Research units A-Z

Workstrand 7 is developing a decision support tool to help policy makers, planners, service providers, researchers, and local communities better explore the likely effects of policy choices.

The range of impacts capable of being modelled by the tool include aspects such as the population’s health, emotional and financial wellbeing, and measures of an area’s economic performance. The tool can also show how these impacts change over time, how the impacts are distributed across the local population, and both when and where unintended consequences might arise.

A crucial part of the tool is showing how alternative policy options lead to different sets of impacts – allowing policy makers to identify options that fit best with their strategic priorities and with public preferences for improving society overall or for reducing inequality.

Workstrand 7 is work closely with practice partners to ensure that SIPHER tools integrate well into the wider toolset available to policy analysts.

Activities fall into three core areas:

• Bringing together data, models and findings from across SIPHER’s workstrands to understand the potential trade-offs and synergies between different outcomes in each policy partner’s topic area;
• Developing and applying new computational methods that can help identify and compare different possible courses of policy action; and,
• Creating an interactive software tool that can be used by policy analysts to monitor their policy challenges, identify trade-offs and synergies, and identify ‘best buy’ policy options to take forward.

SIPHER’s policy themes are cross-sectoral: they involve different policy teams who have interests in different outcomes but where strong interactions can exist between the sectors (e.g. the public health team are interested in health outcomes, the economic strategy team are interested in employment levels, and there is evidence to suggest a strong two-way link between health and employment). Not all of these interactions are presently known.

Workstrand 7 in partnership with Workstrand 1, is developing a map of all the important connections in each policy area. This map is constructed in partnership with topic experts and policy stakeholders in SIPHER’s partner organisations, using systems mapping methods. Workstrand 7 will then compare the map to how policy teams are organised in each partner, identifying for each relevant team:

• the policy levers at their disposal;
• the outcomes and regulatory constraints that are important to them and to the general public;
• the influences from other parts of the map that are not controllable by the team but that can affect their outcomes.

This process of identification will be done in formal mathematical terms to produce what is known as a distributed or multidisciplinary decision problem.

Workstrand 7 is drawing together relevant data and models from Workstrands 2, 3, 4, and 5 to represent each partner’s distributed decision problem – allowing changes in outcomes to be tracked as data becomes available, allowing the effects of potential policies to be estimated and allowing the abilities of our models to be tested and reflected on. We will then be able to estimate how a policy idea generated in one team would potentially affect the outcomes across all other teams. When new data and models become available the latest version of the decision problem is refreshed.

A second core activity for Workstrand 7 is developing computational methods that help policy makers search across the space of possible policy options to find approaches that best meet their strategic priorities and account for public preferences captured in Workstrand 6. These methods broadly belong to what is often called the “computational intelligence” research field. The methods are, essentially, recipes for finding good policy options based on simulation models and data.

In developing these methods there are a number of considerations, in relation to SIPHER’s decision problems:

• That they feature many outcomes, some of which may be in conflict with each other;
• That the outcomes and policy actions are spread across different teams;
• That some factors that can affect outcomes are not under the control of policy makers;
• That we are uncertain about the precise relationships in the systems map and uncertain about what will happen in the future to aspects outside the control of policy makers; and,
• That our simulation models may take a long time to run, which makes them unwieldy for analysing lots of different options at the same time.

Workstrand 7 is identifying ‘adaptive’ policies – those with the capacity to be changed after they have been implemented, in response to changes in external factors.

A software tool that policy analysts can use to set up a multisector decision problem, integrate data and simulation models, and perform interactive analyses is in development. The aim is to develop intuitive visualisations that help identify tensions and synergies across outcomes. Economic appraisal of options will be supported by implementing cost-effectiveness and cost-benefit metrics in the tool.

Workstrand 7 is developing decision support tools that are grounded in the realities of policy making, are useful to SIPHER’s policy partners, and have potential for wider take-up by other policy audiences in the UK and internationally.

These tools will enable policy makers to identify ‘best-buy’ options for improving wellbeing and reducing health inequalities that make economic and financial sense when set in a multi-sectoral context that is broader than traditional public health perspectives.

We also hope to stimulate a search for adaptive policy options that offer robust outcomes in the face of the deep uncertainties faced by policy makers.