Adoption of the World Health Organization’s best practices in clinical trial registration and reporting among top public and philanthropic funders of medical research in the United States

Abstract

Background/Aims: Clinical trial funders in the United States have the opportunity to promote transparency, reduce research waste, and prevent publication bias by adopting policies that require grantees to appropriately preregister trials and report their results, as well as monitor trialists registration and reporting compliance. This paper has three aims: a) to assess to what extent the clinical trial policies and monitoring systems of the 15 largest public and philanthropic medical research funders in the United States meet global best practice benchmarks as stipulated by the WHO Joint Statement;[1] b) to assess whether public or philanthropic funders have adopted more WHO Joint Statement elements on average; and c) to assess whether and how funders policies refer to CONSORT standards for clinical trial outcome reporting in academic journals. Methods: The funders were assessed using an 11-item scoring tool based on WHO Joint Statement benchmarks. These 11 items fell into four categories: trial registration, academic publication, monitoring, and sanctions. An additional item captured whether and how funders referred to CONSORT within their trial policies. Each funder was independently assessed by 2-3 researchers. Funders were contacted to flag possible errors and omissions. Ambiguous or difficult to score items were settled by an independent adjudicator. Findings: Our cross-sectional study of the 15 largest public and philanthropic funders in the US finds that on average, funders have only implemented 3.8/11 (35%) of World Health Organization best practices in clinical trial transparency. The most frequently adopted policy requirement was open access publishing (14/15 funders, 93%), and the least frequently adopted were (1) requiring trial ID to appear in all publications (2/15 funders, 13%) and (2) making compliance reports public (2/15 funders, 13%). Public funders, on average, adopted more policy elements (5.2/11 items, 47%) than philanthropic funders (2.5/11, 23%). Only 1 funders policy documents mentioned the CONSORT statement. Conclusions: There is significant variation between the number of best practice policy items adopted by medical research funders in the United States. Many funders fell significantly short of WHO Joint Statement benchmarks. Each funder could benefit from policy revision and strengthening. Keywords: Clinical trials, transparency, registration, reporting, publication bias, United States, funders, NIH