1.Occupation classification model based on DistilKoBERT: using the 5th and 6th Korean Working Condition Surveys
Tae-Yeon KIM ; Seong-Uk BAEK ; Myeong-Hun LIM ; Byungyoon YUN ; Domyung PAEK ; Kyung Ehi ZOH ; Kanwoo YOUN ; Yun Keun LEE ; Yangho KIM ; Jungwon KIM ; Eunsuk CHOI ; Mo-Yeol KANG ; YoonHo CHO ; Kyung-Eun LEE ; Juho SIM ; Juyeon OH ; Heejoo PARK ; Jian LEE ; Jong-Uk WON ; Yu-Min LEE ; Jin-Ha YOON
Annals of Occupational and Environmental Medicine 2024;36(1):e19-
Accurate occupation classification is essential in various fields, including policy development and epidemiological studies. This study aims to develop an occupation classification model based on DistilKoBERT. This study used data from the 5th and 6th Korean Working Conditions Surveys conducted in 2017 and 2020, respectively. A total of 99,665 survey participants, who were nationally representative of Korean workers, were included. We used natural language responses regarding their job responsibilities and occupational codes based on the Korean Standard Classification of Occupations (7th version, 3-digit codes). The dataset was randomly split into training and test datasets in a ratio of 7:3. The occupation classification model based on DistilKoBERT was fine-tuned using the training dataset, and the model was evaluated using the test dataset. The accuracy, precision, recall, and F1 score were calculated as evaluation metrics. The final model, which classified 28,996 survey participants in the test dataset into 142 occupational codes, exhibited an accuracy of 84.44%. For the evaluation metrics, the precision, recall, and F1 score of the model, calculated by weighting based on the sample size, were 0.83, 0.84, and 0.83, respectively. The model demonstrated high precision in the classification of service and sales workers yet exhibited low precision in the classification of managers. In addition, it displayed high precision in classifying occupations prominently represented in the training dataset. This study developed an occupation classification system based on DistilKoBERT, which demonstrated reasonable performance. Despite further efforts to enhance the classification accuracy, this automated occupation classification model holds promise for advancing epidemiological studies in the fields of occupational safety and health.
2.Occupation classification model based on DistilKoBERT: using the 5th and 6th Korean Working Condition Surveys
Tae-Yeon KIM ; Seong-Uk BAEK ; Myeong-Hun LIM ; Byungyoon YUN ; Domyung PAEK ; Kyung Ehi ZOH ; Kanwoo YOUN ; Yun Keun LEE ; Yangho KIM ; Jungwon KIM ; Eunsuk CHOI ; Mo-Yeol KANG ; YoonHo CHO ; Kyung-Eun LEE ; Juho SIM ; Juyeon OH ; Heejoo PARK ; Jian LEE ; Jong-Uk WON ; Yu-Min LEE ; Jin-Ha YOON
Annals of Occupational and Environmental Medicine 2024;36(1):e19-
Accurate occupation classification is essential in various fields, including policy development and epidemiological studies. This study aims to develop an occupation classification model based on DistilKoBERT. This study used data from the 5th and 6th Korean Working Conditions Surveys conducted in 2017 and 2020, respectively. A total of 99,665 survey participants, who were nationally representative of Korean workers, were included. We used natural language responses regarding their job responsibilities and occupational codes based on the Korean Standard Classification of Occupations (7th version, 3-digit codes). The dataset was randomly split into training and test datasets in a ratio of 7:3. The occupation classification model based on DistilKoBERT was fine-tuned using the training dataset, and the model was evaluated using the test dataset. The accuracy, precision, recall, and F1 score were calculated as evaluation metrics. The final model, which classified 28,996 survey participants in the test dataset into 142 occupational codes, exhibited an accuracy of 84.44%. For the evaluation metrics, the precision, recall, and F1 score of the model, calculated by weighting based on the sample size, were 0.83, 0.84, and 0.83, respectively. The model demonstrated high precision in the classification of service and sales workers yet exhibited low precision in the classification of managers. In addition, it displayed high precision in classifying occupations prominently represented in the training dataset. This study developed an occupation classification system based on DistilKoBERT, which demonstrated reasonable performance. Despite further efforts to enhance the classification accuracy, this automated occupation classification model holds promise for advancing epidemiological studies in the fields of occupational safety and health.
