Four methods for estimating hepatitis C incidence using extant testing data

Area of Science:

• Epidemiology
• Public Health
• Infectious Disease Modeling

Background:

• Accurate hepatitis C (HCV) incidence estimation is vital for achieving World Health Organization (WHO) elimination targets.
• Current methods using HCV antibody (Ab) and RNA test data have poorly understood impacts of cohort specification and case definition on validity and generalizability.

Purpose of the Study:

• To evaluate the influence of different cohort specifications and case definitions on HCV incidence estimation using administrative health data.
• To compare four distinct methods for estimating HCV incidence in a large population-based cohort.

Main Methods:

• A cohort of 15.8 million Ontarians (aged 18-80) from 1999-2018 was constructed using linked ICES administrative health data.
• Four HCV incidence estimation methods were employed: population-based, test-negative, RNA-based, and antibody-inclusive, utilizing Poisson regression where applicable.

Main Results:

• HCV incidence estimates varied substantially by method, with RNA-based being lowest (approx. 30/100,000 PY) and test-negative highest (7.9-fold higher).
• Population-based estimates were sensitive to changes in HCV case definitions over time.
• The test-negative cohort's high prevalence of HIV and substance use disorder limited generalizability; RNA-based and antibody-inclusive methods had potential under- and overestimation issues, respectively.

Conclusions:

• Cohort and case definition significantly influence HCV incidence estimates derived from existing testing data.
• These findings are crucial for accurately measuring progress toward WHO HCV elimination goals.