Economic and impact modeling to guide introduction of new tools against malaria

Background
Tremendous gains have been made over the last decades in scaling-up malaria preventive and curative interventions. Through this effort, new partnerships and funding streams were mobilized to support innovation toward global malaria targets. As a result, there are many promising new tools to detect, treat, and prevent the disease in clinical development and under evaluation for regulatory approval and eventual adoption by programs. The aim of this thesis was to generate evidence on costs and impact of a new vaccine against malaria (RTS,S AS/01) and to inform downstream policy decisions on its implementation in Plasmodium falciparum (Pf) malaria endemic African countries.
Methods
A micro-costing methodology was proposed to estimate costs of the vaccine implementation and several other interventions currently deployed by malaria programs (e.g. mass drug administration, indoor residual spraying, rapid reporting, and reactive case detection). The costing models were developed around regionally relevant implementation scenarios informed with program operational documents and inputs from local partners. The vaccine implementation costs derived were then linked to an individual-based model of malaria transmission dynamics that included a vaccine model parameterized to data from the phase 3 trials. Taking into account country epidemiological and health systems features, the model predicted the incremental impact and cost-effectiveness of the new vaccine when added on top of routine prevention and treatment for each of the 43 Pf endemic countries in Africa. Finally, nationally representative data from the Demographic and Health Surveys (DHS) were analyzed to evaluate the scale-up and the distribution of routine malaria interventions with respect to equity. The analysis sought to identify delivery channels for deployment of the new vaccine that might offer a comparative advantage by expanding access and or by fostering equity in malaria prevention.
Results and significance
The costing studies demonstrated how prospective micro-costing approaches drawing on secondary data could be used to estimate costs of new interventions prior to their implementation in countries. The methodology developed within the thesis yielded highly contextualized and programmatically relevant costs of existing and new interventions to inform decisions at global, country, or other levels. It demonstrated large differences in cost of RTS,S service delivery between Sub-Saharan African (SSA) countries: ranging from 0.72 USD per dose in Burkina Faso to 2.34 USD in Kenya. The impact and cost-effectiveness evaluation of RTS,S indicated that the vaccine was likely to be cost-effective under conventional GDP-based thresholds in most moderate transmission countries; the estimated cost-effectiveness varied with country vaccination coverage, within country heterogeneity in transmission, and cost of service delivery. The analyses of the DHS data further underscored the importance of monitoring adoption and scale-up of new health technologies for equity. Failure to do so could lead to lasting gradients in access to prevention and undermine effectiveness of interventions missing populations at highest risk where the benefits are likely to be greatest. These considerations are immediately relevant for RTS,S currently being evaluated for deployment through the Expanded Program for Immunization that was shown to have higher coverage than other routine malaria interventions but was also more pro-rich than interventions deployed via campaigns (i.e. long-lasting insecticide treated nets (LLINs), IRS). In addition to clinical and field trials, models of transmission dynamics are increasingly used to support evaluation of interventions against malaria. Parameterized with field data the models adequately capture long-term effects of interventions and interactions between properties of new interventions, immunity, and features of the setting on clinical outcomes. The methodologies and the workflows developed within this thesis to support the evaluation of RTS,S for global policy recommendation can be readily adapted to other interventions against malaria. As malaria endemic countries diversify with respect to prevalence, health systems, and economic development, modeling combined with economic analysis become increasingly useful tools to inform unique prioritization of interventions within country malaria programs, including for adoption of new tools.