3.Occupation classification model based on DistilKoBERT: using the 5th and 6th Korean Working Condition Surveys
Tae-Yeon KIM ; Seong-Uk BAEK ; Myeong-Hun LIM ; Byungyoon YUN ; Domyung PAEK ; Kyung Ehi ZOH ; Kanwoo YOUN ; Yun Keun LEE ; Yangho KIM ; Jungwon KIM ; Eunsuk CHOI ; Mo-Yeol KANG ; YoonHo CHO ; Kyung-Eun LEE ; Juho SIM ; Juyeon OH ; Heejoo PARK ; Jian LEE ; Jong-Uk WON ; Yu-Min LEE ; Jin-Ha YOON
Annals of Occupational and Environmental Medicine 2024;36(1):e19-
Accurate occupation classification is essential in various fields, including policy development and epidemiological studies. This study aims to develop an occupation classification model based on DistilKoBERT. This study used data from the 5th and 6th Korean Working Conditions Surveys conducted in 2017 and 2020, respectively. A total of 99,665 survey participants, who were nationally representative of Korean workers, were included. We used natural language responses regarding their job responsibilities and occupational codes based on the Korean Standard Classification of Occupations (7th version, 3-digit codes). The dataset was randomly split into training and test datasets in a ratio of 7:3. The occupation classification model based on DistilKoBERT was fine-tuned using the training dataset, and the model was evaluated using the test dataset. The accuracy, precision, recall, and F1 score were calculated as evaluation metrics. The final model, which classified 28,996 survey participants in the test dataset into 142 occupational codes, exhibited an accuracy of 84.44%. For the evaluation metrics, the precision, recall, and F1 score of the model, calculated by weighting based on the sample size, were 0.83, 0.84, and 0.83, respectively. The model demonstrated high precision in the classification of service and sales workers yet exhibited low precision in the classification of managers. In addition, it displayed high precision in classifying occupations prominently represented in the training dataset. This study developed an occupation classification system based on DistilKoBERT, which demonstrated reasonable performance. Despite further efforts to enhance the classification accuracy, this automated occupation classification model holds promise for advancing epidemiological studies in the fields of occupational safety and health.
4.Evaluation of Temporal Trends of Lead Exposure in Korean Workers Using Workplace Monitoring Data
Dong-Hee KOH ; Sangjun CHOI ; Ju-Hyun PARK ; Sang-Gil LEE ; Hwan-Cheol KIM ; Inah KIM ; Jong-Uk WON ; Dae Sung LIM ; Hoekyeong SEO ; Dong-Uk PARK
Journal of Korean Medical Science 2023;38(34):e271-
Background:
Lead, which is widely used in various industrial settings, is a major health hazard for manufacturing workers. Therefore, control of lead exposure has been implemented in an effort to prevent lead-related health problems. In this study, our aim was to evaluate temporal trends in occupational lead exposure in Korean lead workers using data from monitoring of workplace exposure.
Methods:
A nationwide work environment monitoring database, data from a work environment monitoring institution, and data extracted from a review paper were utilized. Different versions of standard industrial classification codes were aligned with the 10th Korean Standard Industrial Classification, which is generally consistent with the 4th revision of the International Standard Industrial Classification. The multiple data sources were combined and temporal trends over the period from 1994–2021 were estimated. In addition, separate estimation of temporal trends in the storage battery manufacturing industry over the period from 1987–2021 was also performed.
Results:
A total of 444,296 personal airborne lead measurements were used for the estimation process. The temporal trends in occupational exposure to lead declined by −6% annually over the study period. In particular, levels of lead exposure in the storage battery manufacturing industry showed a steeper decline of −12% annually.
Conclusions
Findings of our study showed that occupational exposure to lead declined over the period from 1994 to 2021 in Korea. However, adverse effects of exposure to lead on health should be regarded with caution. The results will be useful in conduct of epidemiological studies examining lead-related effects on health.
