Performance of rapid diagnostic tests for malaria diagnosis in Nigeria: a systematic review and meta-analysis.

BACKGROUND: Malaria remains a significant public health challenge in Nigeria, where diagnostic accuracy is essential for effective treatment and management. This systematic review and meta-analysis aimed to compare the diagnostic performance of malaria rapid diagnostic tests (RDTs) with that of microscopy, the gold standard for malaria diagnosis, in Nigeria. METHODS: A systematic search of Medline (PubMed), Scopus, Web of Science, and Google Scholar and other manually searched articles yielded 502 records. After removing duplicates and screening, 15 studies met the inclusion criteria for the meta-analysis. The risk of bias and applicability were assessed via QUADAS-2. A random effects model was used to pool sensitivity, specificity, positive and negative likelihood ratios (LR + , LR-), diagnostic odds ratios (DORs), and summary receiver operating characteristic (SROC) curves, with analysis conducted via Meta-Disc 1.4 and the metafor package (version 4.6-0) in R (version 4.4.1). Publication bias was assessed via funnel plots and Egger's test. Leave-one-out sensitivity analyses were conducted to assess the robustness of pooled estimates. RESULTS: The pooled sensitivity of RDTs was 69.7% [95% confidence interval (CI)]: 67.7-71.6%, and the specificity was 81.5% (95% CI 79.9-83.1%). The DOR was 21.87 (95% CI 7.45-64.21). The SROC curve was 0.902, though heterogeneity was high (I2 = 96.8%). Subgroup analysis showed regional variability, with better performance in urban settings. Funnel plot asymmetry and Egger's test suggested publication bias. Leave-one-out analyses confirmed robustness, with pooled estimates ranging 68-75% for sensitivity, 88-91% for specificity, and 17.7-27.6 for DOR. CONCLUSION: Malaria RDTs in Nigeria show suboptimal sensitivity and specificity compared with WHO benchmarks. While RDTs remain the most practical frontline tool due to their accessibility and ease of use, their limitations highlight the need for strengthened diagnostic algorithms that integrate RDTs with microscopy or molecular methods where feasible. Leave-one-out sensitivity analyses confirmed the robustness of the pooled estimates, reinforcing the validity of the conclusions. Advancing RDT technology, improving microscopy practices, and ensuring consistent study reporting are essential to enhance malaria diagnosis and control efforts.