Typhoid fevers are infections caused by the bacteria Salmonella enterica serovar Typhi (Salmonella Typhi) and Paratyphi A, B and C (Salmonella Paratyphi). Approximately 17.8 million incident cases of typhoid fever are estimated to occur annually in low- and middle-income countries and incidence is highest in children. This research aims to assess the comparative diagnostic accuracy of diagnostic tests for typhoid fever. We conducted a systematic literature to identify studies that compared diagnostic tests for typhoid fever in children to blood culture result. Diagnostic test network meta-analysis was performed building on models previously proposed by applying a Bayesian latent-class extension to the conventional network meta-analysis (NMA) model. We applied known diagnostic properties of bone marrow culture and the relationship between bone marrow and blood culture as informative priors to facilitate the network meta-analysis. We tested sensitivities for the proportion of negative blood samples that were false as well as bone marrow sensitivity and specificity. NMA was conducted on a subset of 26 studies (involving 47 comparisons) of child studies with blood culture reference tests from South Asia. The lateral flow IgG test performed comparatively well with 89% sensitivity (67% to 98% across scenario analyses) and 83% specificity. The most sensitive test of those investigated for the South Asian pediatric population was reverse passive agglutination with 96% sensitivity, however, scenario analyses yielded large model variability, with the lowest diagnostic sensitivity estimate being 66%. Adding a lateral flow-based IgG rapid test to either of two Typhidot test approaches yielded improvements in sensitivity without substantial declines in specificity. In the South Asian pediatric population, lateral flow IgG and reverse passive agglutination tests had high comparative diagnostic accuracy compared to other diagnostics. Combinations of rapid tests may provide a feasible option to increase diagnostic sensitivity and should be evaluated further.