Bulletin of the World Health Organization

Structured approaches for the screening and diagnosis of childhood tuberculosis in a high prevalence region of South Africa

Mark Hatherill, Monique Hanslo, Tony Hawkridge, Francesca Little, Lesley Workman, Hassan Mahomed, Michele Tameris, Sizulu Moyo, Hennie Geldenhuys, Willem Hanekom, Lawrence Geiter & Gregory Hussey

Volume 88, Number 4, April 2010, 312-320

Table 1. Results of chest radiograph assessment by three independent paediatric reviewers, grouped by certainty of tuberculosis diagnosis, South Africa, 2001–2006

Diagnostic certaintya Reviewer 1
Reviewer 2
Reviewer 3
Final classification
No. % No. % No. % No. %
Highly likely to have tuberculosis 16 1.1 29 2.0 171 11.8
Likely to have tuberculosis 20 1.4 38 2.6 323 22.4
Suspected of having tuberculosis 124 8.6 145 10.0 242 16.7
Positive 160 11.1 212 14.6 736 50.9 271 18.8
Inconclusive 45 3.1 35 2.4 82 5.7
Abnormal but not tuberculosis 102 7.1 139 9.6 312 21.6
Normal 1038 71.8 778 53.9 59 4.1
Negative 1185 82.0 952 65.9 453 31.4 1174 81.2
Not read 100 6.9 281 19.5 256 17.7
Total 1445 100 1445 100 1445 100 1445 100


a “Highly likely to have tuberculosis”, “likely to have tuberculosis” and “suspected of having tuberculosis” were classified as positive; “inconclusive”, “abnormal but not tuberculosis” and “normal” were classified as negative. Final chest radiograph classification was determined by agreement of at least two reviewers.

[an error occurred while processing this directive]