- Posted By: Bethan Copsey
On 24th April 2015, I attended the Clinical Trials of Complex Interventions Conference at the University of Warwick. The conference focused on methodology of complex intervention trials, with topics considering the key challenges of ethical issues, outcome reporting, statistical methodology and evaluation.
The first session presented an overview of complex intervention trials with specific reference to the additional complexities and ethical issues. An intervention was thought to be defined as complex due to the number of, and interaction between, components that make up the intervention, as well as the number of outcomes and possibility of treatment being tailored to the individual patient. The added challenges brought by complex interventions were discussed such as increased ethical issues for example, describing complex interventions in a satisfactory manner to gain fully informed consent. Challenges specific to complex interventions in surgical trials were also discussed, including changing the culture in clinical practice to be more open to new technologies and therapies.
The second session was based on the outcomes in complex intervention trials. Dr Kirstie Haywood appealed for core outcome sets to improve consistency of the measures used and timing of assessment in outcome reporting and stressed the importance of valid rationale for outcome selection. The use of process evaluation within complex intervention trials was also discussed, such as assessing the reach, received dose and level of fidelity, in order to place the study results in context and show whether further work is necessary for successful implementation into practice.
The third session focused on trial design and analysis. It was acknowledged that statistical methodology for drug trials may not be applicable for analysing complex intervention trials because of challenges such as multifactorial structure and individualisation of interventions. Adaptive design for treatment selection was described combining an exploratory study to refine the intervention and a large definitive trial, with focus on assessment of heterogeneity (complex interventions are rarely ‘one size fits all’) and controlling error rates. There was also discussion on clustering, especially the use of the intra-cluster correlation coefficient (ICC), and a brief introduction to health economic evaluation. Professor Sallie Lamb closed the session and described common pitfalls including the failure to engage the right people with specific experience and aiming for unrealistically large effect sizes, and the need for sufficient power, resources, recruitment and retention.
Overall, it was a very interesting and well attended event with speakers from a variety of disciplines discussing a broad range of aspects of trials of complex interventions. For more information, see the MRC guidance on complex interventions or the other key references listed below.
References:
• General:
- MRC guidance: Developing and evaluating complex interventions (http://www.bmj.com/content/337/bmj.a1655)
- The Ottawa Statement on the ethical design and conduct of cluster randomised trials (http://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.10...)
• For core outcome sets:
COMET (http://www.comet-initiative.org/)
COSMIN
CONSORT-PRO (http://www.consort-statement.org/extensions?ContentWidgetId=560)
SPIRIT-PRO
• For process evaluation: Steckler and Linnan (2002) Process evaluation for public health interventions and research (http://blogs.cofc.edu/honorshonduras/files/2013/01/Process-Evaluation-fo...)
• For reporting intervention details: TIDieR guidelines (http://www.equator-network.org/reporting-guidelines/tidier/)