Scientific research is the foundation for which dental and medical treatments are based. Evidence-based dentistry is the practice of performing treatment shown by the literature to be of a particular benefit to the patient.  Providing dental treatments based on anecdotal evidence not supported by scientific research can be ineffective and dangerous.

When reading scientific research, it is important to not take anything at face value.  The type and design of the study has huge implications on the validity of the conclusions presented.  Well-designed randomized clinical trials have the highest level of evidence due to their strict control of variables.  Animal studies have a lower level of evidence due to their questionable application to human beings.

One study design that seems to present very convincing conclusions, but must be carefully considered is the cross-sectional study.  These studies are most often surveys asking people a series of questions.  They can be used to establish a correlation between two or more factors, such as eye color and whether or not you own a motorcycle.  Perhaps more people with brown eyes own a motorcycle than people with green eyes.  Although a trend may be seen in the data, it does not mean that there is a causative relationship between the factors.  

The findings of these studies are passed off as facts and widely distributed throughout the media and internet.  Without well-designed research examining these possible relationships, no causation can be established.  Accepting survey data as fact can lead to potential harmful situations, such as when a patient declines certain dental medical treatment in lieu of unproven remedies.

John Oliver makes some good points:

John Oliver discusses how and why media outlets so often report untrue or incomplete information as science.