5.Self-Esteem Trajectories After Occupational Injuries and Diseases and Their Relation to Changes in Subjective Health: Result From the Panel Study of Workers’ Compensation Insurance (PSWCI)
Seong-Uk BAEK ; Won-Tae LEE ; Min-Seok KIM ; Myeong-Hun LIM ; Jin-Ha YOON ; Jong-Uk WON
Journal of Korean Medical Science 2023;38(37):e284-
Background:
Occupational injuries and diseases are life events that significantly impact an individuals’ identity. In this study, we examined the trajectories of self-esteem among victims of occupational injury and disease and their relation to health.
Methods:
The Panel Study of Workers’ Compensation Insurance conducted annual followups on workers who had experienced occupational injury or disease. A total of 2,000 participants, who had completed medical care, were followed from 2013 to 2017. Growth mixture modeling was utilized to identify latent classes in the self-esteem trajectory.Additionally, logistic regressions were conducted to explore the association between trajectory membership, baseline predictors, and outcomes.
Results:
Three distinct trajectory classes were identified. Total 65.8% of the samples (n = 1,316) followed an increasing self-esteem trajectory, while 31.1% (n = 623) exhibited a constant trajectory, and 3.1% (n = 61) showed a decreasing trajectory. Individuals with an increasing trajectory were more likely to have a higher educational attainment (odds ratio [OR], 1.86; 95% confidence interval [CI], 1.20–2.88), an absence of a moderate-to-severe disability rating (OR, 0.49; 95% CI, 0.25–0.96), no difficulty in daily living activities (OR, 0.81; 95% CI, 0.75–0.88), and were economically active (re-employed: OR, 2.46; 95% CI, 1.52–3.98; returned to original work: OR, 4.46; 9% CI, 2.65–7.50). Those with a decreasing self-esteem trajectory exhibited an increased risk of poor subjective health (OR, 1.89; 95% CI, 0.85–4.85 in 2013 to OR, 3.17; 95% CI, 1.04–13.81 in 2017), whereas individuals with an increasing trajectory showed a decreased risk (OR, 0.54; 95% CI, 0.43–0.68 in 2013 to OR, 0.44; 95% CI, 0.33–0.57 in 2017).
Conclusion
Our findings emphasize the diversity of psychological responses to occupational injury or disease. Policymakers should implement interventions to enhance the self-esteem of victims.
6.Characteristics and Socio-Demographic Distribution of Precarious Employment Among Korean Wage Workers: A Proposition of Multidimensional Approach Using a Summative Score
Seong-Uk BAEK ; Min-Seok KIM ; Myeong-Hun LIM ; Taeyeon KIM ; Jin-Ha YOON ; Jong-Uk WON
Safety and Health at Work 2023;14(4):476-482
Methods:
To operationalize precarious employment, we utilized data from the Korean Working Conditions Survey and focused on three distinct dimensions: employment insecurity, income inadequacy, and a lack of rights and protections. By constructing a summative scale ranging from -16 to 2, with lower scores indicating higher precariousness, we measured employment precariousness among Korean wage workers. To compare employment precariousness according to survey participant characteristics, we employed the Wilcoxon Rank Sum Test.
Results:
We analyzed a weighted number of 38,432 workers. The overall sample showed a median (Q1, Q3) summative scale score of -3 (-6, -1). The median summative score was lower for women compared to men (men: -2; women: -5; p < 0.001), as well as for young or older workers compared to middle-aged workers (young: -4; middle-aged: -2; older: -5; p < 0.001). Similarly, workers with lower educational levels (middle school or below: -8; high school: -5; college or above: -2; p < 0.001) and non-white collar workers (blue collar: -5; service/sales worker: -6; white collar: -2; p < 0.001) experienced higher levels of employment precariousness.
Conclusion
Our findings indicate that certain vulnerable groups, such as women, young or older adults, workers with low educational attainment, and caregiving or low-skilled elementary workers, are disproportionately exposed to high employment precariousness. Active policy interventions are needed to improve the employment quality of vulnerable groups.
