Reading between the lines: what does “a study has shown…” really mean?

We often hear or read ”a study has shown…”. However, what initially sounds convincing or serious, needs to be looked at more carefully, since there are many different types of studies that differ not only with regards to their design but also to their level of scientific evidence.

Whether a health measure can be recommended or not, for example, cannot be decided based on the opinion of a few experts only. In the age of evidence-based medicine, it needs to be supported by the best scientific evidence available. Many studies have been – and are being – carried out to determine this evidence. To evaluate the level of evidence of the various studies, it is helpful to know the most important types of studies, their strengths, and limitations. This is even more important when dealing with the health effects of alcoholic beverages. It is therefore critical for Wine in Moderation to base its communication about wine and health-related issues on the best available scientific evidence.

The most important study types in health research

Out of all the available types of studies and scientific key figures, the glossary of evidence-based medicine alone is more than 60 pages. In this article, we will mention only the most important ones for the topic wine and health issues.

In health research and nutritional medicine, epidemiological studies are predominant. Epidemiology is a scientific discipline which deals with the dissemination, consequences, possible causes and factors of risk concerning health relevant questions in a population (group):

• Cross-sectional study: These studies can only give first clues for possible associations since they are only a snapshot – similar to a holiday photo.
• Case-control studyA case-control study is a study comparing two groups: one with and the other without a certain disea...: where for example, the drinking pattern of patients with or without a certain disease is compared. As for the cross-sectional ones, they can only be a first hint, since they are prone to errors.
• Prospective cohort study: the most meaningful among all the studies since the data are collected over a period of time / follow up and not retrospectively or only once. For example, these studies can investigate whether there is a statistical relationship (correlation) between wine consumption and the later risk for diabetes. When the results show that moderate wine drinkers have a lower risk of diabetes, this is an important indication for a possible health advantage of moderate wine consumption.

Lastly, epidemiological studies can be described as observational and do not allow conclusions about a cause-and-effect relationship.

Studies with a higher level of scientific evidence

To analyse the cause and effect, interventional studies (clinical studies) are necessary. These studies examine the effects of particular changes.

• Placebo-control studyA placebo-control study is a study comparing the cause-effect in two groups, where one receives the ...: It can be, for example, investigated whether a particular change in the lifestyle such as the eating or drinking pattern, has an effect on health. Ideally in such studies, a control group exists, which does not have these changes or receives a placebo. However, the implementation of a placebo-controlled or blinded interventional study in the area of wine and health is next to impossible, since the participants would taste the difference and would know in which group (interventional or control) they are.
• Randomised-controlled trial a gold standard among the studies, where the participants are randomly assigned to the control or interventional group. This system reduces possible errors and increases the scientific evidence level.
• Systematic reviewA systematic review is an analysis comparing and summarizing the data from different studies on a ce...: the next higher level, where the results of several individual studies are summarized according to set criteria.
• Meta-analysisA meta-analysis is a method of summarizing the results of the same research purpose and comprehensiv...: when both the results and the data of individual studies are summarized and re-analysed together. This type of study has the highest-level of scientific evidence, especially when it is a meta-analysis of RCT’s, the gold standard of all studies.

Saying that there is a study for a particular topic is not sufficient for the level of evidence. The type of study needs to be known.