雙診斷工具間異質相依性對於半潛藏篩檢資料準確度估計之影響探討

Effect of heterogeneous dependence between two tests on estimation of diagnostic accuracy for semi-latent screening data

何慶勳(Ching-Hsun Ho)

指導教授 ： 戴政 ； 陳秀熙

國立臺灣大學/公共衛生學院/流行病學研究所/碩士(2009年)

https://doi.org/10.6342/NTU.2009.10101

摘要

於篩檢欲檢驗出某種疾病時，當採用兩種檢驗工具而確診動作僅施予初步診斷結果為陽性的個案，則無法求得此二工具之敏感度與特異度。先前的研究藉由假設檢驗工具間為獨立或是其相依性為同質下，利用最大概似函數法或是捉捕-再捉捕法求出個別之準確度估計。由於有關工具間的相依性假設往往存在許多爭議，因此本研究利用兩種相依模式去討論異質相依性所造成的影響，研究結果顯示相依性的方向並不會干擾準確度之估計，但其強弱卻會產生相當大的影響。

關鍵字

準確度 ； 相依性 ； 檢驗工具

並列摘要

When diagnosing a disease in a screening program, two diagnostic tests may be applied simultaneously. To confirm the accuracy of the two tests, a third gold standard test will usually be also used. However, due to realistic reasons, the confirmatory procedure is limited to individuals with positive results. Previous approaches assumed independence or homogeneous dependence of the two tests and used maximum likelihood or capture-recapture method to estimate the accuracy for each test. However, there are controversies about the dependence assumptions. Therefore, in this thesis we adopted two dependent models for discussing the influence of heterogeneous dependence on diagnostic accuracy. The results indicate that the direction of dependence will not affect accuracy estimates, but the strength will.

並列關鍵字

accuracy ； dependence ； diagnostic test

參考文獻

Adami, H.-O., Day, N.E., Trichopoulos, D., and Willett, W.C. (2001),“Primary and secondary prevention in the reduction of cancer morbidity and mortality,”European Journal of Cancer, 37, S118-S127.

Albert, P.S. and Dodd, L.E. (2008),“On estimating diagnostic accuracy from studies with multiple raters and partial gold standard evaluation,”Journal of the American Statistical Association, 103, 61-73.

Albert, P.S. and Dodd, L.E. (2004),“A cautionary note on the robustness of latent class models for estimating diagnostic error without a gold standard,”Biometrics, 60, 427-435.

Albert, P.S., McShane, L. M., Shih, J. H., et al (2001),“Latent class modeling approaches for assessing diagnostic error without a gold standard：With applications to p53 immunohistochemical assays in bladder tumors,”Biometrics, 57, 610-619.

Begg, C. B. and Greenes, R. A. (1983),“Assessment of diagnostic tests when disease verification is subject to selection bias,”Biometrics, 39, 207-215.