Table 2. Model parameter estimates and values used for sensitivity analyses of intervention strategies for screening and treatment of latent TB infection in immigrants*.
| Parameter | Estimate | Range evaluated in PSA | PSA distribution | References |
|---|---|---|---|---|
| Screening parameters | ||||
| TST sensitivity | 0.782 | 0.69–0.87 | Beta(43,12) | (19) |
| TST specificity, no BCG | 0.974 | 0.963–0.982 | Beta(770,21) | (20,21) |
| TST specificity, BCG | 0.602 | 0.561–0.642 | Beta(239,158) | (20,21) |
| IGRA sensitivity | 0.889 | 0.688–0.993 | Beta(8,1) | (19) |
| IGRA specificity | 0.957 | 0.946–0.968 | Beta(900,40) | (20,21) |
| IGRA indeterminate† | 0.06 | 0.05–0.07 | Beta(83,1286) | (21) |
| Complete TST‡ | 1 | Fixed | Fixed | |
| Complete medical evaluation§ |
1 |
Fixed |
Fixed |
|
| Population characteristics¶ | ||||
| LTBI prevalence | ||||
| Very high incidence | 0.3162 | 0.2686–0.3880 | Varied with reactivation rate | (12,15–17) |
| High incidence | 0.2016 | 0.1706–0.2464 | Varied with reactivation rate | (12,15–17) |
| Moderate incidence | 0.0902 | 0.0763–0.1102 | Varied with reactivation rate | (12,15–17) |
| Low incidence | 0.0159 | 0.0135–0.0195 | Varied with reactivation rate | (12,15–17) |
| Abnormal chest radiograph results or previous TB | ||||
| Very high incidence | 0.039 | Fixed | Fixed | (15) |
| High incidence | 0.028 | Fixed | Fixed | (15) |
| Moderate incidence | 0.029 | Fixed | Fixed | (15) |
| Low incidence | 0.008 | Fixed | Fixed | (15) |
| Adherence to postarrival follow-up# |
0.684 |
0.646–0.721 |
Beta(404.50,186.87) |
(22) |
| Treatment parameters | ||||
| Initiate** | 0.938 | 0.907–0.964 | Beta(180.83,11.95) | (5) |
| Complete, INH | 0.616 | 0.561–0.670 | Beta(131.66,82.07) | (5) |
| Complete, RIF | 0.814 | 0.745–0.876 | Beta(76.85,17.56) | (5) |
| Adverse event, INH | 0.049 | 0.044–0.055 | Beta(249,4789) | (24,25) |
| Adverse event, RIF | 0.021 | 0.018–0.025 | Beta(109,4877) | (24,25) |
| Adverse event hospitalization | 0.01 | 0.0005–0.03 | Beta(1,99) | (25) |
| Death, INH | 0.00000988 | 0–0.00002 | Beta(2,202495) | (26) |
| LTBI risk reduction, INH | 0.90 | 0.78–0.95 | Normal(−2.3,0.5)†† | (27) |
| LTBI risk reduction, RIF | 0.90 | 0.63–0.97 | Normal(−2.3,0.8)†† | (28,24) |
| Partial risk reduction, INH | 0.346 | 0.267–0.490 | Combination of normal distributions††, ‡‡ | Expert opinion, (25) |
| Partial risk reduction, RIF | 0.30 | 0.17–0.40 | Normal(−0.35,0.1)†† | Expert opinion, (24,28) |
| Adverse event duration |
7 d |
0–24 |
Gamma(0.7,10) |
Expert opinion, (25) |
| TB parameters | ||||
| Death from TB | 0.0476 | 0.0391–0.0566 | Beta(76,1523) | (3) |
| Reactivation rate | 0.0011 | 0.0009–0.0013 | Beta(90.92,82545.55) | (15–17) |
| Abnormal CXR risk change | 3.9 | 3.0–4.9 | Normal(1.36,0.15)†† | (29) |
| Extended therapy | 0.124 | 0.029–0.264 | Beta(2.366,16.713) | Expert opinion, (30) |
| Relapse rate | 0.0359 | 0.0197–0.0654 | Normal(−3.327,0.365)†† | (30) |
| Hospitalization duration |
17 d |
Fixed |
Fixed |
Expert opinion, (30) |
| Model parameters | ||||
| BCG vaccination, <30 cases | 0.605 | 0.60–0.61 | Beta(45137,29502) | (12,13) |
| BCG vaccination, ≥30 cases | 0.998 | 0.997–0.999 | Beta(185381,384) | (12,13) |
| BCG vaccination uptake | 0.837 | Fixed | Fixed | (14) |
| Discount rate | 0.03 | Fixed | Fixed | (23) |
| Time horizon | 25 y | Fixed | Fixed | NA |
*AE, adverse event; BCG, bacillus Calmette-Guérin; IGRA, interferon-gamma release assay; INH, isoniazid; LTBI, latent tuberculosis infection; NA, not available; PSA, probabilistic sensitivity analysis; RIF, rifampin; TST, tuberculin skin test; TB, tuberculosis. †Treated as a negative result if it occurred; was equally likely to occur in those with and without LTBI. ‡Without being mandatory, this value is 63.5% (imputed from 43.4% completing screening when 68.4% adhere with a follow-up appointment) (5). §Without being mandatory, this value is 78% (imputed from 43.7 of 56 individuals completing medical evaluation) (5). ¶Very high incidence, >200 cases/100,000; high incidence, >100 and <200 cases/100,000; moderate incidence, >30 and <100 cases/100,000; low incidence, <30 cases/100,000. #From a meta-analysis (22); see also Appendix (https://wwwnc.cdc.gov/EID/article/25/4/17-1630-App1.pdf). **This model assumes all who report postarrival due to a positive preimmigration LTBI diagnostic test are offered treatment. Exploratory analysis adjusts this assumption so that only the number who would complete TST screening begin treatment. ††Results from this distribution are exponentiated. ‡‡Formula: 0.33 × (Normal(−1.168,0.228)) + 0.374 × (Normal(−0.381,0.169)) + 0.293 × 1.