A new study reveals a potential blind spot in the methods doctors use to predict glaucoma risk, particularly for Black patients.
Glaucoma risk models, which primarily rely on data from European and Asian populations, are less accurate at identifying the risk for the eye disease among Black individuals, according to the study in JAMA Ophthalmology.
The limitation could mean patients miss out on treatments to prevent vision loss.
Analyzing over 78,000 patients from the United States and four African countries, researchers examined how well widely used polygenic risk scores predict the risk of glaucoma in African ancestry groups compared to European groups. For Black patients, these scores consistently underperformed.
“These predictive gaps mean that Black patients may not be accurately classified into high-risk categories for primary open angle glaucoma,” said co-first study author Jennifer M. Chang-Wolf, MD, a Duke University School of Medicine graduate who conducted research as her third year project with the Duke Department of Ophthalmology. “This can result in fewer early interventions or preventative strategies for those who need them most.”
People of African descent face higher risks and worse outcomes of glaucoma, a leading cause of blindness, and yet glaucoma risk models are calibrated with genetic data from Europeans and Asians, overlooking the unique genetic factors in other groups.
The study published Nov. 14 illustrates the consequences. Researchers evaluated three risk models—Gharahkhani, Han, and Craig—by analyzing odds ratios and accuracy metrics like the area under the receiver operating characteristic curve (AUROC) score.
While these tools effectively identified high-risk individuals of European ancestry, with odds ratios reaching as high as 7.24, the odds ratios were significantly lower for Ghanaians or African Americans, reflecting weaker predictions.
“Our study demonstrates the limitations of using models primarily developed for European populations,” said Pieter Bonnemaijer, MD, PhD, co-senior author from the Department of Epidemiology at Eramus Medical Center in the Netherlands. “Without more accurate and inclusive risk assessment tools, people of African ancestry may not equally benefit from new advances in medical research that use genetic data to predict glaucoma risk.”
The gap in accuracy goes beyond numbers and reflects a missed opportunity to treat glaucoma early and to help protect eyesight. The shortfall also risks compounding existing health disparities as people of African descent are four times more likely to experience blindness from glaucoma than those of European descent.
Outreach, education, and screening programs can help detect the disease sooner, while addressing medical costs and adherence to treatment might improve outcomes, but growing evidence shows genetics play a major role in disparities in glaucoma.
However, exploring these biological differences will require including a more diverse group of patients in genomics studies.
“If we are to close the gap in health disparities, we need more inclusive research that leads to accurate, representative tools,” Bonnemaijer said.
Study researchers include members of the Genetics in Glaucoma Patients of African Descent Study Group, Genetics of Glaucoma in People of African Descent Study Group, and the Million Veteran Program, one of the largest research programs in the world studying genes and health.
Chang-Wolf, an ophthalmology PhD candidate in the Netherlands, led the analysis as a Duke medical student studying abroad, and collaborated with colleagues at Case Western University and East Carolina University Tyler G. Kinzy, co-first author, and Jessica N. Cooke Bailey, PhD, co-senior author, on the study.
The study was funded by the Bright Focus Foundation, Combined Ophthalmic Research Rotterdam Foundation, Rotterdamse Stichting Blinderbelangen, the Landelijke Stichting voor Blinden en Slechtzienden, the Nederlandse Vereniging ter Voorkoming van Blindheid, with additional support provided by the National Institutes of Health and Veterans Health Administration.