PURPOSE: This study was done to investigate the correlation among symptom experience, self-efficacy, depression, and medication adherence in patients with chronic obstructive pulmonary disease (COPD). METHODS: Participants were 100 patients with COPD recruited in one general hospital in Seoul. A structured questionnaire was used to measure the study variables. Data were analyzed using descriptive statistics, t-test, ANOVA, and Spearman correlation coefficient. RESULTS: Symptom experience and depression were positively correlated (r=.41, p < .001), symptom experience and self-efficacy were negatively correlated (r=-.21, p=.035). Depression was negatively correlated with self-efficacy (r=-.60, p < .001) and medication adherence (r=-.48, p < .001). Self-efficacy and medication adherence were positively correlated(r=.76 p < .001). CONCLUSION: Findings from this study indicate that depression and self-efficacy are important variables related to medication adherence in patients with chronic obstructive pulmonary disease. Therefore, depression and self-efficacy should be assessed, and customized nursing interventions should be provided in order to increase medication adherence.
Depression ; Hospitals, General ; Humans ; Medication Adherence ; Nursing ; Pulmonary Disease, Chronic Obstructive ; Seoul

Purpose: The aim of this study was to assess the epidemiologic and clinical features of maternal Group B Streptococcus (GBS) colonization in vagina during their third trimester. Methods: This study included 644 pregnant women who had undergone GBS culture test in their third trimester in 2018. We collected data from the primary level of care maternity hospital through the retrospective chart review. We compared patients’ demographics, maternal obstetrical complications and neonatal adverse events between GBS positive (n=41) and GBS negative (n=603) groups. To find out clinical predictors of GBS positive result, univariable chi-square test and multivariable logistic regression analysis were applied. Results: The colonization rate of GBS in maternal vagina was 6.4% in their third trimester. GBS positive group showed significant association with the third trimester anemia (hemoglobin level <10.5 g/dL) (p=0.013) and oligohydramnios (p=0.024; odds ratio, 7.32; 95% confidence interval, 1.28–41.31). All specimens were susceptible to penicillin G and cephalosporin. The antibiotic resistance to both erythromycin and clindamycin was 31%. Conclusion The colonization rate of GBS in maternal vagina was 6.4% and third trimester maternal anemia was associated with the GBS carrying status.

Objectives: Developing large language models (LLMs) in biomedicine requires access to high-quality training and alignment tuning datasets. However, publicly available Korean medical preference datasets are scarce, hindering the advancement of Korean medical LLMs. This study constructs and evaluates the efficacy of the Korean Medical Preference Dataset (KoMeP), an alignment tuning dataset constructed with an automated pipeline, minimizing the high costs of human annotation. Methods: KoMeP was generated using the DAHL score, an automated hallucination evaluation metric. Five LLMs (Dolly-v2-3B, MPT-7B, GPT-4o, Qwen-2-7B, Llama-3-8B) produced responses to 8,573 biomedical examination questions, from which 5,551 preference pairs were extracted. Each pair consisted of a “chosen” response and a “rejected” response, as determined by their DAHL scores. The dataset was evaluated when trained through two different alignment tuning methods, direct preference optimization (DPO) and odds ratio preference optimization (ORPO) respectively across five different models. The KorMedMCQA benchmark was employed to assess the effectiveness of alignment tuning. Results: Models trained with DPO consistently improved KorMedMCQA performance; notably, Llama-3.1-8B showed a 43.96% increase. In contrast, ORPO training produced inconsistent results. Additionally, English-to-Korean transfer learning proved effective, particularly for English-centric models like Gemma-2, whereas Korean-to-English transfer learning achieved limited success. Instruction tuning with KoMeP yielded mixed outcomes, which suggests challenges in dataset formatting. Conclusions KoMeP is the first publicly available Korean medical preference dataset and significantly improves alignment tuning performance in LLMs. The DPO method outperforms ORPO in alignment tuning. Future work should focus on expanding KoMeP, developing a Korean-native dataset, and refining alignment tuning methods to produce safer and more reliable Korean medical LLMs.

