Sensitivity and specificity of screening tools and smear microscopy in active tuberculosis case finding

Setting Community based five pulmonary tuberculosis (PTB) surveys among adults. Objectives Estimate sensitivity and specificity of screening tools for PTB and sputum microscopy. Methods For each survey site, we estimated sensitivity and specificity of different screening criteria and microscopy against culture; pooled estimates were obtained using Random Effects Model. Results Sensitivity of cough alone, screening for any symptom (persistent cough ≥2 weeks, fever or chest pain ≥1 month, hemoptysis), any symptom or history of anti-TB treatment (h/o ATT) were 56.2%, 66% and 71.2% respectively; specificities were 95.3%, 93.8% and 92.7% respectively.X-ray when used alone for primary screening had sensitivity and specificity of 76.6% and 97.3% respectively. When used along with screening for cough, these figures were 94.3% and 93.1%, and 100% and 97.3% when used with any symptom and h/o ATT. When used for secondary screening, sensitivity and specificity of X-ray was 66.8% and 87.8% respectively after primary screening for cough, 65.0% and 89.8% after screening for any symptom, and 67.1% and 86.7% when used after screening for any symptom or h/o ATT.Pooled sensitivity and specificity of smear was 46.2% and 99.3% respectively. Conclusion Program managers may use these estimates while evaluating algorithms for active case finding.