李龍騰臺灣大學：預防醫學研究所林敬修Lin, Ching-HsiouChing-HsiouLin2007-11-282018-06-292007-11-282018-06-292005http://ntur.lib.ntu.edu.tw//handle/246246/59199背景：國外很多研究對於礦工罹患塵肺症後其死亡率與特定疾病死亡率並沒有定論，針對導致塵肺症病人死亡的個別危險因子，國外並無相關研究，而塵肺症更是瑞芳鎮2001年男性十大死因之第五名，因此探討上述這些問題，對於瑞芳鎮退休男性礦工罹患塵肺症有其相當重要意義。 目的：本研究主要目的在於利用大樣本及貝氏分析求出塵肺症病人的標準化死亡比，並分析塵肺症會不會增加特定疾病的死亡率，另一目的為利用病例對照研究探討有哪些危險因子會造成塵肺症患者死亡。 方法：本研究一部份為回溯性世代研究，研究對象為548名曾至瑞芳鎮衛生所看診或打流感疫苗的塵肺症男性患者，資料收集期間為2000年7月1日至2004年6月30日，連結到死亡登記檔得到79名患者死亡，利用大樣本及貝氏分析方法計算出標準化死亡比，並求出各項死因有無統計上意義增加或減少，另外從未死亡名單隨機抽取死亡組兩倍人數當對照組，連結到疾病管制局的結核病防治網得到肺結核的罹病率，並利用電話問卷訪問獲得一些資料變項，以邏輯式迴歸分析尋找導致死亡率上升的危險因子。 結果：經貝氏統計分析後，塵肺症患者所有死因的標準化死亡比明顯比一般族群高(SMR:120, 95% CI: 116-124)，主要上升原因是呼吸道疾病死亡(SMR: 402, 95% CI: 320-506)，在所有癌症死亡率並沒有統計意義上升(SMR: 101, 95% CI: 95-107)，但胃癌死亡率卻有統計意義上升(SMR: 132, 95% CI: 100-175)。而在病例對照研究則顯示肺結核、抽菸習慣、年齡及礦坑內工作年數其死亡危險比分別為2.61(95% CI: 1.28-5.30)、2.00(95% CI: 1.04-3.85)、1.09(95% CI: 1.02-1.17)和1.08(95% CI: 1.03-1.14)。 結論：本研究發現塵肺症病人所有疾病、呼吸道疾病和胃癌標準化死亡比有顯著上升，另外肺結核、抽菸習慣、和礦坑內工作年數也是造成死亡的危險因子。Introduction: Whether miners with pneumoconiosis were at increased risk for all-cause or specific-cause death is still controversial. Individual correlates in association with the risk for death among coal workers’pneumoconiosis (CWP) have not been fully addressed. To throw light on these points is of paramount importance to retired miners at Rueifang Township where the cause of pneumoconiosis has been the top five leading causes in 2001. Objectives: The aim of this study is to use SMR analysis with Bayesian approach to assess whether CWP cases lead to an increase in total mortality and specific cause mortality by combing information obtained from previous study with current data. A case-control study was also conducted to assess the effect of exposure to underground mining on the risk for death and to identify significant individual correlates and their effect modification with exposure to coal mine workers. Methods: We first identified a retrospective cohort with 548 retired male miners diagnosed as pneumoconiosis who sought medical consultation or seasonal flu vaccination at Rueifang Health Center. The time period for the collection of the data was from 1st of July, 2000 till 30th of June, 2004. We ascertained 79 deceased patients after linkage with mortality registry. Frequentist or Bayesian Method for Standard Mortality Ratio Analysis was used to calculate the SMR ratio to see if there was any increase in total mortality or a variety of causes of death. The ratio of 1:2 case-control study was designed with 79 deaths as the case and twice number of subjects randomly selected from the CWP cases still alive as the control group,. Information on tuberculosis (TB) was linked to the pulmonary TB prevention web site of the Center for Disease Control, Taiwan for the prevalence of TB. Other information was administered by a telephone-interview questionnaire with proxy subjects for the deceased. A logistic regression model was used to identify individual correlates associated with the risk for death. Results: Using the Bayesian Method, all-cause death in patients with pneumoconiosis was significantly increased in comparison with general population (SMR:120, 95% CI: 116-124). The main causes for the increase was attributed to respiratory diseases (SMR: 402, 95% CI: 320-506). In all cancers’ mortality rate, there was no statistically significant increase (SMR: 101, 95% CI: 95-107), but stomach cancer rate had a remarkable increase in risk for death (SMR: 132, 95% CI: 100-175). In case-control study, the results show that TB, smoking habits, age and the number of year working in the mining pit were three independent significant factors and have respective adjusted odds ratios with the orders of 2.61 (95% CI: 1.28-5.30), 2.00(95% CI: 1.04-3.85), 1.09(95% CI: 1.02-1.17) and 1.08(95% CI: 1.03-1.14). Conclusion： The present study demonstrated that CWP cases have significantly higher risk for all-cause death, stomach cancer, and non-malignant respiratory disease. TB, smoking habit and number of years working in the mining pit were three principal factors responsible for death.