Establishing a Risk Score Model for Oral Potentially Malignant Disorder and Oral Squamous Cell Carcinoma

• Main Authors: Shuai-Lun Lu, 呂帥倫
• Other Authors: Chien-Hung Lee
• Format: Others
• Language: zh-TW
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/77749240300439836231

碩士 === 高雄醫學大學 === 公共衛生學系公共衛生學碩士班 === 105 === Backgound 　　Oral cancer is the fourth most common neoplasm and cancer death among men in Taiwan. But with appropriate treatment at early period, the 5-year survival rate for patients could remain 73.3%-82.4%. Oral cancer had been proved to be associated with the use of chewing betel nuts, smoking and drinking alcohol. The chemical structure of arecoline in nuts is similar to nicotine, so it may be a psychoactive substance as well. Futhermore, oral cancer is one of the few malignances that can be earlier controlled for its precancerous lesions, and placed importance on the preventive and screening stages by the government.Clinically, several risk-scoring systems have been successfully developed and properly applied to predicting diseases and cancers, but still few appliances in oral maglignance disorder(OMD). Purpose 　　This study aims to evaluate the risk of oral malignant disorder for betel nut chewing, smoking, drinking and other factors,and to establish OMD scoring system and validate by using community group data. Material and methods 　　To investigate these issues, we conducted a case-control study in the Kaohsiung Medical University Hospital, Taiwan. The study period was from April,2016 to April,2017. A total of 182 pathologically confirmed oral potential malignant disorder or oral squamous cell carcinoma patients and 214 controls were recruited. A questionnaire which was developed via the DSM-5 substance use disorder was used to measure a total of 11 disorder symptoms of BQ, alcohol and cigarette, as well as other covariates. Participants with 0-1, 2-3, 4-5 and ≥6 DSM-5 symptoms were defined none, mild, moderate and severe disorder, respectively. Polytomous logistic regression models were used to establish OMD risk and its cumulative risk score (CRS). We also randomly recruited 150 BQ chewers and 150 non-chewers in the Kaohsiung community. A OMD risk score was calculated for each community participant. We used receiver operating characteristic (ROC) curve, area under the ROC curve, sensitivity and specificity to assess the validity of discriminating oral potential malignant disorder (OPMD) using OMD risk score. Results 　　In the case-control study, betel nuts disorder(BUD) was found to be associated with a 56.5-fold risk of OMD, with a 157.4-fold higher risk identified among chewers with severe BUD. The CRS defined by the risk predictive model of OMD ranges from 0 to 7. The scoring schemes showed an excellent discrimination for identifying OMD (AU-ROC = 0.930, P <0.05), The optimal CRS cutoff values for risk-scoring systems were 3. The sensitivity and specificity for higher CRS were 86.8% and 87.4 %, respectively. In tis OMD model, high risk group (CRS ≥3) was associated with a 44.6-fold higher risk of OMD. In the community-based evaluation, the CRS scheme showed an acceptable accuracy (AU-ROC = 0.77) for OPMD determination (sensitivity, 95.5%; specificity, 48.9%; and negative predictive value, 99.3%). Conclusion 　　Our study found BUD was strongly associated with OMD. Our findings also provided evidence to support that BUD can be considered a key factor in the development of an efficient risk scoring model for community OPMD screening program.