Overall, the mean economic burden per episode was higher in upper- and middle-income countries (UMICs) than in low-income countries. However, this trend reversed when comparing the aggregated iNTS disease economic burden with the total health expenditure by country.

The proportion of the iNTS economic burden out of the total health expenditure was higher in sub-Saharan Africa than in other regions, meaning that, while the iNTS disease economic burden per incident was lower in sub-Saharan Africa, its relative impact on the country’s total health expenditure would be greater in sub-Saharan Africa compared with other regions. 

Compared to other diseases, the economic burden of iNTS is significant. For example, the cost of malaria ranged from $3.46–⁠$81.08 USD in Mozambique and $15.12–⁠$27.21 in Uganda, versus $92 USD and $125 USD per iNTS incident in Mozambique and Uganda, respectively.

The Importance of Economic Burden Estimates for Vaccine Development

The study’s findings fill data gaps about the economic burden of iNTS in low- and middle-income countries (LMICs), where the number of existing studies on the topic are disproportionately lower. This data gap impedes the development of a vaccine for iNTS, as LMICs’ underrepresentation in the available literature makes it difficult to evaluate the accurate societal impact of future vaccines in these countries, where the disease is highly prevalent.

According to the study, despite the high fatality risks of iNTS infection, there is no available vaccine for the disease. Understanding the economic burden of iNTS is an important step to make standardized comparisons among countries and assess the value of vaccines addressing iNTS.

Modeling and Methodology

The economic burden of iNTS was estimated using several multivariate regression models using data parameters derived from an existing systematic literature review. Specifically, through an econometric modeling framework, the economic burden of iNTS disease was estimated in countries where no primary data points were available; four different types of econometric models were constructed to identify the most suitable model to explain the variance of existing cost data and to estimate the economic burden of iNTS disease in 123 countries. A generalized linear model with gamma distribution with log link was chosen based on model diagnostics.

The model considered GDP per capita, life expectancy, population density, and iNTS. Among the set of covariates included in the final model, life expectancy was the most influential indicator in driving the estimated total cost apart from the binary iNTS variable.