7.Hospital-Based Case Management for Suicide High-Risk Group Using Delphi Survey
Hwan LEE ; Jeungsuk LIM ; Sang Min LEE ; Soung Nam KIM ; Haewoo LEE ; Kyoung-Uk LEE ; Boung Chul LEE ; Hwa-Young LEE ; Jong-Woo PAIK
Psychiatry Investigation 2021;18(10):986-996
Objective:
As of 2019, suicide is serious problem in Korea, with the highest suicide rate among OECD countries. To reduce suicide rates Emergency Department Based Post-Suicide Attempt Case Management carried out with government funding in South Korea, but it is insufficient to address the issue. Aim of this study is to prevent suicide attempts through continuous provision of mental health services even after discharge from acute care.
Methods:
We selected 15 mental health specialists who are multidisciplinary experts in Suicide Prevention. Two-round Delphi survey was conducted on them to reach an agreement for hospital-based case management.
Results:
The first Delphi survey consisted of 8 areas and 39 questions. Among them, 30 questions draw agreement above the reference value. The second Delphi survey, consisted of 37 questions, resulted in 32 above-standard questions.
Conclusion
Consensus was reached in most category of the Hospital Based Case Management for Suicide High-Risk Group. Core of the developed plan was to provide services to patients who visited the hospital, pursue the stability and universalization of services through a medical insurance fee system. In the future, hospital-based case management service will be implemented as a new model contributing to the reduction of suicide rates in Korea.
8.Real-World Experience of Nivolumab in Non-small Cell Lung Cancer in Korea
Sun Min LIM ; Sang-We KIM ; Byoung Chul CHO ; Jin Hyung KANG ; Myung-Ju AHN ; Dong-Wan KIM ; Young-Chul KIM ; Jin Soo LEE ; Jong-Seok LEE ; Sung Yong LEE ; Keon Uk PARK ; Ho Jung AN ; Eun Kyung CHO ; Tae Won JANG ; Bong-Seog KIM ; Joo-Hang KIM ; Sung Sook LEE ; Im-II NA ; Seung Soo YOO ; Ki Hyeong LEE
Cancer Research and Treatment 2020;52(4):1112-1119
Purpose:
The introduction of immune checkpoint inhibitors represents a major advance in the treatment of lung cancer, allowing sustained recovery in a significant proportion of patients. Nivolumab is a monoclonal anti–programmed death cell protein 1 antibody licensed for the treatment of locally advanced or metastatic non-small cell lung cancer (NSCLC) after prior chemotherapy. In this study, we describe the demographic and clinical outcomes of patients with advanced NSCLC treated with nivolumab in the Korean expanded access program.
Materials and Methods:
Previously treated patients with advanced non-squamous and squamous NSCLC patients received nivolumab at 3 mg/kg every 2 weeks up to 36 months. Efficacy data including investigator-assessed tumor response, progression data, survival, and safety data were collected.
Results:
Two hundred ninety-nine patients were treated across 36 Korean centers. The objective response rate and disease control rate were 18% and 49%, respectively; the median progression-free survival was 2.1 months (95% confidence interval [CI], 1.87 to 3.45), and the overall survival (OS) was 13.2 months (95% CI, 10.6 to 18.9). Patients with smoking history and patients who experienced immune-related adverse events showed a prolonged OS. Cox regression analysis identified smoking history, presence of immune-related adverse events as positive factors associated with OS, while liver metastasis was a negative factor associated with OS. The safety profile was generally comparable to previously reported data.
Conclusion
This real-world analysis supports the use of nivolumab for pretreated NSCLC patients, including those with an older age.
9.Investigating the Feasibility of Targeted Next-Generation Sequencing to Guide the Treatment of Head and Neck Squamous Cell Carcinoma.
