Type of Evidence

In determining the validity of food folklore, not only the totality of evidence but also the type and quality of available evidence are important. As shown in

Table 3 Folklore: separating fact from fiction g n ro lore Str kl aUnlikely to be a major threat ^Potential threat to public health

Table 4, the strength of the association between eating a food (cause) and a health outcome (effect) can be ranked according to the type of evidence presented. The best evidence comes from studies that have the most control over the claim or treatment being evaluated and eliminate other factors that may suggest an effect was present, when really it was not. Although randomized double-blind placebo-controlled clinical trials are considered the 'gold standard' for determining diet-health relationships, such studies are rarely available. Lesser levels of evidence must usually be used.

Randomized double-blind placebo-controlled trials When randomized double-blind placebo-controlled trials show a relationship between a specific food and a health effect, the evidence is considered to be very good. These studies exert rigorous control over the claim or treatment being evaluated and over the people who are subjected to

Table 4 Ranking the quality of the evidence it (by randomization) and the assumptions of both the experimenters and the participants (by placebos and blinding). Multiple studies of this type, with an expert review of all other types of data, are considered to be the 'gold standard' for establishing cause-effect relationships. Other types of evidence and studies are lower in the hierarchy, because they are not as definitive in identifying true cause and effect.

Although single randomized trials and non-randomized clinical studies are somewhat less definitive, they are still valuable, because they also permit control over the treatment being evaluated. Often, however, these studies are not large enough, or the study sample is not representative, so the results cannot be generalized to the problem at hand. Other factors that may weaken these studies are not counting of dropouts, lacking or unconvincing placebos, and inappropriate points or biomarkers serving as surrogate end points.

Observational studies Human studies that involve observation rather than direct intervention provide evidence that is satisfactory but less conclusive. These studies are designed to test a relationship between an exposure of interest (folk belief) and a health outcome. Observational studies include both cohort studies (prospective) and case-control studies (retrospective). In a prospective study, a group exposed to the treatment of interest and an unexposed group are followed forward in time. The health outcomes in both groups are observed and evaluated after controlling for confounding factors with the use of statistics. In contrast, retrospective studies compare individuals who have already developed an outcome of interest (case) against those who have not (control). Factors contributing to the development of the outcome are then determined by looking backward in time. Because observational studies cannot be precisely controlled, it is more difficult to establish cause and effect. However, when confounding factors can be adequately controlled for, these studies provide suitable evidence to support diet-health relationships.

Uncontrolled clinical trials Clinical studies in which everyone is treated, in which only those who ask for the treatment are treated, or in which some are treated based on unsubstantiated clinical convictions are suspect. In such studies, no randomization occurs and neither the researcher nor the participant is blinded to the treatment. Therefore, it cannot be determined whether the treatment is actually the cause of the observed results or whether biased

Highest quality

Randomized double-blind

placebo-controlled trials

Observational studies

Prospective studies

Retrospective studies

Uncontrolled clinical trials

Patient reports/Case studies

Non-human trials

In vitro studies

In vivo studies

Lowest quality

Scientific evidence Strong Weak

Table 3 Folklore: separating fact from fiction

Scientific evidence Strong Weak g n ro lore Str kl

Fact3

Fiction/ undetermined^

Emerging science3

Fiction3

aUnlikely to be a major threat ^Potential threat to public health convictions of both experimenters and study participants are falsely contributing to the results. Better evidence is needed before it can be stated with assurance that the folklore based on such observations is true.

Patient reports, case studies, and folklore Even weaker human evidence of cause and effect comes from single medical case reports and anecdotal evidence. These types of evidence are also biased since those who experience success from the treatment are much more likely to report their stories than those who do not.

Animal studies and laboratory experiments Nonhuman studies involving living animals (in vivo studies) or tissue cultures (in vitro studies) are useful in providing information on the possible biological plausibility, dose response, and action of a treatment. However, their ability to predict outcomes in humans is poor. Therefore, these studies are unconvincing and should be used only to support other types of evidence.

Breaking Bulimia

Breaking Bulimia

We have all been there: turning to the refrigerator if feeling lonely or bored or indulging in seconds or thirds if strained. But if you suffer from bulimia, the from time to time urge to overeat is more like an obsession.

Get My Free Ebook


Post a comment