作者: Alessandra Queiroga Gonçalves , Rosa Abellana , Hélio Doyle Pereira-da-Silva , Ivanildes Santos , Paula Taquita Serra
DOI: 10.1016/J.ACTATROPICA.2013.11.026
关键词:
摘要: Performance evaluation of diagnostic tests is critical in the search for accurate diagnoses. A gold standard test usually absent parasitology, thus rendering satisfactory assessment accuracy difficult. Moreover, reliability (assessed by study repeatability) a rarely studied characteristic tests. This compared and evaluated performance (repeatability, concordance accuracy) spontaneous sedimentation technique (SST) Paratest diagnosis Giardia lamblia, Entamoeba histolytica complex, Blastocystis spp., Ascaris lumbricoides, hookworm, Trichuristrichiura Calodium hepaticum. Fecal samples 143 individuals were separated into three replicates each test. Concordance homogeneity results between evaluated. Proportions positives, sensitivity specificity estimated using Bayesian Latent Class Model. High repeatability both was found detection intestinal parasites, except spp. hookworm. generally high (concordance correlation coefficient, 0.72–0.88), hookworm T. trichiura. The detected more cases fewer than SST. quite discordant low (39.4–49.2% SST, 35.8–53.8% Paratest) (93.2–97.2%) presented slightly higher (53.8%), SST did so (49.2%). first on (using approach) two techniques. These suggest underdiagnosis little dense parasitic forms due to technical limitations We conclude that combined repeatability, key strategy better also useful identification limitations.