Objectives: Developing large language models (LLMs) in biomedicine requires access to high-quality training and alignment tuning datasets. However, publicly available Korean medical preference datasets are scarce, hindering the advancement of Korean medical LLMs. This study constructs and evaluates the efficacy of the Korean Medical Preference Dataset (KoMeP), an alignment tuning dataset constructed with an automated pipeline, minimizing the high costs of human annotation. Methods: KoMeP was generated using the DAHL score, an automated hallucination evaluation metric. Five LLMs (Dolly-v2-3B, MPT-7B, GPT-4o, Qwen-2-7B, Llama-3-8B) produced responses to 8,573 biomedical examination questions, from which 5,551 preference pairs were extracted. Each pair consisted of a “chosen” response and a “rejected” response, as determined by their DAHL scores. The dataset was evaluated when trained through two different alignment tuning methods, direct preference optimization (DPO) and odds ratio preference optimization (ORPO) respectively across five different models. The KorMedMCQA benchmark was employed to assess the effectiveness of alignment tuning. Results: Models trained with DPO consistently improved KorMedMCQA performance; notably, Llama-3.1-8B showed a 43.96% increase. In contrast, ORPO training produced inconsistent results. Additionally, English-to-Korean transfer learning proved effective, particularly for English-centric models like Gemma-2, whereas Korean-to-English transfer learning achieved limited success. Instruction tuning with KoMeP yielded mixed outcomes, which suggests challenges in dataset formatting. Conclusions KoMeP is the first publicly available Korean medical preference dataset and significantly improves alignment tuning performance in LLMs. The DPO method outperforms ORPO in alignment tuning. Future work should focus on expanding KoMeP, developing a Korean-native dataset, and refining alignment tuning methods to produce safer and more reliable Korean medical LLMs.

Objectives: Developing large language models (LLMs) in biomedicine requires access to high-quality training and alignment tuning datasets. However, publicly available Korean medical preference datasets are scarce, hindering the advancement of Korean medical LLMs. This study constructs and evaluates the efficacy of the Korean Medical Preference Dataset (KoMeP), an alignment tuning dataset constructed with an automated pipeline, minimizing the high costs of human annotation. Methods: KoMeP was generated using the DAHL score, an automated hallucination evaluation metric. Five LLMs (Dolly-v2-3B, MPT-7B, GPT-4o, Qwen-2-7B, Llama-3-8B) produced responses to 8,573 biomedical examination questions, from which 5,551 preference pairs were extracted. Each pair consisted of a “chosen” response and a “rejected” response, as determined by their DAHL scores. The dataset was evaluated when trained through two different alignment tuning methods, direct preference optimization (DPO) and odds ratio preference optimization (ORPO) respectively across five different models. The KorMedMCQA benchmark was employed to assess the effectiveness of alignment tuning. Results: Models trained with DPO consistently improved KorMedMCQA performance; notably, Llama-3.1-8B showed a 43.96% increase. In contrast, ORPO training produced inconsistent results. Additionally, English-to-Korean transfer learning proved effective, particularly for English-centric models like Gemma-2, whereas Korean-to-English transfer learning achieved limited success. Instruction tuning with KoMeP yielded mixed outcomes, which suggests challenges in dataset formatting. Conclusions KoMeP is the first publicly available Korean medical preference dataset and significantly improves alignment tuning performance in LLMs. The DPO method outperforms ORPO in alignment tuning. Future work should focus on expanding KoMeP, developing a Korean-native dataset, and refining alignment tuning methods to produce safer and more reliable Korean medical LLMs.

