Background A number of recent trials and systematic reviews of public health interventions have had disappointing conclusions despite a major investment in trials. An elegant evaluation design to test a poorly developed intervention is clearly unsatisfactory, so to what extent is the effectiveness of public health compromised by interventions that are poorly developed in the first place? Guidance on the development of complex interventions highlights the need to develop theory-based interventions to the point where they can reasonably be expected to have a worthwhile effect. However, in practice many public health interventions are complex, have arisen from a variety of sources, contain elements the quantity and quality of which arises from chance rather than design, have not been through a pre-intervention development phase, are not theory based, and are relatively poorly defined and described.

Methods Firstly, a systematic search was conducted to find literature which describe methods for the development of public health interventions (as opposed to how to evaluate them). Secondly, a step-wise comparison was carried out between the development of new drugs and the development and introduction of public health interventions.

Results There is little literature on the development of public health interventions which suggests that the “science” of intervention development (in public health) is still at an early stage. Recommended stages in intervention development include: identifying the evidence base; identifying and developing appropriate theory – and, while evidence-based guidance on the development of behaviour change interventions in public health has been produced, there may be several competing theories requiring judgement in respect of which is most appropriate. Moreover, the key components of complex public health interventions, including their implementation, need to be better identified and described in terms of their interrelations and replicability. While pharmaceutical products have a distinct development phase which is well planned and appropriately funded, this is not the case for public health interventions although it is clear that there are lessons for public health from drug development. In part the problem arises because of the lack of a champion or sponsor for the development of public health interventions, and also because of the absence of funding to develop those interventions in the first place.

Discussion For public health interventions to be as successful and effective as they have the potential to be, we must review and revise how such interventions are initially designed and developed. A clear development and sponsorship process for public health interventions of all complexity needs establishing.