Work

Comparison of two Crimean-Congo haemorrhagic fever virus serological tests in Tanzanian febrile patients using a Bayesian latent class approach

Year: 2026

Type: article

Abstract: Background: Crimean-Congo haemorrhagic fever virus (CCHFV) is a tick-borne zoonotic disease that can cause severe disease in humans, has a high fatality ratio, and is present in many African countries. Accurate diagnosis of CCHF remains difficult. Serological and molecular tests are frequently used to detect exposure or infection, but their accuracy varies with viral strain, infection stage, and population factors. Notably, there is no universally recognised gold standard test for CCHFV serological diagnosis. The reference standard for antibody detection requires neutralisation assays, which must be performed in biosafety level 4 (BSL-4) laboratories. These facilities are scarce and largely unavailable in sub-Saharan Africa, including Tanzania, where CCHFV is endemic. This study aimed to evaluate the diagnostic performance of the IDvet ELISA and Luminex serological assays for detecting prior CCHFV exposure in humans using Bayesian latent class modelling in the absence of a reference standard. Methodology: This was a secondary analysis of archived serum samples from a prospective febrile illness surveillance cohort conducted between 2012-2014 in Moshi, Tanzania, at Kilimanjaro Christian Medical Center and Mawenzi Regional Referral Hospital. Bayesian latent class analysis was used to evaluate the performance of IDvet and Luminex assays in detecting CCHFV exposure in the samples collected. Results: Of the 870 febrile admissions enrolled across both sites, serum samples were available for 427 participants (49%). The median posterior estimate for IDvet sensitivity was 0.893 (95% Bayesian credibility interval (BCI) = 0.555-0.996) and specificity was 0.972 (95% BCI = 0.936-0.996). For the Luminex assay, the median posterior estimate for sensitivity was 0.884 (95% BCI = 0.541, 0.995) and specificity was 0.987 (95% BCI = 0.947, 1.00). The model generated seroprevalence estimates for two subpopulations, livestock owners and non-livestock owners. The livestock owner seroprevalence estimate was 0.027 (95% BCI = 0.002, 0.069), and for non-livestock owners, 0.017 (95% BCI = 0.001, 0.044). Conclusion: In the absence of a reference standard and access to BSL-4 neutralisation testing, both the IDvet ELISA and the Luminex assay demonstrated high diagnostic performance with comparable sensitivity in detecting prior exposure to CCHFV among febrile patients in northern Tanzania. Furthermore, the IDvet may offer operational advantages because of its simplicity and suitability for decentralised laboratories.

Source: Frontiers in Microbiology

DOI: 10.3389/fmicb.2026.1894712