observe the effect of daytime surgical hospitalists on the reduction in night shift physicians' workload by analyzing the volume of electronic orders generated at night for postoperative patients. METHODS: A total of 9328 hospitalized patients who underwent colorectal or gastrointestinal surgery for more than 120 minutes were reviewed retrospectively. This study reviewed the nighttime volume difference of electronic orders between patients cared for by a daytime surgical hospitalist and patients cared for by a resident. Multiple logistic regression analysis was performed to analyze the risk factors of nighttime orders during hospitalization (dichotomous end point). Negative binomial regression analysis was also performed to analyze the volume of electronic orders as countable data, and the incident rate ratio was estimated (count end point). RESULTS: The risk of the occurrence of nighttime electronic order for patients treated by a surgical hospitalist was lower than that of patients cared for by a resident (adjusted odds ratio, 0.616; 95% confidence interval, 0.558-0.682; P < 0.001). In the negative binomial regression analysis, the total volume of electronic orders at night was lower in patients cared for by a surgical hospitalist than in patients cared for by a resident (adjusted incident rate ratio, 0.653; 95% confidence interval, 0.623-0.685; P < 0.001). CONCLUSIONS: The introduction of daytime surgical hospitalists is associated with the lower workload of night shift physicians.

A surgical hospitalist system in Korea: a preliminary study of the effects on hospital costs and postoperative outcomes.

PURPOSE: The aim of this study is to investigate the effect of the surgical hospitalist system on postoperative outcomes and hospital costs for surgical patients. METHODS: We reviewed the medical records of 522 patients who were admitted to the divisions of colorectal and gastrointestinal surgery for operation from September to December 2017 at Severance Hospital, Yonsei University College of Medicine in Seoul, Korea. All patients were divided into 2 groups; one that was managed by surgical hospitalists group (HG) and another that was managed by non-hospitalist residents group (NHG) after elective surgery. Postoperative outcomes and hospital costs were analyzed for each group. RESULTS: Two hundred ninety-eight patients were managed by HG and 189 patients were managed by NHG after surgery. The length of hospital stay in the first group was shorter (9.6 ± 5.8 days vs. 12.2 ± 7.9 days, P < 0.001), the incidence of complications was lower (44.6% vs. 55.6%, P = 0.019), and the readmission rate was lower (3.0% vs. 6.9%, P = 0.046) in the HG than in the NHG. The difference in total hospital costs was not significant between the HG and the NHG (₩8,381,304 vs. ₩9,242,493, P = 0.559), but surgery-independent hospital costs were lower in the HG than in the NHG (₩3,020,873 vs. ₩3,923,308, P = 0.001). CONCLUSION: The surgical hospitalist system reduced the length of hospital stay, the incidence of postoperative complications, and the readmission rates of surgical patients. This led to the effect of a reduction in total hospital costs.

Clinical and survival outcomes after hepatectomy in patients with non-alcoholic fatty liver and hepatitis B-related hepatocellular carcinoma.

BACKGROUND: The prevalence of non-alcoholic fatty liver disease-related hepatocellular carcinoma (NAFLD-HCC) has increased parallelly with that of metabolic syndrome. This study aimed to compare the clinical and survival outcomes of NAFLD-HCC and HBV-related HCC(HBV-HCC). METHODS: The medical records of patients who underwent hepatectomy for HCC at Severance Hospital between 2005 and 2015 were retrospectively reviewed. Occult HBV infection was identified by nested PCR. Propensity score matching (PSM) was conducted to minimize lead-time bias caused by the lack of surveillance in NAFLD patients. Surgical and oncologic outcomes were compared between the two groups. RESULTS: There were 32 patients (7%) with NAFLD-HCC, 200 (46%) with HBV-HCC, and 194 (44%) with HBV/NAFLD-HCC (HBV and NAFLD). Before PSM, cirrhosis was more frequently detected in HBV-HCC patients (55% vs 15%, p < 0.001) and the average tumor size was larger in the NAFLD-HCC group than in the HBV-HCC group (4.4 ± 3.3 cm vs 3.4 ± 1.8 cm, p = 0.014). After a median follow-up of 74 months (range 0-157 months), survival analyses before PSM showed better 5-year overall survival (OS) in HBV-HCC patients than in NAFLD-HCC patients (80% vs 63%, p = 0.041). After PSM, 5-year OS rates were similar (60% vs 63%, p = 0.978). There were no differences between the groups in recurrence-free or disease-specific survival before and after PSM. CONCLUSION: Patients with NAFLD-HCC were less likely to have underlying cirrhosis but more likely to have larger tumors at the time of diagnosis than patients with HBV-HCC. The OS of patients with NAFLD-HCC appeared to be worse than that of patients with HBV-HCC. Therefore, active HCC surveillance is recommended in patients with metabolic syndrome for the early detection of HCC.

