The UK Medical Research Council’s widely used guidance for developing and evaluating complex interventions has been replaced by a new framework, commissioned jointly by the Medical Research Council and the National Institute for Health Research, which takes account of recent developments in theory and methods and the need to maximise the efficiency, use, and impact of research.

Despite major investment in both research and policy, many pressing contemporary public health challenges remain. To date, the evidence underpinning responses to these challenges has largely been generated by tools and methods that were developed to answer questions about the effectiveness of clinical interventions, and as such are grounded in linear models of cause and effect. Identification, implementation, and evaluation of effective responses to major public health challenges require a wider set of approaches 1,2 and a focus on complex systems. 3,4 A complex systems model of public health conceptualises poor health and health inequalities as outcomes of a multitude of interdependent elements within a connected whole. These elements affect each other in sometimes subtle ways, with changes potentially reverberating throughout the system. 5 A complex systems approach uses a broad spectrum of methods to design, implement, and evaluate interventions for changing these systems to improve public health.Complex systems are defined by several properties, including emergence, feedback, and adaptation. 3 Emergence describes the properties of a complex system that cannot be directly predicted from the elements within it and are more than just the sum of its parts. For example, the changing distribution of obesity across the population can be conceptualised as an emergent property of the food, employment, transport, economic, and other systems that shape the energy intake and expenditure of individuals. Feedback describes the situation in which a change reinforces or balances further change. For example, if a smoking ban in public places reduces the visibility and convenience of smoking, and this makes it less appealing, fewer young people might then start smoking, further reducing its visibility, and so on in a reinforcing loop. Adaptation refers to adjustments in behaviour in response to interventions, such as a tobacco company lowering the price of cigarettes in response to a public smoking ban.Rhetoric urging complex systems approaches to public health is only rarely operationalised in ways that generate relevant evidence or effective policies. 1,6 Public health problems that emerge as a property of a complex system cannot necessarily be solved with a simple, single intervention, but the interacting factors within the system can potentially be reshaped to generate a more desirable set of outcomes. 7,8 Achievement of meaningful impacts on complex multicausal problems, like obesity, requires more than single interventions, such as traffic light food labelling or exercise on prescription, many of which require high levels of individual agency, have low reach and impact, and tend to widen health inequalities. 9-11 Shifts within multiple elements across the many systems that influence obesity are required, some of which might only have small effects on individuals but can drive large changes when aggregated at population level. 12 Although randomised controlled trials of individual-level interventions are relatively strai...

Randomized trials of complex public health interventions generally aim to identify what works, accrediting specific intervention 'products' as effective. This approach often fails to give sufficient consideration to how intervention components interact with each other and with local context. 'Realists' argue that trials misunderstand the scientific method, offer only a 'successionist' approach to causation, which brackets out the complexity of social causation, and fail to ask which interventions work, for whom and under what circumstances. We counter-argue that trials are useful in evaluating social interventions because randomized control groups actually take proper account of rather than bracket out the complexity of social causation. Nonetheless, realists are right to stress understanding of 'what works, for whom and under what circumstances' and to argue for the importance of theorizing and empirically examining underlying mechanisms. We propose that these aims can be (and sometimes already are) examined within randomized trials. Such 'realist' trials should aim to: examine the effects of intervention components separately and in combination, for example using multi-arm studies and factorial trials; explore mechanisms of change, for example analysing how pathway variables mediate intervention effects; use multiple trials across contexts to test how intervention effects vary with context; draw on complementary qualitative and quantitative data; and be oriented towards building and validating 'mid-level' program theories which would set out how interventions interact with context to produce outcomes. This last suggestion resonates with recent suggestions that, in delivering truly 'complex' interventions, fidelity is important not so much in terms of precise activities but, rather, key intervention 'processes' and 'functions'. Realist trials would additionally determine the validity of program theory rather than only examining 'what works' to better inform policy and practice in the long-term.