Introduction: Health care costs are on the rise in Australia and rational decisions are necessary to prioritise interventions for the allocation of limited resources. As new vaccines are being introduced, budgetary allocations for immunisation are increasing. Economic analyses are useful in prioritising immunisation choices by determining the cost-effectiveness of various interventions. Cost-effectiveness studies on immunisation are scarce and needs updating due to changes in medical practice and vaccination costs in Australia. This thesis describes cost-effectiveness of two selected immunisation interventions in Australia: influenza and hepatitis B. The cost-effectiveness of reducing the influenza vaccination age threshold to 50 from the current 65 was analysed first. Next, the cost-effectiveness of universal hepatitis B infant vaccination, various high-risk vaccination strategies, their marginal costs and benefits, and cost-effectiveness of Indigenous hepatitis B infant vaccination were investigated. Methods: This study is a part of ACE-Prevention project and uses its study protocol: a 3% discount rate, health care perspective, 2003 prices and health benefits measured in Disability Adjusted Life Years (DALYs). A decision tree model was built for influenza to compare universal influenza vaccination of people 50-64 years to current policy of vaccinating people over 65. Two scenario analyses were used as estimations of influenza like illness (ILI), but the primary input for the model was deemed unreliable. The results were compared to a previous study to verify the effect of model structure, assumptions on costs, vaccine uptake and vaccine efficacy using deterministic sensitivity analyses. A micro-simulation model was used to represent the natural history of hepatitis B. The model used data inputs from several Asian countries which was then fitted to Australian data. The cost- effectiveness of universal infant vaccination and targeted high-risk vaccination strategies were measured in comparison to no vaccination. The marginal costs and benefits of these strategies over each other were estimated in an expansion pathway analysis and were compared to current practice. The cost-effectiveness of Indigenous hepatitis B infant immunisation strategies in comparison to no vaccination were estimated separately. Results: Due to the absence of reliable data, the incremental cost-effectiveness ratio (ICER) of reducing the influenza vaccination age threshold to 50 years from the current age of 65 ranged from AU$6,000 to AU$135,000/DALYs, depending on input parameter assumptions used. Universal infant hepatitis B vaccination compared to no vaccination was cost-effective at the population level (ICER = $609/ DALY). Immunisation strategies that target high-risk infants are cost saving and produce health gains compared to no vaccination. These analyses included targeted vaccination of infants whose mothers were born in high endemic countries; maternal screening, vaccination and immunoglobulin for infants of hepatitis B carrier mothers born in high endemic countries and/or carrier mothers in the general population. The results are valid at discount rates of 0%, 3% and 5%. The current hepatitis B vaccination practice of universal infant vaccination, and immunoglobulin and vaccination of infants born to hepatitis B carrier mothers in the general population compared to no vaccination is cost-effective (ICER=$7,500/DALY) as it produces large health gains at a net cost of AU$11 million a year. The current practice in the general population compared to the most cost-effective high-risk strategy is not cost-effective. Maternal screening, vaccination and immunoglobulin to infants born to carrier mothers in the general population saves costs with the least health loss and could be an alternative to current practice. All immunisation strategies in the Indigenous population produce health gains and a net cost saving compared to no vaccination. The current vaccination strategy produces maximum health gains (800 DALYs) and maximum net cost savings (AU$8 million) compared to other strategies. Implications: Data inadequacy was a challenge in constructing models to conduct cost-effectiveness studies that accurately reflect the natural history of the disease. In the absence of good data the validity of results obtained through the models depended upon the validity of assumptions and the applicability of data used in the situation at hand. The large uncertainty in model parameter inputs and the results thus obtained needed to be analysed critically. The influenza model structure, parameter assumptions and data limitations introduced huge uncertainty which is not sufficiently accounted for in existing cost-effectiveness studies.There is insufficient evidence to reliably study the cost-effectiveness of universal influenza vaccination for people aged 50 to 64. The policy implication is that there is a need to collect better data on ILI (or influenza) incidence to enable a reliable cost-effectiveness study. Australia has adopted an expensive but maximum health gain hepatitis B vaccination strategy in the general population. An alternative to the current strategy is to screen pregnant women for hepatitis B carrier state and immunise infants born to carrier mothers with vaccine and immunoglobulin. However, this alternative strategy may not be politically acceptable. The current Indigenous hepatitis B vaccination strategy is optimal and deserves continuation. The economic analysis of other existing immunisation programs is warranted.
