Bias essentially refers to an inclination, diversion, or deviation. In terms of epidemiology and statistics, bias refers to an error in the design or development of a study that leads to inadequate interpretations (deviation from the truth), as a result of a systematic error in the study. However, there are different classifications of study biases.^1,2

The clinical information of many studies is presented at various academic meetings. At the university level, recent publications in the literature are discussed at events such as journal clubs, and case thematic presentations, as well as bedside discussions. The principal objectives of these meetings are to teach the student to understand and make a systematic and accurate analysis of the medical literature; another objective could be to update students and professors on topics of common interest.^3-5 There is no consensus about the modalities of these meetings, and in some cases background questions are asked, while in other cases more elaborate Patient, Intervention, Comparison, Outcome and Time (PICOT) questions are used.

The fact that the student is being assessed involves having an adequate understanding of the article to share that knowledge with his/her peers and professors. The student shall then elaborate on medical, epidemiological, and statistical concepts of the study being discussed; however, being assessed may lead the student to make the decision of omitting some aspects of the study. This may be due to different reasons, including: (1) not wanting to be questioned about a topic he/she does not master; (2) not wanting to stand out showing an in depth knowledge of the article, delivering the missing data when asked; (3) to err in the reading and then in the transcription of the study, accidentally omitting information.

Any of these circumstances may result in a lack of veracity in the presentation of the methods or results of the trials, generating an inappropriate analysis of the results by the recipients of the information which may lead to a deviation from reality.

This is called transcription bias, since although there is a proper way to present the data in scientific articles and this same approach may be used to submit information thereof, occasionally the mode of presentation may vary and hence ignore important and relevant information for the comprehension and analysis of the article or subject matter. The student could argue that the information was not available in the article or in the research protocol (usually published in clinical experiments and systematic reviews); however, this rises some doubts about the reason for ignoring the information. Similarly, it must be emphasized that the meeting should predefine the strategies to ensure transparency in the development of the meeting.

How significant, methodical, and frequent is this limitation? One should try to establish the frequency, the type of students where this happens, under what circumstances, and also the severity of the impact, as this issue has not been previously described. It should be noted that bias is typically "systematic", which is easy to measure in this setting. Understanding that this situation may arise forces us to question the scope and the value of these academic meetings as tools for assessing the mode of presentation, the knowledge about epidemiology, statistics, and medical topics in particular. Consequently, it is possible to establish mechanisms to mitigate the effect of these deviations through controls, either by peers or professors that allow for the identification of omissions in the presentation, as a vehicle for a permanent update of all the participants involved.