Right anterior section graft for living-donor liver transplantation: A case report.

RATIONALE: In living-donor liver transplantation (LDLT), the right lobe graft is commonly utilized to prevent small-for-size syndrome, despite the considerable donor morbidity. Conversely, the feasibility of the left lobe graft and the right posterior section graft in smaller-sized recipients is now commonly employed with comparable outcomes to right lobe grafts. The efficacy of the right anterior section graft has rarely been reported. PATIENT CONCERNS: A 56-year-old man, a heavy alcoholic beverage drinker for 20 years, presented in the emergency department with massive ascites and lethargy. He was previously admitted twice due to bleeding esophageal varices. DIAGNOSIS: He was diagnosed with hepatic encephalopathy coma due to alcoholic liver cirrhosis. The Child-Turcotte-Pugh score was 11 (class C), and the Model for End-stage Liver Disease score was 21.62. INTERVENTION: A LDTL was offered to the patient as the best treatment option available. The patient's 26-year-old son was found to be the only donor-compatible candidate for the LDTL.Preoperatively, the right lobe of the donor occupied 76.2% of the total liver volume exposing the donor to a small residual liver volume. The right posterior section and left lobe volumes were insufficient, providing a graft-to-recipient weight ratio of 0.42% and 0.38%, respectively. However, the right anterior section could fulfill an acceptable GRWR of 0.83%. Thus, a living donor right anterior sectionectomy was performed. OUTCOMES: Clinical signs and symptoms and liver function improved following anterior section graft transplantation without complications. LESSON: The procurement of anterior section graft is technically feasible in selected patients, especially in high-volume liver centers.

Clinical significance of lactate clearance for the development of early allograft dysfunction and short-term prognosis in deceased donor liver transplantation.

This retrospective study evaluated lactate clearance (LC), measured at 6, 12, 18, and 24 hours after reperfusion, as a predictor of early allograft dysfunction (EAD) and short-term outcomes in patients receiving deceased donor liver transplantation. Of 181 transplant recipients, 44 (24.3%) developed EAD and had lower LCs than those who did not develop EAD. A receiver operating characteristic analysis showed that LC determined at 6 hours showed the highest area under curve value of 0.828 (95% confidence interval [CI]: 0.755-0.990) for predicting the development of EAD at a cutoff value of 25.8% with 76.7% sensitivity and 77.9% specificity. LC values that fell below the cutoff values were significantly associated with EAD in a multivariate analysis, with values at 6 hours having the highest adjusted odds ratio (11.891, 95% CI: 4.469-31.639). In-hospital and 6 month mortalities were higher in patients with LC values below the cutoffs compared with those above the cutoff values at each time point. Thus, LC calculated shortly after reperfusion of an allograft is significantly discriminative for the development of EAD and is associated with short-term prognosis after deceased donor liver transplantation.

Time to Initiation of Adjuvant Chemotherapy in Colon Cancer: Comparison of Open, Laparoscopic, and Robotic Surgery.

BACKGROUND: The time to initiation of adjuvant chemotherapy (TIC) can be used as a recovery parameter after surgery. The effect of laparoscopic or robotic surgery on TIC has not been thoroughly studied. This study aimed to compare the impact of open, laparoscopic, and robot-assisted surgery on TIC after colon cancer surgery. MATERIALS AND METHODS: Patients who underwent curative resection for stage II or III colon cancer between January 2007 and June 2013 and who received adjuvant chemotherapy from surgeons capable of performing open, laparoscopic, and robotic surgeries were included in this study. Patient demographics, clinicopathologic variables, and TIC were compared among the three groups. Univariate and multivariate analyses were performed to identify factors affecting TIC. RESULTS: Of the 252 patients, 40, 161, and 51 patients underwent open, laparoscopic, and robotic colectomy, respectively. The postoperative complication rate was lower in the laparoscopic and robotic groups compared to the open group (P = .002). The length of hospital stay was shorter in the laparoscopic group compared with the open and robotic groups (P < .001). Multivariate analysis revealed that the operation method was the only factor affecting TIC, with laparoscopic and robotic surgery being favorable (regression coefficient -5.1, 95% confidence interval -7.6 to -2.6; P < .001). However, there was no difference in TIC between the laparoscopic and robotic group. CONCLUSIONS: Laparoscopic and robotic surgeries were associated with shorter TIC. This study demonstrates another benefit of minimally invasive surgery with regard to early initiation of adjuvant chemotherapy.