Extended Data Table 8 Parameter definitions, values, and sources

From: Estimating the number of incorrect tuberculosis diagnoses in low- and middle-income countries

  1. Parameters assumed to followed the same distribution independently for each country.
  2. * Sensitivity values for smear and RDT defined with respect to culture-negative pulmonary TB, such that sensitivity for all pulmonary TB is equal to \({{se}}^{t1}* ({1-p}^{cult-neg})\). Values for RDT in main analysis based on Xpert Ultra.
  3. # With the binormal model, the sensitivity of clinical diagnosis is defined as a function of specificity, via the relationship \({{se}}^{t2}=\Phi \left({b}^{0}+{b}^{1}{\Phi }^{-1}\left(1-{{sp}}^{t2}\right)\right)\), where \(\Phi\) is the probit transformation (cumulative distribution function of the standard Normal distribution), and \({\Phi }^{-1}\) is the inverse probit transformation.
  4. & Fraction culture-negative based on the fraction clinically-diagnosed in a high-income country setting.
  5. § Equivalent to an odds ratio of 2.0 (1.5, 3.0).
Back to article page