Cost-Utility Analysis of Universal Maternal Pertussis Immunisation in Thailand: A Comparison of Two Model Structures

Abstract

Objectives: This study aimed to assess the cost-effectiveness of introducing universal maternal pertussis immunisation under the national vaccine programme in Thailand.

Methods: We conducted a cost-utility analysis from a societal perspective to compare maternal vaccination with (1) TdaP vaccine, (2) Td vaccine and aP vaccine, and (3) Td vaccine only. We constructed two decision-tree models with Markov elements, each following a different clinical pathway, to allow us to examine structural uncertainty. Costs were converted to 2021 Thai Baht (THB) and a discount rate of 3% was applied to health and cost outcomes, with sensitivity analysis at 0% and 6%. Parameter uncertainty was investigated through deterministic and probabilistic sensitivity analysis, with expected value of perfect information analysis.

Results: Maternal pertussis vaccination would avert 27 cases and up to one death per year. The incremental cost-effectiveness ratio (ICER) for adding aP to the maternal immunisation schedule is 2,184,025 THB/QALY and the ICER for replacing maternal Td vaccination with TdaP is 3,198,101 THB/QALY. Maternal pertussis vaccination only becomes favourable in the probabilistic sensitivity analysis at cost-effectiveness thresholds above 6,000,000 THB/QALY, far above the Thai threshold of 160,000 THB/QALY. If incidence is less than 397 cases per 100,000, maternal pertussis vaccination will not be cost-effective in Thailand, within the plausible range for vaccine effectiveness and probability of hospitalisation. Budget impact is dominated by vaccination costs, which represent 12% and 18% of the 2021 national vaccine programme budget for introducing aP vaccine or for switching Td with TdaP vaccine, respectively.

Conclusions: We have found that maternal pertussis immunisation is not cost-effective in Thailand. Although there may be substantial under-reporting of pertussis cases, comparison with hospital data suggests that most under-reported cases are not hospitalised and therefore have negligible impact on our results. However, considerations such as affordability and local manufacturing may also be important for national immunisation programme decision-making.

Fig. 1

Decision tree model structure for model 1 (base model) (A) and model 2 (structural uncertainty analysis) (B)

Fig. 2

Incremental cost-effectiveness ratio (ICER) plane showing results from the probabilistic sensitivity analysis. Each dot represents a single run of the Monte Carlo simulation. Blue dots represent Td + aP and red dots represent TdaP. The crosses show the results from the deterministic analysis as a reference and the black dashed line represents the Thai cost-effectiveness threshold of 160,000 THB/QALY

Fig. 3

Cost-effectiveness acceptability curves showing the probability of each option (Td—black, Td + aP—blue, TdaP—red) being the most cost-effective option at different cost-effectiveness thresholds

Fig. 4

Tornado diagram for one-way deterministic sensitivity analysis, showing the parameters for which the one-way deterministic sensitivity analysis showed the greatest decrease in ICER