Incidence estimation using a single cross-sectional age-specific prevalence survey with differential mortality

Area of Science:

• Epidemiology
• Biostatistics
• Public Health

Background:

• Estimating disease incidence is crucial for public health planning, particularly for curable conditions like cataract.
• The VISION2020 initiative aims to eliminate avoidable blindness, requiring accurate data on cataract burden and surgical needs.
• Traditional cohort studies for incidence estimation are costly and time-consuming, necessitating alternative methodologies.

Purpose of the Study:

• To present a novel method for estimating the incidence of a curable, non-recurring disease using cross-sectional survey data and mortality rates.
• To extend the illness-death model to a four-state model accommodating disease cure (e.g., cataract surgery).
• To apply the developed method to predict disease burden under various surgical strategies, considering mortality risks.

Main Methods:

• Utilized data from a cross-sectional survey of approximately 10,000 individuals.
• Employed an extended time-homogeneous illness-death model, transitioning to a four-state model to include disease cure.
• Incorporated population-level mortality rates, assuming differential mortality between diseased and non-diseased individuals.

Main Results:

• Developed a method for incidence estimation applicable to curable diseases with limited data.
• Demonstrated the utility of the four-state model for predicting future disease burden and surgical needs.
• Applied the methodology to a Nigerian visual impairment survey dataset.

Conclusions:

• The proposed method offers a feasible approach for incidence estimation of curable diseases using readily available data.
• The four-state model provides a robust framework for disease burden projection and intervention planning.
• Accurate incidence and burden estimates are vital for achieving global health targets like VISION2020.