Guidance for intervention fidelity in nondrug, nonsurgical trials: a scoping review

Intervention fidelity is poorly reported in trials of nondrug, nonsurgical interventions. We aimed to review existing intervention fidelity guidance documents, assess their development methods, and synthesize their terminology, definitions, domains, and recommendations.

We searched 6 databases from inception to September 2023 to identify documents describing explicit, actionable, and overarching recommendations on intervention fidelity. We supplemented this with backward citation searching in April 2024. Two reviewers independently screened titles, abstracts, and full texts. We conducted a qualitative content analysis, following JBI guidance, to extract and synthesize data on document characteristics, scope, methods, terminology, definitions, domains, and recommendations.

After screening 7142 records from database searches and 1768 from citation searching, we included 73 documents. Most documents were unclear about the core methods used to develop recommendations (67/73, 92%), and only one document explicitly reported using consensus (1/73, 1%). We identified definitions of intervention fidelity across 6 categories and 7 intervention fidelity domain categories. Of these, only 2 ("provider delivery-related aspects" and "participant involvement-related aspects") were reported in most documents (>50%). We identified 81 recommendations across 10 categories, with only 14 being reported across more than half of the documents.

Existing documents providing recommendations on intervention fidelity reflect the evolving nature of the field, with consensus still needed on several aspects. The gaps and inconsistencies identified in this review emphasize the need for clearer, more systematic, and globally applicable guidelines. These findings will inform the development of ReFiND, an international, consensus-based reporting guideline for intervention fidelity in nondrug, nonsurgical trials.

When studies are conducted to test how well an intervention works, it is key to evaluate the extent to which the intervention is delivered as planned. This is called intervention fidelity. We looked at 73 guides on how to check and report this. Most of these guides did not explain the methods used to decide what to include. The advice in these guides was often different from each other. As the field of intervention fidelity is still developing, many of these guides represent early attempts, where less structured methods were used before more organized approaches were established. This shows we need a global effort to create clear guidelines for reporting how well interventions are followed.