DECIPHer: Centre for the Development and Evaluation of Complex Interventions for Public Health Improvement

To develop effective interventions to improve population health requires an understanding of what works, for whom, under what circumstances and why? This necessitates the development of epidemiologically and social scientifically informed complex and multi-factorial interventions that are effective across settings and behaviours; the rigorous evaluation of complex interventions, often using pragmatic controlled trial designs with nested process evaluation and including natural experiments of new policy programmes and the use of routine data to develop and target interventions and to provide sources of data on contexts and long term outcomes for intervention studies. The centre for the Development and Evaluation of Complex Interventions for Public Health Improvement (DECIPHer) brings together a team of world class researchers with expertise in a wide variety of quantitative and qualitative evaluation methods and intervention technologies to address these issues. Building on a strong track record in working with partners from public health policy and practice, and in public involvement in research, we will develop, test, evaluate and then implement interventions that are effective in improving the health of the population across settings and behaviours and which address health inequalities, with a particular focus on children and young people.

DECIPHer's aims are to conduct research that:

1. will lead to measurable improvements in the health of children and young people

2. identifies policies and interventions that are readily adopted, implemented and maintained in the real world and are effective in improving health and reducing health inequalities.

We will also provide high quality training programmes to develop the skills and careers of public health researchers, and a focal point for collaboration with policy and practice.

For its second five-year period of funding, DECIPHer's core Centre resources will be focussed on developing a high quality cadre of early career researchers who, with a multidisciplinary team of senior scientists will take forward four research programmes:

The first will examine multiple risk behaviours in young people and their antecedents to develop complex interventions which address multiple risk behaviours and their causes.

The second will develop and conduct a programme of research on organisational approaches to promoting health in schools and other youth settings.

The third will focus on further development of methods to develop and evaluate complex interventions, with an emphasis on applying randomised trials and other high quality research designs to evaluating interventions in a way that produces rigorous evidence that is useful to public health decision makers.

The fourth programme is also methodological in nature, and is concerned with maximising the use of routine data from population surveys and other administrative sources to improve our understanding of the determinants of population health and facilitate the evaluation of interventions without the need for expensive collection of new data.

These four complementary programmes will provide a platform for multiple applied research projects, funded through additional competitive grant funding, which will be strategically co-ordinated within and across three broad priority areas relating to the health of children and young people: Tobacco, Alcohol, Drugs; Obesity, Physical activity, Diet; Mental health and wellbeing. DECIPHer will address health behaviour in its widest sense, encompassing individual risk and protective behaviour and the contexts and structures that improve, sustain or undermine health and well-being. Adopting a socio ecological approach, the focus will be on multiple behaviours and on the design and evaluation of interventions that take full account of the interdependencies between individual, social, family, community, organisational and policy factors.

Technical Summary

Multiple risk behaviours: Complex analyses of large scale longitudinal data sets including latent class analysis, with further analyses attempting to identify longitudinal relationships between multiple correlated determinants at multiple time points on the one hand, and multiple risk behaviours and their development over time on the other. Systematic reviews of the literature, including new methods which systematically identify and synthesise relevant social theory.

Health promoting settings: Transdisciplinary action research partnerships including: repeated measures data collection to inform needs assessment and strategy development, longitudinal evaluation and a data rich environment for system dynamics methods to be developed and applied to the modelling and adaptation of interventions; engagement of stakeholders to maximise ownership and maintenance of interventions; multilevel programmes, following a standard functional design based on evidence and theory, but optimised for local circumstance and need.

Complex intervention research: Mixed methods randomised trials of complex interventions. Further development will be in the analysis and dissemination of how to design and deliver such studies in the real world, contributing to methodological literature in policy trials and implementation science; developing the concept of realist trials; improving reporting and synthesis of complex trials; modelling of complex systems and relationships - agent-based modelling, social network analysis, and of long term cost-effectiveness.

Data linkage and natural experiments: Exploitation of data resources created through data linkage to facilitate natural experiments of population health interventions. This will include development and refinement of novel research designs, informed by engineering based systems science methods, to use repeated measures data to monitor, evaluate and improve interventions. We will further develop the Welsh Electronic Cohort for Children.