Sun Min LIM ; Sang Hee CHO ; In Gyu HWANG ; Jae Woo CHOI ; Hyun CHANG ; Myung Ju AHN ; Keon Uk PARK ; Ji Won KIM ; Yoon Ho KO ; Hee Kyung AHN ; Byoung Chul CHO ; Byung Ho NAM ; Sang Hoon CHUN ; Ji Hyung HONG ; Jung Hye KWON ; Jong Gwon CHOI ; Eun Joo KANG ; Tak YUN ; Keun Wook LEE ; Joo Hang KIM ; Jin Soo KIM ; Hyun Woo LEE ; Min Kyoung KIM ; Dongmin JUNG ; Ji Eun KIM ; Bhumsuk KEAM ; Hwan Jung YUN ; Sangwoo KIM ; Hye Ryun KIM
Cancer Research and Treatment 2019;51(1):300-312
PURPOSE: Head and neck squamous cell carcinoma (HNSCC) is a deadly disease in which precision medicine needs to be incorporated. We aimed to implement next-generation sequencing (NGS) in determining actionable targets to guide appropriate molecular targeted therapy in HNSCC patients. MATERIALS AND METHODS: Ninety-three tumors and matched blood samples underwent targeted sequencing of 244 genes using the Illumina HiSeq 2500 platform with an average depth of coverage of greater than 1,000×. Clinicopathological data from patients were obtained from 17 centers in Korea, and were analyzed in correlation with NGS data. RESULTS: Ninety-two of the 93 tumors were amenable to data analysis. TP53 was the most common mutation, occurring in 47 (51%) patients, followed by CDKN2A (n=23, 25%), CCND1 (n=22, 24%), and PIK3CA (n=19, 21%). The total mutational burden was similar between human papillomavirus (HPV)–negative vs. positive tumors, although TP53, CDKN2A and CCND1 gene alterations occurred more frequently in HPV-negative tumors. HPV-positive tumors were significantly associated with immune signature-related genes compared to HPV-negative tumors. Mutations of NOTCH1 (p=0.027), CDKN2A (p < 0.001), and TP53 (p=0.038) were significantly associated with poorer overall survival. FAT1 mutations were highly enriched in cisplatin responders, and potentially targetable alterations such as PIK3CA E545K and CDKN2A R58X were noted in 14 patients (15%). CONCLUSION: We found several targetable genetic alterations, and our findings suggest that implementation of precision medicine in HNSCC is feasible. The predictive value of each targetable alteration should be assessed in a future umbrella trial using matched molecular targeted agents.
Biomarkers
;
Carcinoma, Squamous Cell*
;
Cisplatin
;
Epithelial Cells*
;
Head*
;
Humans
;
Korea
;
Molecular Targeted Therapy
;
Neck*
;
Precision Medicine
;
Statistics as Topic
10.The role of customer service manual on workplace emotional burden in nationwide cross sectional study
Sehyun YUN ; Sung Shil LIM ; Jihyun KIM ; Young Kwang KIM ; Jong Uk WON ; Jin Ha YOON
Annals of Occupational and Environmental Medicine 2019;31(1):5-
BACKGROUND: We aim to discuss the overall effect of customer service manual (CSM) on service industry workers using Korean Working Condition Survey. METHODS: Out of 50,007 total survey participants, 11,946 customer service workers were included in the current study (5613 men, 6333 women). Answers to survey questions were used to define the use of CSM, emotional burden, emotional dissonance, engaging angry customers and other covariates. Emotional burden included either depressive event or stress level. Odds ratio (OR) with 95% confidence interval (95% CI) of experiencing emotional burden was calculated by logistic regression model. Interaction effect between CSM and engaging angry customer on emotional burden was also estimated. RESULTS: Out of 11,946 subjects, total of 3279 (27.4%) have experienced emotional burden. OR (95% CI) of experiencing emotional burden was 1.40 (1.19–1.64) in men and 1.25 (1.09–1.44) in women. There was gender difference in interaction effect between the use of CSM and engaging angry customers. In men, OR (95% CI) was 3.16 (1.38–7.23) with additive effect when always engaging angry customers with CSM compared to rarely engaging without CSM, while in women OR (95% CI) was 8.85 (3.96–19.75) with synergistic effect. Moreover, the risk of depressive event increased only in women with OR (95% CI) 2.22 (1.42–3.48). CONCLUSIONS: Our current study highlighted association between emotional burden and CSM in both men and women service workers. Furthermore, women were affected more severely by CSM. The results from current study suggest that CSM should be changed appropriately to benefit workers.
Female
;
Humans
;
Logistic Models
;
Male
;
Mental Health
;
Odds Ratio

Result Analysis
Print
Save
E-mail