The study team reported that targeted adult vaccination could produce large reductions in U.S. tuberculosis burden. Vaccinating all identified high-risk groups prevented 5,385 TB cases per year in an optimistic scenario, which the authors reported as a 51.8% reduction (95% uncertainty interval 35.2–66.8). Under a conservative scenario the same combined strategy prevented 1,348 cases annually, a 13.0% reduction (95% UI 6.4–19.8). The investigators also provided number needed to vaccinate (NNV) estimates: 366 in the optimistic scenario and 510 in the conservative scenario for the combined high-risk strategy. These figures quantify both impact and program efficiency that clinicians and policymakers can use to compare targeted approaches against broader vaccination programs.
Authors developed a deterministic compartmental transmission model to simulate Mtb infection, onward transmission, and progression to clinical TB across the United States. The model was stratified into four risk groups; the abstract names high-risk sets (persons living with HIV, non-U.S.-born persons, and persons with medical comorbidities) but does not list the fourth group explicitly in the abstract text. The team calibrated model outputs to 2024 U.S. TB surveillance data. The abstract does not report the model equations, parameter values, or the full definitions for each risk strata; those methodological details were not provided in the abstract.
Investigators ran two principal scenarios described as optimistic and conservative. They varied vaccine efficacy, duration of protection, and vaccination coverage across those scenarios. The abstract specifies only that those elements were varied; it does not supply the precise efficacy percentages, durations, coverage rates, or how those inputs map to each scenario. Readers should note that the reported case reductions and NNVs correspond specifically to the scenarios as constructed by the study team and are conditional on the unreported parameter choices.
The analysis estimated three primary outcomes: annual cases prevented, percent reduction in TB cases, and NNV. For the combined high-risk vaccination approach (including PLWH and non-U.S.-born persons), the optimistic scenario yielded 5,385 cases prevented per year and an NNV of 366. The conservative scenario produced 1,348 cases prevented per year and an NNV of 510. The authors presented uncertainty intervals for percent reductions (optimistic 35.2–66.8; conservative 6.4–19.8), indicating probabilistic or sensitivity-derived ranges around those point estimates. The abstract does not detail whether program costs, logistic constraints, or vaccine safety signals were included in the efficiency calculations; such considerations were not reported in the abstract.
The study team reported that strategy rankings were preserved across all sensitivity and uncertainty analyses. In other words, the ordering of which targeted strategies were most impactful and efficient remained consistent when the authors varied key model inputs. The abstract does not enumerate each sensitivity test or provide the ranking sequence beyond stating that strategies including non-U.S.-born persons and PLWH were the most efficient and impactful. The absence of full sensitivity matrices or tornado plots in the abstract means readers must consult the full paper for the detailed robustness checks and how alternate assumptions shifted absolute estimates.
The abstract makes clear high-level model outputs and comparative strategy performance but leaves several methodological and operational questions unanswered in the abstract itself. The authors did not report specific vaccine products, dosing regimens, age-based eligibility, exact definitions of the four modeled risk groups, or the numeric parameter values used to define optimistic versus conservative scenarios. The abstract also does not describe how the model handled reinfection risk, heterogeneity in transmission by locality, or the potential impact of programmatic barriers to vaccine uptake. Finally, while uncertainty intervals are provided for percent reductions, the abstract does not report uncertainty intervals for the NNV or for the absolute case counts in tabulated form. These gaps indicate the need to review the full text for the model code, parameter tables, and operational assumptions before applying results to policy design or implementation planning.
The study authors concluded that targeted adult vaccination strategies that include non-U.S.-born persons and PLWH are both efficient and impactful for reducing U.S. TB incidence. The abstract documents numeric scenario outcomes and robustness of rankings but omits the granular methodological inputs and implementation-level data that would be required for program rollout decisions. For those details and for any discussion of costs, safety, or specific vaccine candidates, consult the complete manuscript and supplementary materials.