Input for cost–benefit analysis
| Variables category | Variable name | Distribution# | Value (base case) | Relative change (range) | Reference |
| Costs of first-line drugs per day € | cTBD | Triangular | 6.3 | ±20% (5.04–7.56) | [6] |
| Costs of WHO standard MDR-TB drugs per day € | cMDR_TB | Triangular | 101.04 | ±20% (80.83–121.25) | [6] |
| Costs of Ziehl–Neelsen microscopy € | cZN | Triangular | 6.41 | ±20% (5.12–7.69) | GOÄ no. 4513 (online supplementary section 2.b) |
| Costs of mycobacterial culture € | cCulture | Triangular | 23.31 | ±20% (18.65–27.97) | GOÄ no. 4540 (online supplementary section 2.b) |
| Costs of Xpert MTB/RIF € | cXpert | Triangular | 110.75 | ±20% (88.6–130.9) | GOÄ nos. 4780 and 4784 (online supplementary section 2.b) |
| Opportunity costs of blocking twin bed € | cOpp | Triangular | 314.71 | ±20% (253.97–380.95) | Calculated from InEK (http://www.g-drg.de/cms/) data (online supplementary section 2.a) |
| Costs of diagnostic work-up € | cDcs | Triangular | 306.81 | ±20% (245.45–368.17) | Table 4 (online supplementary section 2.d) |
| Latency pending sputum culture result in smear-positive day | dCulture_SP | Normal | 10.32 | ±sd 8.9 (5.62–15.02) | Assessed (table 2) |
| Latency pending DST result in smear-positives day | dResistance_SP | Normal | 23.9 | ±sd 8.9 (15.0–32.8) | Assessed (table 2) |
| Latency pending sputum culture result in smear-negatives day | dCulture_SN | Normal | 13.17 | ±sd 5.15 (8.02–18.32) | Assessed (table 2) |
| Latency pending DST result in smear-negatives day | dResistance_SN | Normal | 29.2 | ±sd 7.2 (18.94–36.46) | Assessed (table 2) |
| Probability of MDR-TB in TB patients | pMDR_TB | Linear | 0.1966 | 0%/30% (0.0–0.3) | Calculated (table 1) |
| PPV of positive sputum smear | PPV_Smear | Linear | 0.9474 | ±95% CI (0.8814–0.9827) | Calculated (table 1) |
| NPV of negative sputum smear | NPV_Smear | Linear | 0.9199 | ±95% CI (0.8955–0.9402) | Calculated (table 1) |
| PPV of Xpert MTB/RIF in smear-positives | PPV_Xpert_SP | Linear | 1.0 | ±95% CI (0.9598–1.0) | Calculated (table 1) |
| PPV of Xpert MTB/RIF in smear-negatives | PPV_Xpert_SN | Linear | 0.8372 | ±95% CI (0.6930–0.9319) | Calculated (table 1) |
| NPV of Xpert MTB/RIF in smear-positives | NPV_Xpert_SP | Linear | 1.0 | ±95% CI (0.4782–1.0) | Calculated (table 1) |
| NPV of Xpert MTB/RIF in smear-negatives | NPV_Xpert_SN | Linear | 0.9772 | ±95% CI (0.9623–0.9878) | Calculated (table 1) |
| PPV of Xpert MTB/RIF irrespective of smear status | PPV_Xpertoverall | Linear | 0.9474 | ±95% CI (0.8946–0.9786) | Calculated (table 1) |
| NPV of Xpert MTB/RIF irrespective of smear status | NPV_Xpertoverall | Linear | 0.9747 | ±95% CI (0.9612–0.9879) | Calculated (table 1) |
| Probability that WHO standard regimen is effective | pMDR_Stand | Linear | 0.9272 | ±20% (0.741–1) | [11] |
| Time left to discharge from hospital days | dDischarge | Triangular | 26.78 | ±20% (21.24–32.14) | [6] |
| Costs of intrahospital contact investigation per TB index case € | cContact | Triangular | 105.81 | ±20% (84.65–126.97) | Adapted from [12] |
| Number of contacts to be investigated | pContact | Triangular | 10 | 5–20 | Assumption (online supplementary section 2.e) |
WHO: World Health Organization; MDR: multidrug-resistant; TB: tuberculosis; DST: drug susceptibility testing; PPV: positive predictive value; NPV: negative predictive value; GOÄ: Gebührenordnung für Ärzte (German medical fee schedule). #: in probabilistic sensitivity analysis.