影響齲齒與口腔黏膜病變檢查一致性因素之探討

背景：國民口腔健康狀況篩檢需要校正訓練使檢查達成良好一致性。研究目的：比較齲齒篩檢時檢查位置、齲齒指數、齲齒盛行率差異影響因素；比較口腔黏膜病變調查之檢查員一致性影響因素。研究方法：將兩次篩檢行前一致性訓練之資料進行分析，模式一為國小學童齲齒篩檢牙醫師校正訓練。模式二為口腔黏膜病變篩檢之專科牙醫師、一般牙醫師、非牙醫師組檢查員校正訓練。結果：模式一國小學童齲齒篩檢之檢查員三次校正訓練結果顯示，篩檢差異隨著次數增加而顯著減少（17.7%至10.3%； p=0.0027）。差異發生部位皆以後牙區顯著大於前牙區（p<0.0001），且隨次數增加則前牙差異顯著減少（5.7%至1.8%；p=0.0045）。比較三位牙醫師校正篩檢結果齲齒指數，其kappa值為0.75至0.78；全口齲齒盛行率之kappa 值為0.80至0.83。模式二為口腔黏膜病變篩檢之檢查員一致性訓練，受訓人員答對比例以專科牙醫師組最高（疾病名稱方面78%、嚴重類別方面78%、轉診與否方面89%），非牙醫師組最低（疾病名稱26%、嚴重類別51.5%、轉診與否65.8%）。轉診與否方面比較敏感度以專科牙醫師組最高（96.9%），非牙醫師組最低（48.1%）；精確度以專科牙醫師組最高（85.7%），一般牙醫師組最低（75.0%）。結論：模式一顯示隨校正次數增加檢查者篩檢一致性顯著進步，後牙差異顯著居多篩檢時須注意光源。模式二顯示一般牙醫師組訓練時簡化診斷項目增加判別能力；非牙醫師組訓練以轉診與否判斷，達到流行病學之初步篩檢目的。

關鍵字

口腔調查篩檢；校正；流行病學

並列摘要

Background: All epidemiological surveys of dental caries and oral mucosal lesions depend on their credibility of the use of standardized diagnostic criteria and the diagnostic reproducibility of examiners. Objective: The purposes of this study were to investigate the factors related to differences in caries screening about area, caries index and caries prevalence, and to investigate the factors related to differences in calibration of oral mucosal lesion screening. Methods: There were two models of calibration trainings in this study. Model I was inter-examiner agreement of three dentists who were undertaking three calibration’s exercises of caries survey with students from elementary school students. Model II was the evaluation of agreement in oral surgery specialists, general dentists and non-dentists (health care workers or nurses) by calibration methods based on oral mucosal lesions. Results: Model I－Inter-examiner agreement of three dentists was revealed that significantly decreased error rate by increased calibration’s frequency (17.7% to 10.3%; p=0.0027). The most errors were happened at posterior teeth area than anterior teeth area (p<0.0001), and errors decreased by increasing calibration practices (anterior teeth 5.7% to 1.8%. p=0.0045). The kappa values of inter-examiner agreement for caries index were 0.75 to 0.78. The kappa values of agreement for caries prevalence were 0.80 to 0.83. Model II－The results from oral mucosal lesions calibration showed the specialists had highest scores (diagnosis:78%, severity:78%, referable:89%); non-dentists had lowest scores (diagnosis:26%, group:51.5%, referable:65.8%). The sensitivity in the specialists were highest (96.9%), in non-dentists were lowest (48.1%). The specificity in the specialists were highest (85.7%), in the general dentists were lowest (75%). Conclusions: It is concluded that the number of caries calibration’s increased would improve inter-examiner agreement. The errors over posterior teeth were more than anterior teeth area that can be improved by lightness. Calibration of oral mucosal lesions revealed that general dentists needed more training for diagnosis of disease, non-dentists can be trained for referable of mass screening in oral mucosal lesions.