UBC Theses and Dissertations
Nonpublication and publication bias in clinical trials in Canada : a qualitative interview study Morrow, Richard Lee
Background: Selective publication of clinical trials is common and leads to publication bias, but factors contributing to selective publication are not well understood. It is also unclear whether trialists believe they have a duty to trial participants to report their research. Objective: To understand (i) whether and how industry sponsors of clinical trials influence decisions to report trial results, (ii) factors contributing to nonpublication and publication bias, and (iii) how the experiences and views of trial participants, trial investigators, and others connected to clinical trial research relate to whether researchers have a duty to trial participants to report research findings. Design: Qualitative interview study. Participants: 34 participants including 17 clinical trial investigators, 1 clinical research coordinator, 3 research administrators, 3 research ethics board members, and 10 clinical trial participants. Setting: Semistructured interviews conducted in person or by telephone with participants from Alberta, British Columbia, and Ontario, Canada. Analysis: Data analysis was informed by grounded theory, including coding of interview transcripts, memo-writing, and developing key themes. Results: Industry sponsors may influence whether clinical trials are reported through stopping trials early and not reporting results, ownership and control of data, clinical trial agreements which do not fully protect an investigator’s right to publish, control of internal company trials, and funding dependency. While companies have a commercial incentive to selectively report trials, other incentives within clinical research also appear to favour publication of positive over negative trials. Positive findings are perceived to be easier to publish, to help investigator’s ability to access industry and nonindustry research funding, and to be rewarded by research institutions in hiring, promotion and recognition. Interviews suggested that when participants enter a trial, there is often an implicit understanding between researchers and participants involving a responsibility to report results. Accounts of trial investigators suggested reporting research results is a necessary part of honouring informed consent. Conclusions: While clinical trial reporting is valued in Canada, selective reporting of clinical trials arises for a variety of reasons. Policy to promote full reporting of trials may be strengthened by recognizing factors that contribute to nonpublication and publication bias.
Item Citations and Data
Attribution-NonCommercial-NoDerivatives 4.0 International