Other terms
- Biologic
- Biologic License Application
- Biologic Response Modifiers (BRMs)
- Biomarker
- Biomarker-Driven Randomization to Target Subjects
- Biomarker-Driven Therapies
- Biosimilars
- Birthmark
- BLA
- Blisters
- Blood-Brain Barrier (BBB)
- Blue Light Therapy
Bias
Bias is an unintended, systematic error of measurement, assessment, research process, interpretation, or reporting of data caused by the investigator’s preferences or preconceived beliefs.
Bias can exist in pre-trial planning, during the trial, and after the trial. Some major sources of bias in clinical research include: flawed study design, selection bias, interviewer bias, recall bias, transfer bias, performing bias, confounding, and publication bias.
One can avoid bias by taking steps to define risk and outcomes clearly, using standardized and blind data collection methods, randomizing subjects, using objective data sources or verifying subjective sources, planning for lost-to-follow-up, and controlling for known confounders. Unlike random errors, the impact of bias may not be reduced by increasing the sample size.
Why is understanding bias important?
Bias in research encourages one outcome over another even if it is incorrect. Understanding bias can help investigators identify and avoid creating errors during the research process. It can also assist treatment providers in critically interpreting scientific literature and making the best choices for their patients.
Source: Meuli, L., & Dick, F. (2018). Major Sources of Bias. European Journal of Vascular and Endovascular Surgery, 55(5), 736. https://doi.org/10.1016/j.ejvs.2018.02.017
