A new general biomarker-based incidence estimator

Area of Science:

• Epidemiology
• Biostatistics
• Public Health

Background:

• Estimating disease incidence using recent infection biomarkers in cross-sectional surveys is valuable.
• Current methods for handling biomarker data, especially for HIV, lack consensus, particularly for long-term infection classification.

Purpose of the Study:

• To develop a generalizable method for estimating recent disease incidence.
• To overcome limitations of existing estimators that rely on specific assumptions about biomarker dynamics and population context.

Main Methods:

• Derived a general expression for a weighted average of recent incidence.
• Introduced an explicit timescale (T) to truncate the averaging period, removing assumptions about biomarker dynamics and epidemiologic context.

Main Results:

• The estimator's dynamics depend on two parameters: mean duration of recent infection and false-recent rate.
• Identified dimensionless parameters to quantify bias and assess estimator utility based on test performance and population state.
• Simulated data demonstrated accurate estimation of test characteristics and incidence with coverage within 1% of intended levels.

Conclusions:

• Biomarker-based incidence estimation can be adapted to general contexts.
• The new method avoids strong assumptions required by previous approaches regarding biomarker dynamics and historical data.