摘要 i Abstract iii Contents vi Table List vii Appendix ix 1. Introduction 1 1-1. Pneumoconiosis 1 1-2. Mining Industry and Pneumoconiosis in Taiwan 3 1-3. Rationale of the Study and Research Background 4 1-4. Purpose of Study 6 2. Literature Review 7 2-1. Characteristics of the Miners 7 2-2. The Radiological Changes of Pneumoconiosis 8 2-3. Pulmonary Function Tests of Pneumoconiosis 10 2-4. The Relationship Between Miners and Smoking 11 2-5. The Prevalence of Pneumoconiosis 12 2-6. The Relationship Between Pneumoconiosis and Tuberculosis 15 2-7. The Mortality of Pneumoconiosis 16 2-8. The Analysis of Rueifang Township Residents' Cause of Death 20 3. Materials and Methods 23 3-1. Study Subjects 23 3-2. Study Design 24 3-3. Measurement of the Variables and Data Collection 25 3-4. Statistical Analysis 28 4. Result 32 4-1. Frequentist Approach 32 4-2. Bayesian Approach 34 4-3. Case-Control Study 36 5. Discussion 39 5-1 Major Finding 39 5-2 Comparison with Previous Studies 40 5-3. Strength of the Study 46 5-4. Limitation and Further Direction 47 6. Conclusion 50 7. Reference 51 Table List Table 1 International Labour Office Classification of Pneumoconiosis 57 Table 2 The Prevalence of Pneumoconiosis and Pulmonary Tuberculosis in Coal Miners. Taiwan 1983-88 58 Table 3 Top 10 Cause-Specific Deaths of Rueifang Male in 2001 59 Table 4 Literature Review of Standard Mortality Ratio (SMR) in Miners With Pneumoconiosis 60 Table 5 Literature Review of Risk Factors in Miners With Pneumoconiosis 61 Table 6 Number of Retired Miners With Pneumoconiosis and Which Year Enrolled into the Study 62 Table 7 Age Distribution of 548 Miners and Mean Person-Years of Follow up 63 Table 8 Vital Status Among 548 Miners With CWP 64 Table 9 Cause-Specific Observed Deaths in 548 Retired Miners With Pneumoconiosis, Rueifang 2000-2004 65 Table 10 Cause-Specific Observed Deaths and SMR in Retired Miners With Pneumoconiosis, Taiwan 2000-2004 66 Table 11 The Results of SMR Analysis on All-Cause Deaths for CWP Cases Using Bayesian Approach 67 Table 12 The Results of SMR Analysis on All malignant Neoplsma Deaths for CWP Cases Using Bayesian Approach 68 Table 13 The Results of SMR Analysis on All-Cause Except Malignant Neoplasmas Deaths for CWP Cases Using Bayesian Approach 69 Table 14 The Results of SMR Analysis on Respiratory Diseases Deaths for CWP Cases Using Bayesian Approach 70 Table 15 The Results of SMR Analysis on Gastric Cancer Deaths for CWP Cases Using Bayesian Approach 71 Table 16 The Results of SMR Analysis on Esophageal Cancer Deaths for CWP Cases Using Bayesian Approach 72 Table 17 The Results of SMR Analysis on Lung Cancer Deaths for CWP Cases Using Bayesian Approach 73 Table 18 The Results of SMR Analysis on Infectious Diseases Deaths for CWP Cases Using Bayesian Approach 74 Table 19 The Results of SMR Analysis on Pulmonary Tuberculosis Deaths for CWP Cases Using Bayesian Approach 75 Table 20 The Comparison of Frequentist and Bayesian SMR in Retired Miners With Pneumoconiosis, Taiwan 2000-2004 76 Table 21 Characteristics of 79 Died and 160 Alive Miners With Pneumoconiosis 77 Table 22 Univariate Analysis of Case Control Study 78 Table 23 Model Selection Using Entry Method 79 Table 24 Model Selection Using Stepwise Regression Method 80 Table 25 Logistic Regression Analysis of Case Control Study 81 Table 26 Assessment of Interaction Between TB and Years Association With Death Among CWP Deaths 82 Table 27 Assessment of Interaction Between Smoke and Years Association With Death Among CWP Deaths 83 Table 28 Assessment of Interaction Between TB and Smoke Association With Death Among CWP Deaths 84 Appendix The Map of Rueifang Township 85 The Questionnaire of Case Control Study 861606819 bytesapplication/pdfen-US煤礦工塵肺症標準化死亡比coal workers’ pneumoconiosisstandard mortality ratio[SDGs]SDG3thesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/59199/1/ntu-94-R92846011-1.pdf