The interactive tool allows users to view and amend unit cost assumptions by country and intervention, and to visualise the sensitivity of cost and cost-effectiveness results to these assumptions.

The tool shows a separate interface for each intervention study, selected using a drop-down list. Within each study, a country drop-down field is used to show country-specific assumptions and results. By default, cost assumptions have been calculated using the relativities outlined in Figure 2 and can be adjusted by double-clicking and entering alternative values into the unit cost fields.

Aggregated total costs by treatment arm and the incremental costs of intervention relative to control arms are shown for the selected intervention and country. Reported costs are aggregated to four main categories:
•    Healthcare costs (including hospital costs, GP and nurse visits, physiotherapy and occupational therapist sessions);
•    Social care costs (including residential and nursing care, home care and day care);
•    Other costs (including transport costs and replacement/opportunity costs for informal care)
•    Intervention costs (reflecting the cost of the main or comparator intervention).

Several studies also provide estimates of Quality Adjusted Life Year (QALY) outcomes. In these cases, cost-effectiveness is measured using an Incremental Cost-Effectiveness ratio (ICER): the additional cost of the intervention (relative to the comparative intervention option) divided by the corresponding difference in QALY outcomes. Among interventions that demonstrate a positive QALY effect, the option is considered cost-effective if incremental costs per QALY gained are less than a user-defined ‘willingness-to-pay’ (WTP) threshold.

Cost-effectiveness results are plotted on a cost-effectiveness plane (Figure 3). Interventions that are estimated to be more costly and more effective than their comparators reside in the North-East quadrant of the plane, with those lying below the WTP line considered to be cost-effective.

Figure 3: Cost-effectiveness plane