BACKGROUND: The efficacy of bisphosphonates for osteoporotic fracture has been consistently reported in recent randomized controlled trials (RCTs) enrolling hundreds of patients. The objective of this study was to update knowledge on the efficacy of available bisphosphonates in the prevention of vertebral and non-vertebral fractures. METHODS: An approach “using systematic reviews” on PubMed and Cochrane Library was taken. Twenty-four RCTs investigating the effects of bisphosphonates for the prevention of osteoporotic fracture were included in final analysis. A pairwise meta-analysis was conducted with a random effects model. Subgroup analysis was performed according to the type of bisphosphonate. RESULTS: The use of bisphosphonate decrease the risk of overall osteoporotic fracture (odds ratio [OR] 0.62; P<0.001), vertebral fracture (OR 0.55; P<0.001) and non-vertebral fracture (OR 0.73; P<0.001). Subgroup analysis indicated that zoledronic acid showed the lowest risk reduction (OR 0.61; P<0.001) for overall osteoporotic fractures but no significance was observed for etidronate (OR 0.34; P=0.127). CONCLUSIONS: This update meta-analysis re-confirmed that bisphosphonate use can effectively reduce the risk of osteoporotic fracture. However, there is a lack of evidence regarding etidronate for the prevention of osteoporotic fracture.
Diphosphonates* ; Etidronic Acid ; Humans ; Meta-Analysis as Topic ; Osteoporosis ; Osteoporotic Fractures* ; Risk Reduction Behavior

BACKGROUND: Pulmonary sclerosing hemangioma is a rare thoracic tumor, and pathophysiology or clinical course of this tumor is not yet fully described. Furthermore, there is no consensus on the standard operative procedure for this tumor. MATERIAL AND METHODS: Medical records of thirty-two patients, who underwent surgical resection of pulmonary sclerosing hemangioma from 1996 to 2007, were retrospectively reviewed. RESULTS: Nineteen patients underwent lobectomy and thirteen patients underwent limited resection. Video-assisted thoracoscopic surgery was performed in 9 patients in the latter group. Lymph node dissection was done in 21 patients, and one patient was found to have lymph node metastasis of the tumor. There was no postoperative complication, no early death and no tumor-related late mortality. The mean follow-up duration was 39.3 months (2 months~129 months), and all patients were free of local recurrence and distant metastasis during this period. There was no significant difference in patient's characteristics between the two groups, except that the mean hospital stay was shorter in limited resection group than in lobectomy group (p=0.0031). CONCLUSION: Pulmonary sclerosing hemangioma usually requires surgical resection for both diagnosis and treatment. Limited resection can decrease hospital stay with a surgical outcome comparable to lobectomy, and may be preferred to lobectomy if sufficient resection margin can be achieved.
Consensus ; Follow-Up Studies ; Hemangioma ; Humans ; Length of Stay ; Lung Neoplasms ; Lymph Node Excision ; Lymph Nodes ; Medical Records ; Neoplasm Metastasis ; Postoperative Complications ; Pulmonary Sclerosing Hemangioma ; Recurrence ; Retrospective Studies ; Surgical Procedures, Operative ; Thoracic Surgery, Video-Assisted

