Publication bias
Publication bias refers to the tendency of researchers and publishers to make decisions about sharing research findings based on the outcomes of the studies. This bias often manifests when researchers opt not to submit articles about studies that yield negative or inconclusive results because they deem them unimportant, or when publishers choose not to include such studies in their journals. The history of medical publishing dates back centuries, evolving from individual observations to specialized journals that share findings on various medical and non-medical topics.
Bias in publication can arise intentionally or unintentionally, influencing the overall understanding of research effectiveness. For instance, if studies showcasing negative results are underreported, it could lead to an inflated perception of the efficacy of a treatment or drug. Factors contributing to publication bias include time constraints, funding pressures, and the prestige associated with publishing in high-profile journals. Additionally, biases can occur against research from certain countries or in different languages, further complicating access to comprehensive knowledge. Recognizing publication bias is crucial for researchers and practitioners as it can skew the interpretation of data and ultimately impact patient care and scientific advancement. Efforts to mitigate this bias, such as requiring the registration of trials, are seen as steps toward ensuring a more balanced and representative body of research.
On this Page
Subject Terms
Publication bias
Publication bias occurs when researchers or publishers make decisions on whether to share printed information concerning research based on the outcome of the study. Researchers often submit articles to professional journals to share the results of their studies. However, researchers sometimes decide not to write up and submit the information because the results do not seem significant, do not turn out as they expected, or some other reason. Sometimes even when information is submitted, publishers do not include it in their publications because of the content or source of the article. In either case, the research does not get shared because of publication bias.
Background
Medical practitioners have shared their findings in writings for centuries. Originally this was done through book-length writings that contained observations about numerous medical conditions. A physician would write books that contained only his own observations, and people who wanted to learn from the findings would have to sort through all the material to find the relevant sections for a particular condition or treatment.
This began to change in the seventeenth century. As physicians and scientists became increasingly interested in and devoted to applying formal practices for observing patients and experimenting with treatments, they began sharing their experiences on individual diseases, injuries, or treatments. Medical journals came into existence in the 1600s, as people gathered together findings from different physicians and compiled them in a magazine-type format that physicians could then access as a resource for caring for their own patients.
Physicians would often treat a patient, observe how the patient responded to that treatment, and then report what happened in a journal article. Eventually, journals began to specialize in particular types of conditions or treatments. For instance, journals came into existence dedicated to general health concerns, cardiac health, neuroscience, etc. A similar practice was followed with non-medical research. Scientists in sociology, archeology, and other fields would conduct an experiment, record their findings, and then write articles to share what they had learned.
Overview
Bias is defined as exhibiting some form of prejudice for or against someone or something. Bias can be intentional, such as not hiring someone of a certain race, or unintentional, such as consistently choosing one brand of peanut butter over another. People can be biased toward cats over dogs, for instance, without having any specific negative feelings about dogs. However, biases affect how decisions are made and can have negative consequences on the outcomes of those decisions.
In the case of journal articles, bias arose in several forms almost as soon as journals existed. Sometimes, researchers created their own bias by deciding not to publish an article about their findings. This can happen for a variety of reasons. It could be that the researcher does not think the information he uncovered is important enough. This is often the case when an experiment or attempted treatment does not work or does not work the way the researcher intended. For example, if a physician attempted a new surgical procedure on a patient with a hard-to-treat condition and the patient did not benefit from the surgery, the physician might decide not to publish anything about the attempt. However, this bias against negative findings deprives the medical world of the knowledge that was gained; in this example, that would be the knowledge that the particular procedure was ineffective. As a result, others might duplicate the same procedure because they do not know that it has been tried without success. In this case, bias against publishing also means that others do not know about the failed attempt, and deprives them of the opportunity to build on the original failure and try a new approach. Another possible consequence of not publishing studies with negative findings is that it can skew the analysis of other studies. For example, if ten studies are done testing the effectiveness of a new drug and two show promising results but eight show no results, but only one of those eight studies is submitted for publication, the drug could appear to be more effective than it really is.
There are many reasons why researchers may create this bias for not submitting an article. They may lack time or not want to waste time on publishing negative findings when their time could be put to use doing new research. The funding source for the study may also be an issue; companies funding studies because they hope to promote a new product, such as a drug or medical device, are likely to discourage sharing information from research that does not show that product in a positive light.
Another form of publication bias occurs when the researcher chooses to send studies with positive results to more prestigious or larger journals and send negative findings to lesser known publications with smaller audiences. Researchers can gain notoriety and their own prestige for having new breakthrough studies published in important journals. As a result, they may choose to submit their articles to more notable journals when the findings are positive but choose journals with smaller distribution to report negative or less significant findings.
The journal publishers can also contribute to publication bias. They may choose to publish larger studies with potential breakthrough treatments or information over those with negative findings or those that simply confirm previous research. There can also be bias against articles written by researchers from different countries, or against those who write in certain languages. Another type of bias occurs when articles from smaller publications are not included in databases that compile articles for distribution to physicians or other scientists.
Regardless of the source or form of bias, publication bias hampers the effectiveness of those who are the end users of the research. As a result, those who use journal articles as a source of information are advised to analyze the information in articles, being wary of situations in which the findings from multiple studies are similar. This could indicate that bias is present and opposing findings have not been published. Since medical trials must be registered, it is also possible to compare the number of registered trials against the number of published studies from completed trials to uncover potential bias against opposing findings. Experts believe that requiring similar registration for other studies and experiments for which researchers intend to publish their findings would help limit bias.
Bibliography
Ingre, Michael, and Gustav Nilsonne. “Estimating Statistical Power, Posterior Probability and Publication Bias of Psychological Research Using the Observed Replication Rate.” Royal Society Open Science, 12 Sept. 2018, rsos.royalsocietypublishing.org/content/5/9/181190. Accessed 1 Dec. 2018.
Joober, Ridha, et al. “Publication Bias: What Are the Challenges and Can They Be Overcome?” Journal of Psychiatry and Neuroscience, vol. 37, no. 3, May 2012, pp. 149–52.
Marta, Monica Mihaela. “A Brief History of the Evolution of the Medical Research Article.” Clujul Medical, vol. 88, no. 4, 2015, pp. 567–70.
Montori, Victor M., et al. “Publication Bias: A Brief Review for Clinicians.” Mayo Clinic Proceedings, 2000, www.mayoclinicproceedings.org/article/S0025-6196(11)63030-9/fulltext. Accessed 1 Dec. 2018.
Peplow, Mark. “Social Sciences Suffer from Severe Publication Bias.” Nature, 28 Aug. 2014, www.nature.com/news/social-sciences-suffer-from-severe-publication-bias-1.15787. Accessed 1 Dec. 2018.
Song, Fujian, et al. “Publication Bias: What Is It? How Do We Measure It? How Do We Avoid It?” Open Access Journal of Clinical Trials, vol. 5, no. 1,July 2013, pp. 51–81.
“What Is Bias?” Psychology Today,www.psychologytoday.com/us/basics/bias. Accessed 1 Dec. 2018.
“What Is Publication Bias?” Statistics How To, www.statisticshowto.datasciencecentral.com/publication-bias/. Accessed 1 Dec. 2018.