What were we trying to do?
As described in the paper Process evaluations nested in randomised controlled trials of complex interventions: a scoping review of approaches and reporting, we were trying to find out how process evaluations have been carried out alongside randomised controlled trials (RCTs) of complex interventions, and how these process evaluations have been reported. 

 

• Complex interventions are a set of actions designed to solve a complicated health problem.
• Process evaluations are studies done alongside trials to understand how and why an intervention works and if this is influenced by certain contextual factors. 
• Randomised Controlled Trials are experiments carried out in two (or more) groups to find out what the impact of an intervention is when some participants receive it and others don’t. 

 

 

Why was this important?
Doing a process evaluation during an RCT is useful because it helps us understand whether an intervention works as expected, and the reasons why or why not. Since 2015, Medical Research Council (MRC) guidance has recommended doing a process evaluation alongside an RCT of a complex intervention. However, there has been little review of how the guidance has been used in practice. 

 

Finding out how this MRC guidance has been used is important, because it will show whether it’s being followed and whether more detailed advice is needed.

 

How did we do it?
We searched for all published papers that mentioned the 2015 MRC guidance. We then used a filter to include all papers about RCTs from any field of health or social care. We located any data in the papers relating to:

 

• study details
• methods used for data collection and analysis in the process evaluation
• integrations of process evaluation data with RCT data
• use of theory in process evaluation design
• reporting of the process evaluation (as part of the RCT paper/report, or as a separate study)

 

We analysed all the data we had collected to find out whether the 2015 MRC guidance on the use of process evaluations was being followed. 

 

Findings

• We found 160 papers about full RCTs and 53 papers about pilot or feasibility RCTs. 
• Most process evaluations included both interview-based (qualitative) and statistics-based (quantitative) data collection.
• Data analysis methods included descriptive statistics and analysis of themes.
• Most studies didn’t report the use of theory in designing the process evaluation.
• There was little description of how the process evaluation and RCT outcomes were linked.
• Not all details about process evaluations were reported in the papers. 

 

Findings showed that while process evaluations can be helpful in explaining the outcomes of RCT interventions, the information that’s gathered in a process evaluation doesn’t often get fully reported in papers. We recommend that authors consider including supplementary information to provide a fuller picture of the process evaluation used during an RCT. 

 

There are currently no suitable reporting guidelines for process evaluations linked to RCTs, so work to develop such guidelines would also be valuable.

 

 

More information

 

 

Programme Manager
Gill Rizzello
Gill.rizzello@manchester.ac.uk