Research units A-Z

Workstrand 5 brings to life the synthetic populations generated in Workstrand 3, estimating how those populations are likely to change over time within each community.

They also assess how policy scenarios or changes in the environment may influence the characteristics of the population, including people’s health and wellbeing. For example, we may have data and evidence about how improving social housing or increasing employment rates affects health.

Micromodelling allows us to assess what the impact might be on different sub-groups of the population (e.g. single parent households) within different geographical areas (such as neighbourhoods).

The effect of policy interventions or changes in individual circumstances are being assessed as “what if” experiments running into the future. For example, asking questions such as “what would happen to health if employment rates in young men could be increased and how does this compare to increasing employment rates in older men or young women?” or “what would happen to productivity in this area if older adults were to stay healthy for longer?”.

The policy scenarios to be assessed are being chosen on the basis of stakeholder priorities.

Using the synthetic population, the policy microsimulation models in Workstrand 5 is estimating how people move between different ‘states’, for example, a transition from well to ill, or employed to unemployed.

We are finding out about what these transitions look like from:

• Causal links identified through the complex systems mapping in Workstrand 1 and the evidence reviews in Workstrand 2;

• Analyses of partner data; and,

• Analyses of large datasets collected by the government, which follow the same individuals or families over a long time and asks them about changes in their lives (e.g. Understanding Society, the English Longitudinal Study of Ageing, and Healthy Ageing in Scotland).

The most important part of this process is to ensure that the transitions used for each individual are appropriate, given the individual’s specific set of characteristics (e.g. gender, age, socio-economic status, and so on).

The microsimulation is dynamic, which means that individuals in our synthetic populations age over time and their attributes are updated as their circumstances change.

Microsimulations, similar to the ones created by Workstrand 5 are deployed in a wide range of contexts, both within the UK and internationally. For example, microsimulations are used to assess the outcomes of tax reforms or the health of the population after certain interventions (e.g. changing taxes on alcohol or tobacco).

SIPHER are using the models to analyse the effects of different policy scenarios being considered by our policy partners for their topic areas (e.g. an inclusive economy). The models provide additional insight to the population-level models produced in Workstrand 4 because they uniquely focus the modelling at the level of individual persons.

A person-specific approach provides a great deal of flexibility in assessing differential effects across a wide range of population groups of interest to policy makers, allowing impacts on inequalities to be examined. With careful integration, Workstrands 4 and 5 are providing robust insight into the effect of interventions across a wide range of sub-populations and policy areas.