Background The Medical Research Council (MRC) guidance for the development and evaluation of complex interventions describes such an intervention as one which has several interacting components; the MRC recommends that complex interventions are developed through literature identification, modelling of processes and theory development. We describe the utilisation of this guidance to develop the SAFER 2 (Support and Assessment for Fall Emergency Referrals) study intervention—a multi-centre randomised controlled trial (RCT) looking to improve the care Ambulance Services provide to elderly patients who fall.

Methods Literature was identified using MedLine and through applicant's previous work in this field. Two pilot studies were undertaken in two UK ambulance services; an economic evaluation was identified. A workshop, incorporating stakeholders from ambulance services, acute trusts and primary care, allowed further development and finalisation of the intervention. Specialist sub-groups were set-up to develop specific components of the intervention

Results Literature review identified previous RCTs and a recent systematic review, highlighting the need for a comprehensive evaluation of the ability of paramedics to safely make decisions to not convey elderly patients who fall to hospital and the effect of direct referral to multi-factorial community interventions on such patients. Pilot studies reinforced the need for an intervention that improves paramedic's decision making. The workshop resulted in the amalgamation of all evidence and agreement of the final intervention by all stakeholders.

The final definition of the intervention comprised of paramedic training, patient assessment tool, referral tool, process of referral, clinical support and Falls Prevention Service make-up and response.

Discussion Thorough intervention development and effective implementation are vital to the success of SAFER 2 and other studies in this field. In order to produce generalisable evidence about clinical and cost effectiveness that is reproducible, the intervention needs to be clearly defined with input and agreement from study stakeholders.