Fig. 2

One-way sensitivity analysis by country. a South Africa (current scenario, drug price 1USD per day). We show variations in incremental cost-effectiveness ratio (ICER) to analyse the influence of different assumptions on our conclusions of cost-effectiveness. High/low refers to a higher/lower value of the parameter being considered compared to the baseline value. The x-axis shows the change in the ICER where 0 represents no change (ie baseline ICER). The double red line represents the change in ICER when the result is cost saving (i.e. negative ICERs). Negative ICERs are not at scale and this is indicated by a double slash. b Brazil (guidelines current, drug price 1USD per day). We show variations in incremental cost and incremental effect as opposed to changes in incremental cost-effectiveness ratio (ICER) because for Brazil, the ICER remains negative in this scenario (ie cost saving). The purpose is to investigate the impact of our assumptions on two components of the ICER: incremental cost and incremental effect. High/low refers to a higher/lower value of the parameter being considered compared to the baseline value. The x-axis shows the change in incremental costs or incremental effects (DALYs averted) compared to the baseline result (a negative value in incremental costs means less cost differentials, same applies to the DALYs), 0 represents no change (ie baseline). c Bangladesh (current scenario, drug price 1USD per day). We show variations in incremental cost-effectiveness ratio (ICER) to analyse the influence of different assumptions on our conclusions of cost-effectiveness. High/low refers to a higher/lower value of the parameter being considered compared to the baseline value. The x-axis shows the change in the ICER where 0 represents no change (ie baseline ICER). The single red line represents the change in ICER when the result becomes cost-effective (one GDP as willingness-to-pay threshold). The double red line represents the change in ICER when the result is cost saving (i.e. negative ICERs). Negative ICERs are not at scale and this is indicated by a double slash. d Tanzania (current scenario, drug price 1USD per day). We show variations in incremental cost-effectiveness ratio (ICER) to analyse the influence of different assumptions on our conclusions of cost-effectiveness. High/low refers to a higher/lower value of the parameter being considered compared to the baseline value. The x-axis shows the change in the ICER where 0 represents no change (ie baseline ICER)