Retrieval of individual patient data depended on study characteristics: a randomized controlled trial

Abstract

ObjectivesThe aim of the study was to examine the effect of providing a financial incentive to authors of randomized clinical trials (RCTs) to obtain individual patient data (IPD). Study Design and SettingParallel-group RCT with authors identified in the RCTs eligible for two systematic reviews. The authors were randomly allocated to the intervention (financial incentive with several contact approaches) or control group (using the same contact approaches). Studied outcomes are proportion of authors who provided IPD, time to obtain IPD, and completeness of IPD received. ResultsOf the 129 authors contacted, 37 authors suggested or contacted a person or funder providing relevant details or showed interest to collaborate, whereas 45 authors directed us to contact a person or funder, lacked resources or time, did not have ownership or approval to share the IPD, or claimed IPD was too old. None of the authors shared their IPD. We contacted 17 sponsors and received two complete IPD datasets from one sponsor. The time to obtain IPD was >1 year after a sponsor's positive response. Common barriers included study identification, data ownership, limited data access, and required IPD licenses. ConclusionIPD sharing may depend on study characteristics, including funding type, study size, study risk of bias, and treatment effect, but not on providing a financial incentive.