行政院國家科學委員會專題研究計劃成果報告:口腔白斑與口腔癌之臨床及流行病學研究--配對病例對照及病例追蹤研究

前官：利用除去口腔癌前病兆，如口腔白斑，來預防口腔癌，是一種可行的防治策略• 但必須了解從正常到白斑及口腔癌的自然病史，方能評估各種危險因子對發生白斑及其轉變成口腔癌的影響。而先前研究常因個案數太少、追蹤時間過長、及受到白斑治療的影響，而有所限制。本研究提出另一種回溯性追蹤方式，來為克服上述問題。 目標：本研究之目的為：利用三階段．馬可夫鏈模式，同步估計白斑發生及轉變成癌症的速率。 材料與方法：從台北某一家醫學中心，回溯追蹤自1988 至1998 間三梱疾病世代，利用口腔白斑及牙周病兩世代，以白斑世代資料連結癌症登記資料及病歷回顧方式，發展三階段病例對照研究方法，由口腔白斑、口腔癌及牙周病三世代中抽樣，探討第三個研究目的。統計方法為：發展三階段，馬可夫鏈模式，同時估計口腔白斑發生率及其轉變癌症的速率。 結果‘經分析資料結果顯示：同時估計白斑發生率為0 . 20 % ( 95 % Cl : 0 . 16 ％-0 . 25 % ) 及其轉變癌症速率為5 . 42 % ( 95 % Cl : 4 . 22 ％-6 . 62 ％》 。嚼食檳榔發生白斑為不吃檳榔的17 . 7 倍（9 . 03 - 34 . 5 ）。吸菸造成白班的相對危險率為4 . 26 ( 2 . 21 -8 . 23 ) ，而飲酒僅對白斑轉變成。腔癌有影響，相對危險為2 . 37 ( 1 . 47 -3 . 82 ）。 結論：由疾病的自然病史可以用來估計去除危險因及及白斑治療對預防a 腔癌的效果，除去檳榔可減少94 ％的白斑及61 ％的白斑癌病變，避免飲酒僅可減少82 % 白斑癌病變對白斑的發生並無影響，戒除吸終可降低77 ％的白斑個案及70 ％的癌轉變。三階段馬可夫鏈模式應用於回溯性世代之病例對照方法對疾病轉變速率的估計，將可運用於其他慢性疾病研究。Introduction: To prevent oral cancer, elimination of precancerous lesion i.e. leukoplakia is feasible preventive strategy. It is necessary to make a clear understanding of natural history from normal, leukoplakia, to oral cancer to know the efficacy of dental check-up for leukoplakia in reducing oral cancer. But, several limitations in previous studies are rare cases, long-term follow-up, and treated effect of leukoplakia. An alternative approach is proposed to tackle these problems. Objectives: The aim of this thesis is to estimate the incidence of leukoplakia and malignant transformation simultaneously based on a three-state Markov chain model. Material and Methods: Three retrospective follow-up cohorts from 1988 to 1998 are identified from one medical center in Taipei. Using leukoplakia and periodontal disease cohort, a matched case-control study is design to answer the first objective. The leukoplakia cohort is linked to cancer registry and chart review to identify oral carcinoma to answer the second objective. A three-state Markov chain model adapted to case-control study is developed to estimate the incidence of leukoplakia and malignant transformation. Results: Using of three-state case-control method, the incidence of leukoplakia for this cohort is 0.20% (95%Cl:O.16%-0.25%), and malignant transformation is 5.42%(95%Cl:4.22%-6.62%). The adjusted odds ratios for chewing betel nut and smoking on the occurrence of leukoplakia are 17.7 (95%Cl: 9.03-34.5) and 4.26 (95%Cl: 2.21-8.23). Alcohol drinking is the only factor to effect the malignant transformation of leukoplakia, and the odd ratio is 2.37(95%Cl: 1.47-3.82) Conclusion: Getting rid of chewing betel nut would avoid 94% occurrence of leukoplakia and 66% of malignant transformation. Removing alcohol drinking habit would avoid 82% malignant transformation, but no effect to prevent leukoplakia. Quitting smoking would avoid 77% occurrence of leukoplakia and 74% of malignant transformation. From methodological viewpoint, three-sate Markov chain model adapted to estimate transition parameters based on a retrospective case-control study, can be applied to other chronic diseases