Typhoid vaccine schedule depends on incidence, waning, and fatality risk
A multi-model analysis for typhoid conjugate vaccine scheduling has pushed the debate beyond the question of whether vaccination works and into the more difficult question of when and how to deploy it for maximum benefit. Across four transmission models, the optimal program changed with local epidemiology, the pace of immune waning, and the economic assumptions used to value prevented illness and death.
The central message is blunt. A single vaccination schedule does not fit every endemic setting.
Typhoid fever remains a major problem in low- and middle-income countries, where fecal-oral transmission persists in the setting of inadequate water and sanitation infrastructure. The burden falls heavily on children, especially those younger than 15 years, and in high-incidence settings the peak can occur as early as ages 2 to 4 years. Antimicrobial resistance adds another layer of urgency, because weakening antibiotic efficacy can erode the gains from treatment and make prevention more valuable than ever.
Against that backdrop, investigators compared two agent-based and two compartmental dynamic models of typhoid transmission under harmonized age-specific incidence data from medium, high, and very high incidence settings. The analysis was built to inform World Health Organization deliberations on updated vaccine recommendations, including whether a booster dose should be added to routine schedules.
Routine vaccination at 9 months performed best overall, but not everywhere
Over a 10-year horizon, the strategy that combined routine vaccination at 9 months with a catch-up campaign to age 15 years and a booster dose at 5 years produced the largest predicted public health impact across settings. Across the models, that approach reduced typhoid cases by a median of 40% to 68%, depending on the incidence setting.
The details matter because the same vaccine can look very different once transmission intensity changes. In very high incidence settings, routine vaccination at 9 months or at 2 years of age was cost-saving, and adding a booster at 5 years was cost-effective at most willingness-to-pay thresholds across all scenarios. In medium incidence settings, the addition of a catch-up campaign was cost-effective above a willingness-to-pay threshold of $1,250 per disability-adjusted life-year averted when illness costs and case fatality risk were high, and in high incidence settings when those same assumptions were low.
Age at vaccination also shifted with waning. If vaccine protection declined quickly, the optimal strategy was to delay vaccination to 2 or 5 years of age, depending on the age at which typhoid incidence peaked. That pattern reflects a basic immunization tradeoff: vaccinating too early can leave a vulnerable window if protection fades before children reach the years of highest exposure, while vaccinating later may miss early childhood disease in settings where risk begins sooner.
Waning immunity changes the value of a booster dose
Emerging evidence that protection from typhoid conjugate vaccines may wane after 5 years gives this analysis its practical edge. Medium-term efficacy data from Malawi and Bangladesh have already hinted that duration of protection is not uniform across settings, and that local force of infection may influence how long immunity remains clinically useful. In the models, that uncertainty was not a side issue; it determined whether a booster dose added enough benefit to justify the programmatic burden.
In settings where waning is slow, early routine vaccination remains attractive because protection persists through much of the high-risk period. In settings where waning is fast, a schedule anchored too early can leave older children and adolescents exposed after immunity declines. The booster dose at 5 years is meant to close that gap, especially where burden continues well beyond infancy and toddlerhood.
The models were generally aligned on the direction of effect, which strengthens confidence in the broad conclusions. Differences in model structure did not erase the core pattern: incidence, fatality risk, and waning rate all pulled the preferred schedule in different directions.
Cost-effectiveness shifted with local burden and the price of disease
Economic results tracked the epidemiology. The same program that was clearly worthwhile in one context could be less compelling in another, depending on the willingness-to-pay threshold and the assumptions used for treatment costs and death risk. The analysis used a net-monetary-benefit framework to compare health and economic impact under two representative scenarios for vaccination and treatment costs.
This is where typhoid control becomes a policy problem rather than only a vaccine problem. When treatment costs are high, case fatality risk is high, or both, prevention carries more economic value because each avoided case prevents downstream spending and loss of life. When those inputs are lower, vaccine schedules still may be cost-effective, but the margin narrows and the preferred age at vaccination can shift.
The catch-up campaign also changed the calculus. Reaching children already beyond infancy substantially increased the modeled impact of introduction, which is especially relevant in endemic settings where a large share of disease is already occurring before a routine infant schedule can fully protect the population.
Why the models converged despite different structures
Two agent-based and two compartmental models are not identical tools. They represent transmission in different ways, and each carries assumptions about environmental exposure, chronic carriage, and age-specific contact patterns. Yet the convergence across models is meaningful because it suggests the results were not driven by a single analytic framework.
Uniform input data helped. All four models were fitted to harmonized age-specific incidence data from medium, high, and very high incidence settings, which gave the comparison a common foundation. Standardization does not eliminate uncertainty, but it reduces the chance that differences in outputs are simply artifacts of mismatched inputs.
The remaining uncertainty sits where vaccine policy always becomes difficult: indirect protection, waning, and simplifying assumptions about transmission dynamics. Sparse data on indirect protection limit how precisely the herd effects of vaccination can be estimated. In practical terms, the estimated benefit of a schedule may be conservative if transmission falls more than the model can fully capture, or overly optimistic if protection wanes faster than assumed.
What program planners can take from the analysis
The study points toward a settings-based vaccination policy rather than a single universal schedule. Routine vaccination at 9 months with a catch-up campaign looks strongest overall, but the best option depends on local typhoid incidence, case fatality risk, and the pace at which vaccine protection fades.
In very high incidence settings, the economics support action even more strongly, with 9-month or 2-year introduction remaining cost-saving and a 5-year booster often adding value. In medium incidence settings, the case for introduction strengthens when disease is severe or costly. In settings where waning is fast, delaying the first dose to 2 or 5 years may produce a better fit with the age pattern of risk.
The policy implication is straightforward: vaccine schedules should be matched to local typhoid epidemiology, not copied wholesale from one country to another. The analysis supports use of typhoid conjugate vaccines in most endemic settings, but it also shows that booster timing and the age of routine vaccination can change the balance of benefit and cost enough to alter the preferred program. The unresolved issue is not whether vaccination has value, but how quickly each endemic setting can match its schedule to the local pattern of risk, waning, and affordability.