OBJECTIVE: Despite the good prognosis of borderline ovarian tumors (BOTs), a few BOT patients experience the relapse of disease, either borderline or malignant. However, the risk of recurrence of BOTs is somewhat controversial. We intended to find out the specific characteristics and prognosis of the recurrence of BOTs. METHODS: Between 1995 and 2012, 130 women were diagnosed with BOTs at a single institution. Eleven patients diagnosed and treated for the recurrence of BOTs including seven cancerous and four borderline relapses were included for the analysis in this retrospective study. Clinicopathological characteristics and surgical procedures as well as follow-up data with overall survival were assessed. Statistical analyses was performed using the chi2 test, t-test and log-rank test with Cox regression. RESULTS: One hundred and thirty patients with mean follow-up of 65.8 months were evaluated, of whom half were below 40 years old at their first diagnosis of BOTs. Among 11 recurrent cases (8.5%), 7 cancerous transformations (5.4%) and 4 borderline recurrences (3.1%) were detected with median time of 6 and 71 months after the primary surgery, respectively. Nine out of 11 recurrences were happened at their equal or below 40-years-old age at the primary diagnosis (P=0.027). Also, all 7 cancerous relapses arose from premenopausal women of median age of 35 years old. Twenty (15%) patients had laparoscopic surgery and they were all treated conservatively. Among those 20, 5 were suffered from relapse of BOTs and laparoscopic approach was proved to be a significant risk factor for disease relapse in our study (P=0.013). Although laparoscopic surgery was revealed had meaningful influence on disease free survival (P=0.024), it was not significant on overall survival (P=0.226). CONCLUSION: Laparoscopic conservative surgery for BOTs can be accepted under close follow-ups. And younger patients should be followed for long period to evaluate recurrence.
Diagnosis ; Disease-Free Survival ; Epithelial-Mesenchymal Transition ; Female ; Follow-Up Studies ; Humans ; Laparoscopy ; Ovarian Neoplasms ; Prognosis ; Recurrence* ; Retrospective Studies ; Risk Factors

Background: This study aimed to determine if sleep disturbances may mediate the relationship between panic symptoms and depression in patients with panic disorder (PD). Methods: Electronic medical records were retrospectively reviewed for 110 consecutive patients with diagnosed PD in an outpatient clinic between October 2018 and December 2019. Measurements include the PD Severity Scale, Beck Depression Inventory-II (BDI-II) and Insomnia Severity Index (ISI). Statistical analyses were performed to assess any potential relationship between PD, insomnia and depression. Results: Of the PD patients, 88 (80%) and 89 (80.9%) had comorbid depression (BDI-II ≥ 14) and insomnia (Korean version of the ISI ≥ 8), respectively. In a mediation model using insomnia as the mediating variable, the total effect of panic symptom severity on depression was significant (t = 7.23, P < 0.001). There were significant effects of panic symptoms on insomnia (t = 4.62,P < 0.001) and of insomnia on depression (t = 6.69, P < 0.001). The main effect of panic symptom severity on depression was also significant, after controlling for the effect of insomnia (t = 5.10, P < 0.001), suggesting partial mediation. Conclusion Both depressive symptoms and insomnia are common in patients with PD and depression was partially mediated by insomnia in these patients. These results suggest that an intervention for insomnia in patients with PD might help prevent the development of depression.

Objective: Heart rate variability (HRV) reflects the regulation of the autonomic nervous system. Panic disorder is highly associated with autonomic dysfunction, and is often accompanied by depression. The aim of this study is to determine the association between depression and HRV indices in patients with panic disorder. Methods: A total of 110 outpatients diagnosed with panic disorder participated in this study. The medical records of patients with panic disorder who visited the outpatient clinic of Konkuk University Hospital between December 2018 and March 2020 were retrospectively reviewed. Measurements used in this study include the Panic Disorder Severity Scale-Self Report, Beck Depression Inventory (BDI-II), Insomnia Severity Index, and HRV. Patients were divided into depressive and non-depressive groups based on their BDI-II scores. The association between HRV indices and depressive symptoms was statistically analyzed. Results: The low frequency/high frequency (LF/HF) ratio was reduced in patients with depression (mean = −0.095, p = 0.004 in the above moderate depressive group, mean = −0.120, p = 0.020 in the severe depressive group). Significant correlations were found between depressive symptoms and standard deviation of NN interval (SDNN) (ms) (−0.19, p = 0.044), very low frequency (VLF) (ms2 /Hz) (−0.22, p = 0.021), LF (−0.25, p = 0.008), HF (−0.19, p = 0.043), and LF/HF (−0.25, p = 0.009). Multiple linear regression analysis showed that BDI predicted SDNN (ms), VLF (ms2 /Hz), LF, HF, and LF/HF. Conclusion We confirmed that the LF/HF ratio decreases when depression is accompanied by panic disorder. HRV indices may be useful markers for detecting depressive symptoms in patients with panic disorder.