Clinical practice should ideally rely on robust scientific evidence, the standard for which are systematic reviews and meta-analyses of all randomised controlled trials (RCTs). (1,2) Therapeutic decisions in healthcare must be informed by clinical research findings, and patients and prescribers must be able to trust the presented research evidence. However, this evidence can be considered valid only if the studies included in reviews and meta-analyses signify the complete publications, without any bias. (2) Recently, the reliability of much of the evidence base for several popular therapeutic and preventive interventions has been challenged due to the publication bias. (3)
What is ‘publication bias’? It refers to data distortion in scientific journals mainly due to the increased likeliness of publications of those studies with significant and positive results compared to those with unfavourable or negative or insignificant results. (2) Publication bias, in essence, is a threat to the core principle of evidence-based medicine that is based on systematic reviews of published evidence providing accurate assessments of an intervention’s actual safety and efficacy data. (4) Moreover, the methods used to conduct the systematic review represent their validity. The presence of a systematic bias of favouring the publication of studies with statistically significant or positive findings substantially threatens the validity of the conclusions of a systematic review. (5) Study selection based on their status of publication (submitted or accepted), duplicate, undetected publications, and selective reporting are some of the factors that lead to publication bias. In addition, the bias also occurs when the publication selectively depends on the nature and direction of the study findings, (6) which will then be thoroughly different from those of unpublished studies.(2)
Evidence from literature confirms that the probability of the publication of studies with positive and favourable results is more than that of studies with negative or unfavourable results.(1,2,7) For instance, a systematic review and meta-analysis of 85 cohorts assessed their likelihood of publication based on different variables, such as favourability of results, statistical significance, study sponsorship, number of study centers, study phases, study design, study size, and country of origin; with the outcome of interest being complete publication in a peer-reviewed journal.(4) The authors found that favourability of results, statistical significance, and multicenter status were all significantly impacted the probability of publication. Moreover, the likeliness of publication of the studies presented as abstracts was significantly higher if a sponsor funded the studies. Therefore, the likeliness of publication for favourable study findings was significantly influenced by the study’s funding status.(3) This only goes on to suggest a strong association of differential publication due to the favourability of study findings and the status of sponsorship from the industry.(4)
The decision-making in clinical and medical practice should depend on the entirety of research evidence and not on a sample that is biased by selective publication only of studies showing favourable findings.(6) To overcome the hurdle of bias and support the complete, unbiased publication of studies, researchers are suggesting a mandate for clinical trials to register before recruiting patients so that the authors of systematic reviews know about all potentially eligible studies, notwithstanding their findings. It will also help if authors of systematic reviews ensured the assessment of the potential problems of publication bias in their review, thus considering methods for addressing this issue and confirming an inclusive search for both published and unpublished trials. (5) Finally, precise and robust measures, if taken by the scientific and medical organizations, ethical committees, regulatory bodies, journal editors, and the industry itself, will ensure that commercial interests of pharmaceutical companies do not weaken the knowledge of scientifically correct study planning, study execution, and publication.(2)
- Melander H, Ahlqvist-Rastad J, Meijer G, Beermann B. Evidence b(i)ased medicine: selective reporting from studies sponsored by pharmaceutical industry; review of studies in new drug applications. BMJ. 2003; 326:1171-3.
- Kerekovska A, Galunska B. Publication bias in clinical research sponsored by pharmaceutical industry. Scripta Scientifica Pharmaceutica. 2014; 1:7-13.
- Jefferson T. Sponsorship bias in clinical trials: growing menace or dawning realisation? Journal of the Royal Society of Medicine. 2020; 113(4):148-157.
- Canestaro WJ, Hendrix N, Bansal A, et al. Favorable and publicly funded studies are more likely to be published: a systematic review and meta-analysis. J Clin Epidemiol. 2017; 92:58-68.
- Hopewell S, Loudon K, Clarke MJ, et al. Publication bias in clinical trials due to statistical significance or direction of trial results. Cochrane Database of Systematic Reviews 2009, Issue 1. Art. No.: MR000006.
- Song F, Eastwood AJ, Gilbody S, et al. Publication and related biases. Health Technology Assessment. 2000; 4(10):1-105.
- McGauran N, Wieseler B, Kreis J, et al. Reporting bias in medical research – a narrative review. BioMed Central Trials. 2010; 11:37.