Beyond prevalence: diagnostic heterogeneity in population-based glaucoma studies in Africa-a systematic review

The burden of glaucoma is disproportionately high in Africa. While numerous population-based studies have reported prevalence estimates, little attention has been given to the diagnostic criteria underpinning these findings.Objective: This review sought to systematically examine the diagnostic criteria used to define glaucoma in population-based studies in Africa, and to synthesize corresponding estimates of glaucoma prevalence.

A systematic search of PubMed, Scopus, Google Scholar, and African Journals Online identified population-based studies on glaucoma prevalence in Africa. Diagnostic approaches were classified by adherence to the International Society of Geographical and Epidemiological Ophthalmology (ISGEO) criteria. The primary outcome was the diagnostic framework used, with secondary outcomes including prevalence estimates for major glaucoma subtypes.

Nine studies met the inclusion criteria. The use of ISGEO definitions varied considerably across the studies. Key sources of heterogeneity included intraocular pressure thresholds, optic nerve head assessment, use of visual field testing, and gonioscopy application. Studies applying the full ISGEO definitions generally had a trend of reported prevalence estimates within a narrower range (4.3% to 5.3%), whereas studies employing partial ISGEO or non-ISGEO diagnostic approaches reported a broader range of prevalence estimates (2.1% to 8.4%). The proportion of previously diagnosed glaucoma cases was low, ranging from 3.3% to 14.3% among studies that reported previously diagnosed glaucoma rates (n = 5).

Diagnostic heterogeneity remains a major limitation to the comparability of glaucoma prevalence estimates across Africa. Standardized reporting of diagnostic protocols, together with clear justification of the diagnostic approaches employed, is essential to improve the interpretation and comparability of future epidemiological studies while recognizing the practical constraints of resource-limited settings.

[https://www.crd.york.ac.uk/PROSPERO/view/ CRD42024592160], identifier [CRD42024592160].