The incidence of Hepatitis A (HA) has declined in most developed nations over the past century. Some countries, such as the United States, have instituted universal vaccination in regions of high disease burden and have seen further reductions in incidence. Other countries, such as Canada and most European countries, rely on targeted vaccination. Given a falling disease burden on the one hand and a safe and effective vaccine on the other, it is not clear whether universal vaccination should also be implemented low-endemicity countries. The purpose of this study is to evaluate the cost-effectiveness of universal HA vaccination in Canada.

Methods

It has been shown in principle that cost-effectiveness analysis can be more accurate if dynamic models are used, since dynamic models can account for herd immunity effects. For the present study, age-structured compartmental model was developed and parameterized using (1) clinical data on the course of HA infection, (2) a systematic review of HA seroprevalence studies in Canada, (3) catalytic modeling to estimate the true infection rate of HA in Canada by adjusting for subclinical infection and underreporting, and (4) survey data on infection rates due to travel in HA endemic countries. Costs were obtained from a systematic literature review and Ontario and Quebec administrative data. Vaccine strategy costs (eg, vaccine and vaccine administration), direct infection costs (eg, hospitalization, physician visits, liver transplants, death), and time costs (eg, cost of lost productivity) were estimated.

Results

Addition of a universal HA vaccination to the existing school-based HB program at the age of 9 via combined vaccine provides $52,000 per QALY for the payer perspective and $34,000 per QALY for the societal perspective, relative to no vaccination. One HA-only dose at age 4 in clinic followed by one combined HA/HB dose at age 9 in school provides ratios of $54,000 and $44,000 per QALY for payer and societal perspectives respectively. Results are most sensitive to vaccine costs. If herd immunity effects are not included, these cost-effectiveness ratios increase two- to three-fold.

Conclusion

Universal vaccination appears to be economically attractive in low endemicity countries relative to no vaccination. Definitive policy recommendations must wait until universal vaccination can be compared to the current targeted approach which appears to have been effective in reducing HAV disease burden.