Case reports/series & bias considerations

Abstract

It was French philosopher Claude Lévi-Strauss who correctly stated, “The scientist is not the person who gives the right answer, (they're) the one who asks the right questions”.1Claud Levi-Strauss quotes. Brainy Quote. Available at: https://www.brainyquote.com/quotes/claude_levistrauss_154677. Accessed May 24, 2021.Google Scholar The essence of good scientific research hinges on asking insightful questions and then developing the proper methodology to test and answer them. Unfortunately, bias is often unintentionally introduced into the scientific research process. Cambridge Dictionary defines bias as, “Any trend/deviation from the truth in data collection, analysis, interpretation and publishing, which can cause false conclusions”.2Bias. Cambridge Dictionary. Available at: https://dictionary.cambridge.org/us/dictionary/english/bias. Accessed May 24, 2021.Google Scholar So, what forms of bias exist? Over the past 20+ years of practice, I have gathered an ever-growing list of the various forms of bias in scientific research. Unfortunately, I do not have the specific references for each of these; however, I do not claim any of the following definitions are original to me.•ADMISSION BIAS: A form of VOLUNTEER BIAS (see below); the population being studied does not reflect the general population that should be considered.•CONFIRMATION BIAS: Information that confirms your preconceptions/beliefs and avoids/undervalues/disregards information inconsistent with your own beliefs.•DATA ANALYSIS BIAS: Fabrication of nonexistant data, abusing data by eliminating outliers that do not support your hypothesis, and/or manipulating the data by using inappropriate statistical tests or “fishing” for tests that will support your desired outcome.•DATA INTERPRETATION BIAS: Discussion of observed differences that were not statistically significant (a.k.a. ”borderline significant”) or may be statistically significant but are clinically irrelevant. Also includes extrapolation of data outside the range of data actually studied and overgeneralization of study conclusions to general population.•FUNDING BIAS: Prevailing studies come from one company who funds them, and the results/outcomes consistently support the companies own marketing interests.•HINDSIGHT BIAS: Researcher(s) convince themselves after an event has already occurred that they had accurately predicted it before it happened.•MISCLASSIFICATION BIAS: Occurs with poorly defined pathology that is not easy to precisely diagnose and has no “gold standard.”•PUBLICATION BIAS: Journals are much more likely to publish manuscripts presenting a positive or neutral outcome than a negative one.•RECENCY BIAS: Tendency to project recent experiences of observations into the future causing predictions about the future to be based on events of the recent past.•SAMPLING/ASCERTAINMENT BIAS: Some subjects are systematically more likely to be selected into a sample population than others thereby limiting generalizability.•SELECTION BIAS: Unequal representation of subjects being studied.•SURGEON/CENTER OF EXCELLENCE BIAS: Outcomes for design surgeons/centers of excellence are skewed towards the positive even in complex situations so the outcomes are not generalizable to the common surgeon/facility.•SURVIVOR BIAS: A form of SAMPLING BIAS; patients/data points that make it past the selection process are focused on and those that did not are overlooked/ignored.•VOLUNTEER BIAS: Over recruitment of one type of subject/population being studied. What does this have to do with Foot & Ankle Surgery: Techniques, Reports & Cases (FASTRAC)? Well, think back to interesting cases you were involved with that might be worth publishing or you in fact published. Now think of how you would or did construct the manuscript. Now think of the bias you potentially introduced. Stuck? By definition, a case report has ADMISSION and SELECTION bias because the subjects included are considered outliers by being unique and have been selected specifically for this reason. CONFIRMATION, HINDSIGHT, PUBLICATION and RECENCY bias all exist because you are personally responsible for the subject's positive outcome, which has already occurred in the past and has been confirmed by a recent follow-up visit. SURVIVORSHIP and VOLUNTEER bias exist because you selected the patient to undergo the surgery based on their uncommon pathology present. So, now what? Well, after reviewing myriad manuscript submissions to FASTRAC, I have recognized while most authors present the general weaknesses of the case presented in the discussion section, to date, not one author has presented their attempts to control for bias or present the questions the readers should be asking about the real or perceived bias present. In order to facilitate this process, I have added to the FASTRAC “Guide for Authors” (https://www.elsevier.com/journals/foot-and-ankle-surgery-techniques-reports-and-cases/2667-3967/guide-for-authors) the need for the authors to specify what attempts were made to control for bias. If the authors were unable to control for bias, then they'll be asked to present questions they believe the reader should consider prior to addressing the same or similar pathology with the approaches presented. As a result, the readers of FASTRAC will be able to follow the teachings of Claude Lévi-Strauss and simultaneously benefit from the good scientific research contained within the manuscript that provides us with the right answer while simultaneously being reminded to address bias by asking the right questions. Be well, Tom The author declared no potential conflicts of interest with respect to research, authorship, and/or publication of this article.