Diagnostic accuracy of clinical illness for bovine respiratory disease (BRD) diagnosis in beef cattle placed in feedlots: A systematic literature review and hierarchical Bayesian latent-class meta-analysis

详细信息    查看全文

• 作者：E. Timsit^a ; ^b ; ^{eftimsit@ucalgary.ca} ; N. Dendukuri^c ; ^{nandini.dendukuri@mcgill.ca} ; I. Schiller^c ; ^{Ian.Schiller@clinepi.mcgill.ca} ; S. Buczinski^d ; ^{s.buczinski@umontreal.ca}
• 关键词：BRD ; bovine respiratory disease ; BCI ; Bayesian credible interval ; CAB ; Commonwealth Agricultural Bureau ; CI ; clinical illness ; DIC ; deviance information criterion ; HSROC ; hierarchical summary receiver operating characteristic curve ; LU ; lung lesion ; Se ; sensitivity ; Sp ; specificity
• 刊名：Preventive Veterinary Medicine
• 出版年：2016
• 出版时间：1 December 2016
• 年：2016
• 卷：135
• 期：Complete
• 页码：67-73
• 全文大小：731 K
• 卷排序：135

文摘

Diagnosis of bovine respiratory disease (BRD) in feedlots is based on clinical illness (CI) detected by pen-checkers. Diagnostic accuracy of CI was evaluated using lung lesions at slaughter as an imperfect reference test. A hierarchical Bayesian latent-class meta-analysis (7 studies included) was used to model test accuracy. Pooled estimates of CI’s sensitivity and specificity were low (0.27; 0.12-0.65) and high (0.92; 0.72-0.98), respectively. Diagnostic accuracy of CI was highly heterogeneous among studies.