Participants' and investigators' experiences and views of potential adverse effects of an educational intervention: Protocol for a qualitative evidence synthesis

Introduction
To help students think critically about health information and decisions, we developed the Informed Health Choices (IHC) secondary school intervention. We are evaluating the intervention in cluster-randomised trials, and linked process evaluations, in Kenya, Rwanda, and Uganda. The study planned herein is a qualitative evidence synthesis (QES), using data about experiences and views of adverse effects from all three process evaluations. The QES overlaps with the process evaluations. The QES will allow us to comprehensively explore, report, and discuss experiences and views of potential adverse effects, and potential mechanisms. The findings are intended to help inform decisions about whether or how to redevelop, re-evaluate, or implement the intervention. The methods and findings might also be helpful to developing, evaluating, or implementing other educational interventions—especially interventions intended to improve critical thinking, within health or other fields.

Objective
Explore participants’ and investigators’ experiences and views of potential adverse effects of the IHC secondary school intervention, and potential mechanisms of those effects.

Methods
An independent researcher will assess methodological limitations of the included studies, based on a list of domains used by the Cochrane Effective Practice and Organisation of Care (EPOC) group. We will merge and modify framework analysis as outlined by Ritchie and Spencer, framework synthesis as described by Barnett-Page et al., and “best fit” framework synthesis as outlined by Carrol et al. All three are pragmatic approaches with a deductive analysis using an a priori framework, followed by an inductive thematic analysis. We will note possible differences in how participants generally conceptualise adverse effects of educational interventions, compared to the study team, and possible differences in adverse effects across trial settings, for the purposes of future research. To assess confidence in the synthesis findings, we will apply the Grading of Recommendations Assessment, Development and Evaluation Confidence in Evidence from Reviews of Qualitative research (GRADE-CERQual) approach, using the interactive Summary of Qualitative Findings (iSoQ) tool. We will produce a Qualitative Evidence Profile, and Summary of Qualitative Findings Tables.

Discussion
The QES and the process evaluations overlapping and being part of the same project has methodological implications that amount to both strengths and limitations. Like in prospective meta-analyses, when planning the process evaluations, we harmonised the objectives, facilitating synthesis, while investigators in each setting still had autonomy to explore phenomena specifically for their study. Like individual patient data (IDP) meta-analyses, this study will be based on data from the process evaluations, facilitating more reliable analysis and synthesis than if it was only based on reported findings. As far as we are aware, this QES will be the first of its kind methodologically, and first empirical study of its size and rigour focusing on potential adverse effects of an educational intervention. The QES and process evaluations overlapping and being part of the same project also introduces risk of bias. Another challenge is that we are including the study team’s experiences and views. In general, to address these challenges, we will be transparent, and apply reflexivity throughout.