Lassa fever outbreaks, mathematical models, and disease parameters: a systematic review and meta-analysis

Abstract

BackgroundLassa fever, caused by Lassa virus (LASV), poses a significant public health threat in West Africa. Understanding the epidemiological parameters and transmission dynamics of LASV is crucial for informing evidence-based interventions and outbreak response strategies.MethodsWe conducted a systematic review (PROSPERO CRD42023393345) to compile and analyse key epidemiological parameters, mathematical models, and past outbreaks of LASV. Data were double extracted from published literature, focusing on past outbreaks, seroprevalence, transmissibility, epidemiological delays, and disease severity.FindingsWe found 157 publications meeting our inclusion criteria and extracted 374 relevant parameter estimates. Although LASV is endemic in West Africa, spatiotemporal coverage of recent seroprevalence estimates, ranging from 0.06% to 35%, was poor. Highlighting the uncertainty in LASV risk spatially. Similarly, only two basic reproduction number estimates at 1.13 and 1.19 were available. We estimated a pooled total random effect case fatality ratio of 33.1% (95% CI: 25.7 – 41.5, I2= 94%) and found potential variation in severity by geographic regions typically associated with specific LASV lineages. We estimated a pooled total random effect mean symptom-onset-to-hospital-admission delay of 8.3 days (95% CI: 7.4 – 9.3, I2 = 92%), but other epidemiological delays were poorly characterised.InterpretationOur findings highlight the relative lack of empirical LASV parameter estimates despite its high severity. Improved surveillance to capture mild cases and approaches that integrate rodent populations are needed to better understand LASV transmission dynamics. Addressing these gaps is essential for developing accurate mathematical models and informing evidence-based interventions to mitigate the impact of Lassa fever on public health in endemic regions.FundingUK Medical Research Council, National Institute for Health and Care Research, Academy of Medical Sciences, Wellcome, UK Department for Business, Energy, and Industrial Strategy, British Heart Foundation, Diabetes UK, Schmidt Foundation, Community Jameel, Royal Society, and Imperial College London.Research in ContextEvidence before this studyWe searched PubMed up to August 2, 2023 for ((lassa fever) or (lassa virus)) and (epidemiology or outbreak or (models not image) or transmissibility or severity or delays or (risk factors) or (mutation rate) or seroprevalence). We found ten systematic reviews. Three on ribavirin as a Lassa fever treatment, two on Lassa virus (LASV) vaccine candidates, and one each on historical importations of Lassa fever cases from West Africa to non-endemic countries, clinical characteristics for protocol development, and Lassa fever in pregnancy. Two systematically reviewed epidemiological parameters. One on basic reproduction number estimates which ranged from 1.1 to 1.8 for human-to-human and 1.5 to 1.7 for rodent-to-rodent transmission. However, no meta-analyses were conducted. The other focused on LASV infection case fatality ratios (CFRs): 29.7% (22.3–37.5) in humans and prevalence: 8.7% (95% confidence interval: 6.8– 10.8) in humans, 3.2% (1.9–4.6) in rodents, and 0.7% (0.0–2.3) in other mammals. There were no systematic reviews on LASV transmission models.Added value of this studyWe provide a comprehensive overview of published outbreaks, transmission models and epidemiological parameters for LASV. We highlight the sparsity of key epidemiological parameter estimates such as the serial interval or generation time. The discrepancy between the high overall severity and the high seroprevalence in the general population suggests a high proportion of infections are asymptomatic or only result in mild disease. Therefore, current surveillance systems may need refining to better characterise LASV transmission dynamics.Implications of all the available evidenceEpidemiological models are useful tools for real-time analysis of outbreaks, assessing epidemic trajectories and the impact of interventions. Our study is a useful basis to inform future LASV models, but highlights uncertainties and knowledge gaps that need to be filled in LASV transmission and natural history. Future LASV studies will benefit from integrating human and rodent reservoir surveillance.