Process Evaluations nested in randomised controlled trials of complex interventions: A scoping review of approaches and reporting

Background The United Kingdom Medical Research Council (MRC) guidance for complex interventions recommends the nesting of process evaluations in randomised controlled trials (RCTs) of complex interventions. To date, there has been limited review of how the MRC guidance has been operationalised in RCTs of complex interventions. Objectives This scoping review aimed to identify how process evaluations nested in RCTs on complex interventions have been conducted and reported since the 2015 publication of MRC guidance. Methods We identified articles in Pubmed and Scopus that cited the 2015 main journal article on the MRC guidance and applied an RCT filter to these articles. Studies reporting on RCT nested process evaluations in any field of health or social care were included. Data was extracted about study details, process evaluation study design, including methods used for data collection, data analyses, integration with RCT data and use of theory in process evaluation design. Results We identified 160 RCTs and 53 pilot/feasibility RCTs. Most process evaluations included a combination of qualitative and quantitative data collection methods, either triangulated (mixed method studies) or not (multi-method studies). Most studies did not report the use of theory in shaping how the process evaluation was designed. Common methods for data analysis included descriptive statistics and thematic analysis for quantitative and qualitative studies respectively. There was limited triangulation of process evaluation data with RCT outcomes. Conclusions While process evaluations can be helpful in explaining outcomes of RCT interventions and subsequent implementation of these intentions, information gathered from process evaluations is often not fully reported in papers. There are currently no suitable reporting guidelines for process evaluations linked to RCTs; work to develop such guidelines, potentially as a StaRI checklist extension would be valuable.