Bayesian evaluation of the performance of three diagnostic tests for Chlamydophila psittaci in humans

Abstract

Chlamydophila psittaci (C. psittaci) remains a significant threat to the health of farming communities in close contact with psittacine birds yet its infection burden remains poorly understood owing to the low accuracy of available diagnostic tests. This study aimed to evaluate the accuracy of chicken embryo (CEI) and mice inoculation (MI) tests and a PCR assay for the detection of C. psittaci in humans. Sputum specimens from 70 Egyptian individuals in contact with psittacine birds were screened for the presence of the pathogen using the three tests. A Bayesian latent class model was used to estimate the Se and Sp of the three tests. The PCR assay had a higher Se (85%; PCI: 42.4% - 99.4%) than CEI (68.5%; PCI: 24.6% - 95.6%) and MI (47.0%; PCI: 12.3% - 85.1%) tests together with a higher Sp (98.9%; PCI 94.1% - 100%) than CEI (98.6%; PCI: 93.8% - 99.9%) and MI (98.6%; PCI: 93.8% - 99.9%) tests. To our knowledge, this is the first attempt at evaluating the accuracy of these tests for the detection of C. psittaci in humans. The PCR assay clearly outperforms the inoculation tests and hence holds better promise for routine use in surveillance programs for psittacosis.