Chest radiograph reading and recording system: evaluation in frontline clinicians in Zambia

Abstract

BackgroundIn Zambia the vast majority of chest radiographs (CXR) are read by clinical officers who have limited training and varied interpretation experience, meaning lower inter-rater reliability and limiting the usefulness of CXR as a diagnostic tool. In 2010–11, the Zambian Prison Service and Ministry of Health established TB and HIV screening programs in six prisons; screening included digital radiography for all participants. Using front-line clinicians we evaluated sensitivity, specificity and inter-rater agreement for digital CXR interpretation using the Chest Radiograph Reading and Recording System (CRRS).MethodsDigital radiographs were selected from HIV-infected and uninfected inmates who participated in a TB and HIV screening program at two Zambian prisons. Two medical officers (MOs) and two clinical officers (COs) independently interpreted all CXRs. We calculated sensitivity and specificity of CXR interpretations compared to culture as the gold standard and evaluated inter-rater reliability using percent agreement and kappa coefficients.Results571 CXRs were included in analyses. Sensitivity of the interpretation “any abnormality” ranged from 50–70 % depending on the reader and the patients’ HIV status. In general, MO’s had higher specificities than COs. Kappa coefficients for the ratings of “abnormalities consistent with TB” and “any abnormality” showed good agreement between MOs on HIV-uninfected CXRs and moderate agreement on HIV-infected CXRs whereas the COs demonstrated fair agreement in both categories, regardless of HIV status.ConclusionsSensitivity, specificity and inter-rater agreement varied substantially between readers with different experience and training, however the medical officers who underwent formal CRRS training had more consistent interpretations.Electronic supplementary materialThe online version of this article (doi:10.1186/s12879-016-1460-z) contains supplementary material, which is available to authorized users.

Prevalence of pulmonary tuberculosis among prison inmates: A cross-sectional survey at the Correctional and Detention Facility of Abidjan, Côte d'Ivoire

“…As expected, cough and fever lasting more than two weeks were significantly associated with the risk of confirmed TB [ 26 , 38 , 39 ]. Other classical factors such as HIV seropositivity [ 18 , 25 , 30 , 40 , 41 ], low body mass index [ 17 , 18 , 25 ], active smoking [ 38 , 42 ] and chest pain [ 25 ] were not significantly associated with TB, due to the lack of power resulting from the relatively low absolute number of confirmed TB cases. The prevalence of HIV infection was comparable to that in the general adult population of Côte d’Ivoire [ 43 ].…”

“…Each of these tests considered separately have their own limitations, and taken together they provide us with a higher bound estimate for TB prevalence that should be considered with caution. Chest X-ray interpretation is a function of X-ray quality and readers experience, and has low sensitivity [ 40 ]. Sputum-smear microscopy sensitivity is much lower than that of GeneXpert, a test that we did not use [ 44 ].…”

Comparative effectiveness of a mnemonic-use approach vs. self-study to interpret a lateral chest X-ray

“…Today, medical educators find it challenging not only to teach how to read CXR, but also to evaluate proficiency, accuracy, and repeatability in reading CXRs (11,18,23). Even among seasoned clinicians, there is significant variability in interrater reliability, proficiency, and accuracy (4,10,26).…”

Targeted active screening for tuberculosis in Zimbabwe: are field digital chest X-ray ratings reliable?

“…Finally, as digital CXRs images were used, the images were of high quality and may have reduced inconsistencies in reporting due to poor quality images. 16,17 Our main limitation was a lack of culture results, the gold standard for TB diagnosis. We were therefore not able to calculate sensitivity, specificity and predictive values of the field ratings.…”

“…Studies have reported consistent and improved rating performance among trained medical officers and when TB reference images are used. 16,21 Electronic transmission of CXR images to expert radiologists needs to be explored as it ensures more accurate TB diagnosis, quality assurance of field CXR ratings, and savings on Xpert cartridges.…”