STRIVE modelling meeting report.pdf
This report synthesises messages and recommendations from a two-day consultation involving 37 experts - mathematical modellers, epidemiologists, economists and policy-makers - from academia, civil society, bilateral development partners and multi-lateral organisations.
Objectives
STRIVE and the HIV modelling consortium convened the consultation, with support from the Global Fund to Fight AIDS, Tuberculosis and Malaria, in order to:
-
Review the state of the evidence on the effectiveness, costs andc ost-effectiveness of a range of structural interventions;
-
Investigate how these issues are currently addressed by available models, and identify limitations and potential improvements;
-
Discuss alternative modelling solutions, notably treatment and prevention cascades;
-
Learn from each other’s approaches and from approaches used in country processes to model structural interventions;
-
Generate recommendations about how models could better incorporate these interventions in the short term, and how this agenda should develop over time.
Background
The consultation took place against a backdrop of flat-lining international HIV financing, heightened focus on prioritised investment approaches and roll-out of country investment cases and growing global recognition of structural drivers in explaining high risk among adolescent girls and young women in sub-Saharan Africa. Policy-makers, funders and analysts all recognise their importance, but interventions that address structural drivers of risk and service uptake/adherence still tend to be excluded from resource allocation models and, therefore, under-prioritised in investment cases. This has largely been explained by the limited evidence available on their effectiveness for HIV endpoints and their low cost-effectiveness for the HIV budget. However, the evidence base has been growing and merits revisiting in light of renewed prioritisation efforts at national level.
Findings
Participants discussed short-term improvements to current models - and the data and process requirements these would involve - as well as, in the longer term, developing models for combination interventions with structural components. The final discussion focused on both short-term actions and longer-term ambitions.
-
On the basis of immediate endpoints, and how they are currently measured and modelled, structural interventions cannot compete with other HIV interventions, but there are likely to be other justifications for scaling them up further.
-
Structural factors research should be reorganised by population groups to understand the impact that each structural factor exerts on a population group, and how each relates to one or more HIV endpoints.
-
Modelling structural factors is a developing area, and there is a need to better understand which model structures can be used and their implications.
A major conclusion was to conduct a comprehensive review and re-order the current evidence in line with the HIV treatment and prevention cascades, and to then explore methods of integrating this into investment models, through a process guided by expert opinion."
