Dokładność testów przesiewowych do wykrywania epilepsji w placówkach zdrowia pierwszego kontaktu i społeczności w krajach Afryki Subsaharyjskiej: przegląd systematyczny i plan meta-analizy

INTRODUCTION: Circumstantial evidence suggests that a high proportion of cases of epilepsy in countries across sub-Saharan Africa (SSA) remain undiagnosed. The magnitude of the burden is unknown. Screening tools offer promise for early detection and prevalence estimation that will enable evidence-informed management of epilepsy in SSA. This review will systematically assess the accuracy and reliability of screening tools for detecting epilepsy in communities and primary care settings in SSA. METHODS AND ANALYSIS: Relevant databases, non-database sources and grey literature will be searched for studies on epilepsy screening tools. PubMed, LILACS, CINAHL, PsycINFO and Google Scholar, from inception to 31 May 2026, will be searched for studies on screening tools (questionnaires) administered by non-expert physicians to populations or hospital/clinic-based cohorts with no language restrictions. The following search terms will be used: screening tool, screening questionnaire, screening test, screening instrument, diagnostic tool, diagnostic accuracy, epilepsy, sensitivity, specificity, true positive, false positive, true negative and false negative and SSA. All countries in SSA will be included as search terms. Cochrane databases, African Journals Online, African Index Medicus, HINARI and Preprint and Thesis repositories will also be searched. Reference lists of potentially relevant studies will be reviewed, and experts will be contacted to identify additional studies missed in our searches. Study selection (using a pretested study selection flow chart), data extraction (using a validated data extraction form) and risk-of-bias assessment (using the revised Quality Assessment of Diagnostic Accuracy Studies-2) will be performed independently by at least two reviewers, and any discrepancies will be resolved through discussion. The pooled sensitivity, specificity and diagnostic odds ratio (DOR) will be estimated from 2-by-2 tables of true positives, false positives, true negatives and false negatives. Possible causes of heterogeneity between studies will be assessed through pre-specified subgroup analyses. A meta-analysis will be conducted using a bivariate random-effects model to summarise sensitivity, specificity and DOR per patient. A summary receiver operating characteristic curve will be plotted to determine the overall diagnostic performance of the index tests. Sensitivity analyses will be conducted to test the robustness of pooled estimates of screening accuracy, and all estimates will be presented with their 95% CIs. ETHICS AND DISSEMINATION: This study will synthesise empirical evidence from publicly available published and unpublished studies, and hence no ethical approval is required. An eligible study with serious ethical issues will be excluded from the analysis and the reasons for exclusion will be documented. The review findings will be shared with all relevant stakeholders, including healthcare providers, patient advocate groups, agencies involved in implementing epilepsy care and policies, civil society, social services providers and researchers. The review findings will be shared widely at scientific symposia and conferences, and the final report will be published in a high-impact-factor peer-reviewed journal. PROSPERO REGISTRATION NUMBER: CRD42024566976.