Would universal antenatal screening for HIV infection be cost-effective in Australia?
Background: If cases of mother-to-child transmission of HIV infection are observed and there exists sufficient healthcare infrastructure to screen and supply optimal treatment, then the economics of universal antenatal screening should be explored. Existing cost-effectiveness analyses of universal antenatal HIV screening conducted for high income countries have either reported or assumed a prevalence of >0.02% among the antenatal population. We model the data for Australia, where prevalence is very low, to examine whether universal screening in this setting is cost-effective. Methods: A static decision model of the incremental cost-effectiveness of universal antenatal HIV screening as compared to existing practice was developed. A societal perspective was adopted, data were derived from secondary sources and different scenarios were tested by sensitivity analyses. Outcomes include: the incremental number of infections avoided and discounted-life-years gained; monetary valuations of the economic costs and benefits; and, the prevalence of HIV infection among the currently unscreened population where incremental economic cost and benefit are equalised. Results: The intervention is cost-effective at the prevalence 0.003257% with incremental costs exactly offset by the monetary valuation of incremental benefits; 5.17 new diagnoses of HIV are predicted; 1.29 infections would be avoided; and, 46.64 discounted-life-years would accrue. Applying favourable and unfavourable values for key variables suggest the prevalence at which the intervention is cost-effective lies in a range between 0.0016-0.0079%. Conclusions: We demonstrate that universal screening for HIV infection is cost-effective at very low prevalences; universal screening will generate net benefits under many scenarios included here; accurate statistics on the true prevalence of HIV in the currently unscreened antenatal population are required.
The Journal of Infectious Diseases 190 (1) 166-174 [doi:10.1086/421247]