Researchers must give the full picture

Human life and health are at stake when clinical research findings are presented in an incomplete, obscure or misleading manner. Many countries are therefore attempting to remedy the problems – despite considerable obstacles.

Medical and Social Science & Practice

The SBU newsletter presents and disseminates the results of the SBU reports, describes ongoing projects at the agency, informs about assessment projects at sister organisations, and promotes interest in scientific assessments and critical reviews of methods in health care and social services.

Reading time approx. 5 minutes Published: Publication type:
Share:

Man with binocularWhen researchers present their findings, they must often be selective, but the overall picture must still be accurate. All relevant studies must be openly reported with their main findings highlighted, regardless of whether they support or undermine a particular intervention.

This describes the ideal scenario. However, selective and biased reporting of results is in fact a major problem in treatment research. [1] Studies that fail to support the tested intervention clearly and through statistical reliability are sometimes published after considerable delay - or in the worst case scenario, not at all. Moreover, papers with unfavourable or unclear results rarely appear in prestigious journals and are cited less frequently by other researchers. Such skewing in the literature is referred to as publication bias and provides policymakers, professionals and patients with an overall distorted view concerning the benefits of treatment. [2] Distorting the overall picture in this way may influence healthcare decisions and lead to misallocation of scarce resources.
Such bias may also be noted in how researchers select what results to report from individual studies, known as outcome reporting bias. [2 ]For example, authors may choose to move the focus from the central outcomes in the study to more peripheral outcomes that have demonstrated greater impact. The consequence is an exaggeration of treatment effects. [3,5] It is therefore considered highly inappropriate for researchers to use their own findings to reformulate what was initially a central research question into one that is peripheral, and thereby instead give greater weight to less important outcomes. For example, if a treatment does not affect morbidity and mortality, and these were the central issues addressed by the study, the author of the article should not focus on isolated promising lab results.

Biased publication of findings, also referred to in the literature as spin, is nothing new in research. [4] Many possible reasons have been proposed, [5] including that manufacturers and sponsors want the research to benefit sales of their products. Another possible reason is that scientific journal editors and researchers are looking for exciting results that will draw attention and benefit their business and career.
However, biased publication is not the only reason that certain research results remain in the shadows. Some research reports are difficult to access. Grey literature, which exists alongside traditional scientific journals, sometimes remains invisible in mainstream research databases. [6] This category may include unpublished corporate information, academic theses from various countries, government studies and reports from agencies, authorities, regions and municipalities. Moreover, many scientific journals still charge hefty fees and lock articles behind paywalls, despite international and national open access initiatives.

However, many countries are fighting for greater transparency, especially in regard to publicly funded research, and for full disclosure of such results. The European Commission is working to achieve open science, including by facilitating access to and re-use of research data and findings. [7] According to the Commission, open science should promote transparency and publicly available findings.
The UK’s “Make it Public” strategy, launched by the National Health Services (NHS) in 2020, aims to ensure open access to research information by improving registration of studies, information to participants and reporting of results. [8] The NHS Medical Research Authority has adopted a ten-point action plan. See sidebar.

Critics in sweden have argued for clarification of requirements to be met by Swedish researchers when publishing final reports from clinical trials. [9] In January 2021, the organisation Transparimed and Cochrane Sweden issued proposals [9,10] pertaining to this area, including that Swedish authorities and research funders should endorse and adopt the World Health Organisation’s statement [11] concerning public access to clinical research results. In conjunction with the pandemic, the Swedish Research Council specifically endorsed [12] the British research fund Wellcome Trust’s call [13] for researchers, journals and research funders to ensure rapid sharing and open access to results and data relevant to management of the coronavirus. [14] In this context, the Swedish Research Council also recommended that publicly funded research data should be published openly online within a reasonable period after publication of the results – with the exception of cases such as copyright-protected data. [12]
As long as the affected policymakers, researchers, professionals and patients are only allowed access to selected portions of the scientific evidence, the risk that the overall picture will be biased remains. This situation could allow treatments that are actually ineffective or even harmful to erroneously be perceived as beneficial or cost-effective.

 

References

  1. Dwan K, et al, for the Reporting Bias Group. Systematic review of the empirical evidence of study publication bias and outcome reporting bias. PLoS ONE 2013;8:e66844.
  2. Page MJ, et al. Investigating and dealing with publication bias and other reporting biases in meta-analyses of health research: A review. Res Syn Meth 2021;12:248-59.
  3. Bruckner T. Clinical trial transparency: A guide for policy makers. Transparency International, et al, 2017. Downloaded from https://www.transparimed.org/reports
  4. Boutron I, et al. Reporting and interpretation of randomized controlled trials with statistically nonsignificant results for primary outcomes. JAMA 2010;303:2058-64.
  5. Chiu K, et al. ”Spin” in published biomedical literature: A methodological systematic review. PLoS Biol 2017;15:e2002173.
  6. Schöpfel J. Towards a Prague definition of grey literature. Twelfth international conference on grey literature: Transparency in grey literature. Downloaded from https://archivesic.ccsd.cnrs.fr/sic_00581570
  7. EU Commission. Open Science Policy Platform Recommendations, https://ec.europa.eu/research/openscience/pdf/integrated_advice_opspp_recommendations.pdf
  8. NHS Health Research Authority. Make it public. Transparency and openness in health and social care research. July 2020. Downloaded from https://www.hra.nhs.uk/
  9. Engelmark S. Stora brister i rapporteringen av kliniska studier. Tidningen Curie, Vetenskapsrådet, 2021-01-13. Downloaded from https://www.tidningencurie.se
  10.  Bruckner T. Results are missing for over two hundred drug trials involving Swedish patients. Article downloaded from https://www.transparimed.org/single-post/results-are-missing-for-over-two-hundred-drug-trials-involving-swedish-patients
  11.  WHO, https://cdn.who.int/media/docs/default-source/clinical-trials/ictrp-jointstatement-2017.pdf
  12. Vetenskapsrådet. Dela forskningsdata och resultat för hantering av Coronaviruset. News at www.vr.se Feb. 2020.
  13. Wellcome Trust. Sharing research data and findings relevant to the novel coronavirus (COVID-19) outbreak. Press release 31 Jan. 2020 at https://wellcome.org/
  14. Vetenskapsrådet. Samordning av öppen tillgång till forskningsdata. Statusrapport. ISBN 978-91-88943-33-0.
Page published