Mass hepatitis C virus screening among persons in custody in Hong Kong: a modelling cost-effectiveness analysis based on a prospective cohort study.

BACKGROUND

Persons in custody (PICs) have a disproportionately high rate of hepatitis C virus (HCV) infection, yet correctional elimination programmes remain scarce. We evaluated the cost-effectiveness of different prison-targeted tailored strategies.

METHODS

Using prison-based cohort data, we modelled 10,000 PICs via a decision tree and age-stratified Markov framework to compare one-off screening or periodic screening, against no screening. Each ran under hepatologist-led teleconsultation, outreach face-to-face consultation (on-site visits) and hospital-based consultation. Costs and quality-adjusted life years (QALYs) were estimated from a societal perspective, with incremental cost-effectiveness ratios (ICERs) benchmarked to the per-capita GDP (USD 54,134).

FINDINGS: Compared with no screening, one-off screening strategies reduced HCV-related compensated cirrhosis by 0.4-5.0%, decompensated cirrhosis by 0.6-6.8%, hepatocellular carcinoma by 0.4-5.2%, and HCV-related mortality by 0.4-3.7%. Mean costs increased from US $1326 to US $1375 and QALYs from 7.536 to 7.554 per person. ICERs for one-off strategies ranged from US $2680 to 2920 per QALY. Triennial and annual screening yielded higher health gains but with higher ICERs of US $7837 and 15,370 per QALY. Results were robust across care delivery models, prevalence scenarios, and sensitivity analyses.

INTERPRETATION: Prison-based HCV screening with one-time screening strategies is cost-effective in a metropolitan setting. Periodic screening can further reduce disease burden but requires a higher commitment of the healthcare budget. Integration of telemedicine-supported, prison-based screening within the national hepatitis elimination strategy could accelerate HCV micro-elimination and deliver a substantial public health impact.

FUNDING: Health Bureau, Government of HKSAR under Health and Medical Research Fund (reference: 20211881).