Confirmation bias and quantitative approach in psychiatry: should ideological competing interests be declared?
Front. Psychiatry. 2024-04-24; 15:
DOI: 10.3389/fpsyt.2024.1365733
Lire sur PubMed
Blay M(1)(2), Gauld C(3), Espi P(3), Falissard B(2).
Author information:
(1)ADDIPSY, Outpatient Addictology and Psychiatry Center, Santé Basque
Développement Group, Lyon, France.
(2)Université Paris-Saclay, Université de Versailles Saint-Quentin-en-Yvelines
(UVSQ), INSERM, Centre de Recherche en Epidémiologie et Santé des Populations
Team « DevPsy », Villejuif, France.
(3)Hôpital Femme-Mère-Enfant, Hospices Civils de Lyon, Lyon, France.
DOI: 10.3389/fpsyt.2024.1365733
PMCID: PMC11077325
PMID: 38721611
Conflict of interest statement: 4We are aware that this proposition is
provocative and may be difficult to implement in real life. Researchers have
necessarily their own opinions, and the boundary is subtle between ideology and
personal theoretical or political opinions. Thus, many practical challenges can
be associated with the implementation of such a policy. For example, there may
be a risk of stigmatization of certain viewpoints leading to discouragement of
researchers from exploring controversial topics. Indeed, if working on a “hot”
and polarizing topic may include personal opinions’ disclosure, researchers
having one opinion but evolving in a context favorable to another may start to
withdraw from this specific topic exploration to not suffer any consequences
(like rejection from social or academic spheres), which could in turn enhances
the risk of publication bias (with only research matching the context’s opinion
being conducted). However, this risk could be decreased by restricting the
content of this disclosure only to the researchers’ preconceived opinions on the
actual subject of the study, without compelling them to fully describe all their
personal opinions outside of the research subject. Also, there may be a risk of
increasing the actual global mistrust in psychiatric science, both from public
and practitioners [e.g., (33)]. Indeed, if ideological interests are disclosed,
and if the results found are in line with these interests, readers may be prone
to over conclude that these results are inherently biased, especially when
considering that these readers may also be prone to confirmation bias, and when
considering the overall low level of knowledge on how research is conducted and
statistical analyses work (34, 35). If this risk should be carefully considered,
we also believe that this mistrust partly relies on the actual non-disclosure of
this type of competing interests that have long been the subject of concerns,
and providing a way to address these concerns may thus be a way to enhance the
readers’ trust. Finally, given that the present paper can be seen as a practical
example of confirmation bias and ideology, we wanted to lay the foundation stone
of ideological competing interests’ disclosure. Indeed, the four authors have
their own opinions regarding these topics, and notably on the risk of influence
of ideologies on the conduction and production of science. Thus, the examples
and the references we chose can be seen as subjective, partials, and selected to
support our preconceived opinions. We described the latter in the Author’s Note
section. Overall, we hope that this article will contribute to pursue the
reflection on this important topic.The authors declare that the research was
conducted in the absence of any commercial or financial relationships that could
be construed as a potential conflict of interest.