Assessing the Performance of Diagnostic Tests When a Gold Standard Does Not Exist

Abstract

Situations are encountered that gold standard doesn’t exist and two or more imprecise referencetests are available in the diagnosis of a specific disease clinically. There are three basicapproaches including Discrepant Resolution (DR), Composite Reference Standard (CRS) andLatent Class model (LCA) in literature for assessing the performance of diagnostic tests. Thisstudy is aimed to evaluate the three basic approaches. LCA has a more complex structure thanDR and CRS approach. LCA is more advantageous than CRS and DR methods without anyassumptions about resolver test or other imperfect reference tests specify. As a result, LCAmethods which are developed for assessing the performance of new diagnostic test and reducingthe bias because of imperfect reference tests give more reliable results than other methods

Keywords

• Diagnostic test,Composite reference standard,Latent class analysis

ALTIN STANDART TESTİN OLMADIĞI DURUMDA TANI TESTLERİNİN PERFORMANSLARININ DEĞERLENDİRİLMESİ

Abstract

Klinikte belirli bir hastalığın tanısının konulmasında, altın standart testin olmadığı ama iki veyadaha fazla kesin olmayan referans testin var olduğu durumlarla karşılaşılmaktadır. Yeni testinkesin olmayan referans testlere göre performanslarını değerlendirmek için literatürde DiscrepantResolution (DR), Composite Referans Standart (CRS) ve Latent Class model (LCA) olmaküzere üç temel yaklaşım bulunmaktadır. Bu çalışma ile üç temel yaklaşımın değerlendirilmesiamaçlanmıştır. LCA, DR ve CRS yaklaşımından çok daha karmaşık bir yapıya sahiptir. LCA,çözücü test ya da diğer bir kesin olmayan referans testin seçiciliği hakkında herhangi birvarsayım gerektirmeksizin CRS ve DR yöntemlerinden daha avantajlıdır. Sonuç olarak, yenitanı testinin performansını değerlendirmek ve kesin olmayan referans testler sonucunda ortayaçıkan yanlılığı azaltmak adına geliştirilen yöntemlerden LCA, diğer yöntemlere nazaran dahagüvenilir sonuçlar vermektedir

Keywords

• Tanı testi,Composite referans standart,Gizli sınıf analizi

References

1. Reitsma JB, Rutjes AWS, Khan KS, Coomarasamy A, Bossuyt PM. A Review of solution for diagnostic accuracy studies with an imperfect or missing reference standard. J Clin Epidemiol. 62: 797-806, 2009. DOI: 10.1016/j.jclinepi.2009.02.005.
2. Genç Y. Tanı testi çalışmalarında metodolojik standartların kullanılması. Ankara Üniversitesi Tıp Fakültesi Mecmuası. 56 (4): 259-64, 2003.
3. Albert PS. Estimating diagnostic accuracy of multiple binary tests with an imperfect reference standard. Stat Med. 28: 780- 97, 2009. DOI: 10.102/sim.3514.
4. Zhou XH, Obuchowski NA, McClish DK. Statistical Methods in Diagnostic Medicine. Wiley Series in Probability and Statistics. New York: John Wiley& Sons, Inc. pp:71-77, 2002.
5. Staquet M, Rozencweig M, Lee YJ, Muggia FM. Methodology for the assessment of new dichotomous diagnostic tests. Journal of Chronic Diseases. 34:599–610, 1981. http://dx.doi.org/10.1016/0021-9681(81)90059-X.
6. Baker SG. Evaluating a new test using a reference test with estimated sensitivity and specificity. Communications in Statistics: Theory and Methods. 20:2739–52, 1991.
7. Whiting P, Rutjes AWS, Reitsma JB, Glas AS, Bossuyt PM, Kleijnen J. Sources of variation and bias in studies of diagnostic accuracy: a systematic review. Ann Intern Med. 140(3):189–202, 2004.
8. Valenstein PN. Evaluating diagnostic tests with imperfect standards. Am J Clin Pathol. 93(2):252–8, 1990.

9. Begg CB. Biases in the assessment of diagnostic tests. Stat Med.6:411–423, 1987.
10. Nerette P, Stryhn H, Dohoo I, Hammel L. Using pseudogold standards and latent-class analysis in combination to evaluate the accuracy of three diagnostic tests. Prev Vet Med. 85: 207- 25, 2008. DOI: 10.1016/j.prevetmed.2008.01.011.
11. Baughman AL, Bisgard KM, Cortese MM, Thompson WW, Sanden GN, Strebel PM. Utility of composite reference standards and latent class analysis in evaluating the clinical accuracy of diagnostic tests for pertussis. Clin Vaccine Immunol. 15 (1): 106-14, 2008. DOI:10.1128/CVI.00223-07.
12. Alonzo TA, Pepe MS. Using a combination of reference tests to assess the accuracy of a new diagnostic test. Stat Med. 18: 2987-3003, 1999.
13. Miller W. Editorial response: can we do better than discrepant analysis for new diagnostic test evaluation? Clin Infect Dis. 27(5): 1186-93, 1998.
14. Hadgu A. The discrepancy in discrepant analysis. Lancet. 348 (9027): 592-93, 1996. DOI: http://dx.doi.org/10.1016/S0140- 6736(96)05122-7.
15. Lipman HB, Astles JR. Quantifying the bias associated with use of discrepant analysis. Clin Chem. 44(1): 108-15, 1998.
16. Hawkins MD, Garrett AJ, Stephenson B. Some issues in resolution of diagnostic tests using an imperfect gold standard. Stat Med. 20: 1987-2001, 2001.
17. Jang D, Sellors JW, Mahony JB, Pickard L, Chernesky MA. Effects of broadening the gold standard on the performance of a chemiluminometric immunoassay to detect Chlamydia trachomatis antigens in centrifuged first void urine and urethral swab samples from men. Sex Transm Dis. 19(6): 315-9, 1992.
18. Hess AS, Shardell M, Johnson JK, Thom KA, Strassle P, Netzer G, et al. Methods and recommendations for evaluating and reporting a new diagnostic test. Eur J Clin Microbiol Infect Dis. Published online: 29 March 2012. DOI: 10.1007/s100096-012-1602-1.
19. Rutjes AWS, Reitsma JB, Coomarasamy A, Khan KS, Bossuyt PMM. Methodological research in diagnostic tests: Evaluation of diagnostic tests when there is no gold standard-A review of methods. NHS Research Methodology Programme 2006, UK (grant: RM04/JH21: No Gold Standard).
20. Toft N, Jorgensen E, Hojsgaard S. Diagnosing diagnostic tests: evaluating the assumptions underlying the estimation of sensitivity and specificity in the absence of a gold standard. Prev http://dx.doi.org/10.1016/j.prevetmed.2005.01.006. 68: 19-33, 2005.
21. Hui SL, Walter SD. Estimating the error rates of diagnostic tests. Biometrics. 36:167–171, 1980.
22. Pepe MS, Janes H. Insights into latent class analysis of diagnostic test performance. Biostatistics. 8(2): 474-84, 2007.DOI: 10.1093/biostatistics/kxl038.
23. Vermunt JK. Latent Class and finite mixture models for multilevel data sets. Stat Methods Med Res. 1-19, 2007. DOI:10.1177/0962280207081238.