Complex Interventions

Meeting of the Primary health Care Study Group in association with the Network of Hubs for Trials Methodology Research

  • Topic: Complex Interventions in Trials and Assessing the Generalisability of the Trial Findings
  • Date: May 9, 2011
  • Venue: Royal Statistical Society, Errol Street, London


  1. Emily Crowe (Coordinator - Network of Hubs for Trials Methodology Research)
  2. Aziz Sheikh (Centre for Population Health Sciences, University of Edinburgh)
  3. Walter Gregory (Cancer Division, Clinical Trials Research Unit, University of Leeds)
  4. Stephan Duffy (Wolfson Institute of Preventive Medicine, Queen Mary’s, University of London)
  5. Group Discussion - ALL


The following issues were identified during the group discussion:

  1. Access to routine data – difficult and not very open to researchers.
    The datasets available need to be mapped (includes primary care and hospital data sets)
  2. What can you reliably get from routine data sets?
  3. More expertise with working with data sets is needed – who has knowledge of working with datasets? Work done in one dataset needs to be validated in another dataset
  4. Need to develop more rigorous methods for comparing data in a subset of routine data and GP reported data.
  5. There are hundreds of complex interventions – is there a way of looking at the intervention and not the disease area as the intervention could be generaliseable? Is it possible to do systematic reviews of similar complex interventions?
  6. We need software to model the effect of different components but the problem is getting accurate data to feed the models. Qualitative analysis to help identify what the interventions are doing individually may help.