Cost-effectiveness analysis of typhoid conjugate vaccines in an outbreak setting: a modeling study

Abstract

Background: Several prolonged typhoid fever epidemics have been reported since 2010 throughout eastern and southern Africa, including Malawi, caused by multidrug-resistant Salmonella Typhi. The World Health Organization recommends the use of typhoid conjugate vaccines (TCVs) in outbreak settings; however, current data are limited on how and when TCVs might be introduced in response to outbreaks.

Methodology: We developed a stochastic model of typhoid transmission fitted to data from Queen Elizabeth Central Hospital in Blantyre, Malawi from January 1996 to February 2015. We used the model to evaluate the cost-effectiveness of vaccination strategies over a 10-year time horizon in three scenarios: (1) when an outbreak is likely to occur; (2) when an outbreak is unlikely to occur within the next ten years; and (3) when an outbreak has already occurred and is unlikely to occur again. We considered three vaccination strategies compared to the status quo of no vaccination: (a) preventative routine vaccination at 9 months of age; (b) preventative routine vaccination plus a catch-up campaign to 15 years of age; and (c) reactive vaccination with a catch-up campaign to age 15 (for Scenario 1). We also explored variations in outbreak definitions, delays in implementation of reactive vaccination, and the timing of preventive vaccination relative to the outbreak.

Results: Assuming an outbreak occurs within 10 years, we estimated that the various vaccination strategies would prevent a median of 15-60% of disability-adjusted life-years (DALYs). Reactive vaccination was the preferred strategy for WTP values of $0-300 per DALY averted. For WTP values > $300, introduction of preventative routine TCV immunization with a catch-up campaign was the preferred strategy. Routine vaccination with a catch-up campaign was cost-effective for WTP values above $890 per DALY averted if no outbreak occurs and > $140 per DALY averted if implemented after the outbreak has already occurred.

Conclusions: Countries for which the spread of antimicrobial resistance is likely to lead to outbreaks of typhoid fever should consider TCV introduction. Reactive vaccination can be a cost-effective strategy, but only if delays in vaccine deployment are minimal; otherwise, introduction of preventive routine immunization with a catch-up campaign is the preferred strategy.

Keywords: Economic evaluation; Preventive vaccination; Reactive vaccination; Typhoid conjugate vaccines; Typhoid fever.

Fig. 1

Cost-effectiveness planes and acceptability frontiers. The cost-effectiveness planes (left) and cost-effectiveness acceptability frontiers (CEAFs; right) are plotted for A, B Scenario 1 (randomized outbreak timing), C, D Scenario 2 (no outbreak), and E, F Scenario 3 (outbreak has already occurred). In the cost-effectiveness planes, each dot represents the incremental costs (in 2020 USD) and DALYs averted for one simulation when compared with the base case of no vaccination. The bold Xs denote the expected additional cost and DALYs averted for each vaccination strategy with respect to no vaccination. Strategies are indicated by the color of the dot or X (purple: preventive routine vaccination; green: preventive routine vaccination plus a catch-up campaign up to age 15; or orange: reactive routine vaccination plus a catch-up campaign to age 15—for Scenario 1 only). In the CEAFs, the preferred strategy (i.e. the strategy that yielded the highest average net benefit) for each willingness-to-pay threshold ($0–1000 per DALY averted; x-axis, 2020 USD) is indicated by the color of the line (black: no vaccination; and same strategy colors as other panels), while the proportion of samples in which that strategy yielded the highest net benefit is indicated by the value on the y-axis; this can be interpreted as our certainty in the optimal strategy

Fig. 2

Cost-effectiveness acceptability frontiers for Scenario 1 with varying delays in reactive vaccination. The cost-effectiveness acceptability frontiers for Scenario 1 (randomized outbreak timing) are shown for a range of willingness-to-pay thresholds ($0–1000; x-axis, 2020 USD for A a 6-month delay in reactive vaccination after the outbreak threshold is exceeded, B a 12-month delay in reactive vaccination after the outbreak threshold is exceeded, and C a 24-month delay in reactive vaccination after the outbreak threshold is exceeded. The preferred strategy (i.e. the strategy that yielded the highest average net benefit) is indicated by the color of the line (black: no vaccination; purple: preventive routine vaccination; green: preventive routine vaccination plus a catchup campaign up to 15 years; or orange: reactive routine vaccination plus a catchup campaign), while the proportion of samples in which that strategy yielded the highest net benefit is indicated by the value on the y-axis (which can be interpreted as our certainty in the optimal strategy)