Quality assurance of serologic testing for Chagas disease in a primary care setting of rural Paraguay

•ELISA kits to detect T. cruzi in different settings showed good agreement.•Performance of ELISA kits in rural areas could be affected by operator training.•The RDT showed a performance similar to an algorithm including an ELISA test.•More evidence is required before recommending the use of an RDT algorithm to confirm diagnosis. The diagnosis of Chagas disease mostly relies on the use of multiple serologic tests that are often unavailable in many of the remote settings where the disease is highly prevalent. In the Teniente Irala Fernández Municipality, in central Paraguay, efforts have been made to increase the diagnostic capabilities of specific rural health centres, but no quality assurance of the results produced has been performed. We comparatively analysed the results obtained with 300 samples tested using a commercial rapid diagnostic test (RDT) and enzyme linked immunosorbent assays (ELISA) at the laboratory of the Teniente Irala Fernández Health Center (CSTIF) with those generated upon repeating the tests at an independent well-equipped research laboratory (CEDIC). A subgroup of 52 samples were further tested at Paraguay's Central Public Health Laboratory (LCSP) by means of a different technique to evaluate the diagnostic performance of the tests carried out at CSTIF. We observed an excellent agreement between the ELISA results obtained at CSTIF and CEDIC (kappa coefficients between 0.85 and 0.93 for every kit evaluated), and an overall good performance of the tests carried out at CSTIF. However, the sensitivity of one kit was lower at CSTIF (81.3 %) than at CEDIC (100 %). The individual use of an RDT to detect the infection at CSTIF showed a similar sensitivity to that obtained combining it to an ELISA test (92.3% vs 88.5, p = 1). Nonetheless, the generalizability of this result is yet limited and will require of further studies. [Display omitted]