Abstract
BackgroundPilot and feasibility studies play an important role in developing and evaluating complex interventions prior to full-scale randomised controlled trials (RCTs). They allow researchers to assess trial processes, intervention delivery, acceptability, and implementation before definitive evaluation. Medical Research Council guidance recognises that interventions and procedures may need refinement during this process. Consequently, modifications to interventions and trial protocols may occur between pilot/feasibility studies and subsequent definitive RCTs. However, the nature, extent, and reporting of these modifications remain unclear. Inconsistent or poorly justified reporting may compromise interpretation of feasibility findings and weaken the rationale for progression to a trial. ObjectivesThis review aimed to identify and categorise modifications reported between pilot/feasibility studies and corresponding main RCTs, quantify their frequency, and determine whether modifications and their reasons were reported. MethodsA systematic search of CINAHL, Embase, MEDLINE, PsycINFO, and PubMed was conducted on 13 June 2025 to identify studies published from 2016 onwards. Eligible studies included completed or ongoing randomised controlled trials with a preceding published external pilot or feasibility study. Study selection was undertaken independently by two reviewers. Data on modifications were extracted, by PICOT domains, and classified as major or minor using a predefined framework. Data extraction was conducted by one reviewer and verified by a second. Formal risk-of-bias and certainty-of-evidence assessments were not undertaken, as the primary objective was to characterise and analyse modifications and reporting practices rather than evaluate intervention effects. ResultsA total of 3,361 records were identified, and 1,015 full-text articles were assessed for eligibility. Fifty-five pilot/feasibility–main trial studies were included. Modifications between pilot/feasibility and main trials were common, particularly for the population (73%) and intervention (64%). Changes were also observed in the comparator (42%), primary outcomes (29%), and timing of outcome assessment (16%). Modifications were classified as major or minor, with the majority classified as major across PICOT domains. Reporting varied across domains, with intervention changes most consistently documented. The review was limited by reliance on published reports and the absence of a validated framework for classifying modifications. These findings highlight a need for greater transparency in reporting modifications across trial components. ConclusionModifications between pilot/feasibility studies and related main RCTs are frequent and affect multiple trial components, particularly interventions and populations. However, reporting is inconsistent and lacks transparency. Poor reporting may make it difficult to judge whether feasibility findings adequately support the design and progression of a definitive trial. Improved and standardised reporting of modifications is needed to enhance the transparency, interpretability, reproducibility, and methodological rigour of trials transitioning from pilot/feasibility work to definitive evaluation. (PROSPERO: CRD420251020958)