National burden of and optimal vaccine policy for Japanese encephalitis virus in Bangladesh: a seroprevalence and modelling study - 19/11/25

Doi : 10.1016/S1473-3099(25)00590-0

Mariana Perez Duque, MD MPH ^a, Kishor K Paul, PhD ^b,^c, Rebeca Sultana, PhD ^b,^d, Gabriel Ribeiro dos Santos, PhD ^a, Megan O’Driscoll, PhD ^a, Abu M Naser, PhD ^b,^e, Mahmudur Rahman, ProfessorPhD ^f, Mohammad Shafiul Alam, PhD ^b, Hasan M Al-Amin, PhD ^b,^g, Mohammed Z Rahman, PhD ^b, Mohammad E Hossain, PhD ^b, Repon C Paul, PhD ^h, Elias Krainski, PhD ⁱ, Stephen P Luby, ProfMD ^j, Simon Cauchemez, ProfPhD ^k, Jessica Vanhomwegen, PhD ^l, Emily S Gurley, ProfPhD ^m,^{^{†
Joint senior authors}}, Henrik Salje, ProfPhD ^a,^{^{†
Joint senior authors},}⁎
^a Department of Genetics, University of Cambridge, Cambridge, UK
^b International Centre for Diarrhoeal Diseases Research, Bangladesh (icddr,b), Dhaka, Bangladesh
^c School of Population Health, University of New South Wales, Sydney, Australia
^d University of California, Berkeley, CA, USA
^e Division of Epidemiology, Biostatistics, and Environmental Health, School of Public Health, University of Memphis, Memphis, TN, USA
^f Institute for Epidemiology, Disease Control and Research (IEDCR), Dhaka, Bangladesh
^g School of the Environment, University of Queensland, Brisbane, QLD, Australia
^h Centre for Big Data Research in Health, University of New South Wales, Sydney, NSW, Australia
ⁱ Statistics Program, CEMSE Division, King Abdullah University of Science and Technology, Thuwal, Saudi Arabia
^j Division of Infectious Diseases and Geographic Medicine, Stanford University, Stanford, California, USA
^k Mathematical Modelling of Infectious Diseases Unit, Institut Pasteur, Université Paris Cité, INSERM U1332, CNRS UMR2000, Paris, France
^l Emerging Diseases Unit, Institut Pasteur, Paris, France
^m Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, USA

^*Correspondence to: Prof Henrik Salje, Department of Genetics, University of Cambridge, Cambridge CB2 3HE, UKDepartment of GeneticsUniversity of CambridgeCambridgeCB2 3HEUK

Sous presse. Épreuves corrigées par l'auteur. Disponible en ligne depuis le Wednesday 19 November 2025

Summary

Background

Bangladesh first reported Japanese encephalitis virus (JEV) in 1977 and has seen regular cases since; however, no JEV vaccination programme currently exists. A barrier to the use of JEV vaccines has been a limited understanding of the underlying burden.

Methods

We conducted a nationally representative serological community study in 70 communities in individuals of all ages from October, 2015, to January, 2016. Serum samples obtained from the community study were tested for IgG antibodies against JEV. We developed spatially explicit binomial regression models to estimate the underlying force of infection across the country, and to identify risk factors associated with infection. We then used mathematical models to estimate the annual Japanese encephalitis disease burden currently and under different vaccination strategies, varying the geographical distribution, target age cohorts, and vaccination coverage.

Findings

Of the 2938 participants included in the serological study, 100 individuals were seropositive for JEV (overall seroprevalence 3·4% [95% confidence interval 2·8–4·1]; range 0–28% across communities). The annual probability of infection was 0·005 (95% credible interval [CrI] 0·003–0·009), with risk being greatest near border regions. We estimated that annually there are 157 000 JEV infections (95% CrI 88 000–261 000), resulting in 157 severe cases (89–253) and 31 deaths (18–52). A vaccination strategy in the ten most affected districts with a catch-up campaign with 60% coverage of people aged 1–15 years would require 5 million doses and avert 0·9 cases per 100 000 doses over 5 years, compared with 35 million doses and 0·5 cases averted for a nationwide campaign. No vaccination scenario was cost-effective under a willingness-to-pay of 3-times gross domestic product.

Interpretation

A spatially targeted vaccine campaign would be most effective in reducing JEV burden; however, it would still not meet standard cost-effectiveness targets.