A classification of methods used to personalize participative interventions revealed inadequate reporting in trial protocols

Objectives: The objective of the study was to develop a classification of methods used to personalize participative interventions in randomized controlled trials (RCTs).

Study design and setting: We conducted a systematic review including protocols of RCTs assessing participative interventions in PubMed and ClinicalTrials.gov between June 2018 and May 2019. Data extraction was performed by two independent reviewers. We developed a precise classification of methods used to personalize interventions. Then, protocols were reviewed to determine whether personalization was sufficiently described to enable replication.

Results: We included 109 protocols. The classification used four components and 13 subcomponents accounting for decision points (when interventions were personalized), tailoring variables (on what interventions were personalized), decision rules (how and by whom interventions were personalized), and nature of the subsequent tailoring (what was personalized in the interventions). In 95% of protocols, at least one component or subcomponent of our classification was not adequately reported to enable the replication of the intervention. Components the least well described were tailoring variables (72% of protocols insufficiently described) and the nature of the subsequent tailoring (46% of protocols).

Conclusion: This study provides the first detailed classification of methods used to personalize interventions. This is required to transparently implement personalization and improve reporting in RCTs.