Professor Trish Greenhalgh, Professor of Primary Care Health Sciences at the Nuffield Department of Primary Care Health Sciences will present the findings of a recently conducted HTA-funded systematic review on models and methodologies for measuring research impact. Universities are now assessed not merely by their publications and grant income but also by their ‘impact’. This is potentially good news for those of us who are concerned about ivory tower research that sits on shelves (or electronic repositories of so-called ‘high impact’ journals) but has little effect on practice or policy. If we are going to take impact seriously, we need to be clear about the philosophical assumptions underpinning different kinds of research and also the different kinds of links between research, practice and policy.
The literature on impact in health services research has strong positivist roots and draws heavily on ‘logic models’ linking upstream research with downstream impact - the basis of the new discipline of ‘implementation science'. But in recent years this literature has become more philosophically diverse, attempting to combine the outputs of a firmly positivist evidence–based medicine (e.g. quantitative findings on the efficacy of tests and treatments) with a broader (constructivist, realist, performative) epistemology of research utilisation that incorporates various social science disciplines (notably, social psychology, organisational sociology, social policy, and science and technology studies). This means that ‘models’, however elegant, typically disappoint and the crafting of an ‘impact case study’ is often more art than science.