Prevalence of death in people with vision impairment from cataracts before treatment: a case study from Kenya

BACKGROUND: Cataract remains the leading cause of blindness globally, which substantially affects quality of life and economic productivity. Despite being a highly cost-effective intervention, cataract surgery remains inaccessible to many, especially in low-resource settings. This study presents a dynamic model that aims to estimate the number of people who will die before receiving cataract surgery, using Kenya as a case study. METHODS: We developed a dynamic simulation model to project the national cataract backlog, surgical interventions, and mortality over a 50-year period (1990-2040). The model integrates demographic and epidemiological data, alongside key parameters including age-specific cataract incidence, severity progression, mortality risk, and surgical throughput. Sensitivity analysis was done to estimate the effect of different cataract surgical rates. FINDINGS: At current surgical capacity, the model estimates that 280 400 (77%) of 360 000 individuals on Kenya's cataract backlog in 2025 will die before receiving surgery, with 236 400 (66%) dying before 2030. Sensitivity analysis shows that doubling cataract surgical rates could enable an additional 24 000 people to receive treatment before death, representing a 16% reduction in untreated mortality. A ten-fold increase (cataract surgical rates of 7020 surgeries per million people per year) would nearly eliminate deaths among those awaiting surgery. INTERPRETATION: This model provides a comprehensive view of the national cataract burden by incorporating incidence, surgical capacity, and mortality estimates for untreated cases. It underscores the urgent need for expanded cataract surgery capacity and improved access to care. The model offers actionable insights for policy makers and health system planners aiming to reduce avoidable blindness and prevent premature deaths from treatable conditions. FUNDING: The Wellcome Trust and Fred Hollows Foundation.