How Do We Know If Supplements Work?

A recent research review sought to eliminate any confounding variables that create bias in supplement studies.

Supplements occasionally get negative results when studied, and we end up questioning whether or not they actually are healthy. This leads to a quandary posed by many people, health professional and average Joe alike: “How do I know the truth, when there’s so much conflicting information?”

In a recent review published in the Nutrition Journal, researchers were interested in what supplement users do during their daily lives. On the surface this might seem confusing or unimportant, but when it comes to lending the support of statistics to science, this kind of study couldn’t be more necessary. Let me explain why.

One of the most important goals in the scientific process is to try to eliminate confounding variables. Without getting too deeply into statistics, a confounding variable is more or less a factor that is important to a study’s results, but is left out of the research because the variable is unknown or underestimated. If a confounding variable is present in a research study, it creates what is called an omitted-variable bias. This bias is exactly what it sounds like. The conclusion to the test is potentially flawed because of the missing variable.

In many studies about health, it is not uncommon to be concerned with confounding variables. Vitamins are a common example of this phenomenon. If people take vitamin supplements, they’re probably health conscious to begin with. If that’s true, perhaps reports of improved health from vitamin use are actually because of the otherwise healthy lives that the users live.

In the case of the Nutrition Journal study, the researchers were looking to find out about the behaviors of supplement users compared to non-users. Specifically, they were interested in variables known to affect health. If we know what regular activities supplement users do that are different from non-users, we also know the confounding variables. We can then review the existing literature or create new scientific studies that have greater statistical strength backing them.

Potentially confounding variables and other considerations were found in the review:

  • Supplement use was more common as people aged, and more common in women.
  • Supplement users also had slightly better diets, especially as they got older. That’s important because it’s possible for people to get too many nutrients, and could account for some negative results found from supplement use. However, the researchers found that supplement use nevertheless improved nutrient intake in general.
  • Supplement users were more likely to exercise and have a healthy body weight, and less likely to smoke.
  • The researchers also noted that heavy supplement users were healthier than non-users and users who took fewer supplements, even over a twenty-year span of time. However, this could be due to greater intensity of the confounding factors listed above.

When it’s all said and done, this review and others like it provide a springboard for examining the strength of past and future research. We can answer that oh-so-common question of “Who can we trust?” by seeing for ourselves how robust the research is. You have to love science.


1. Annette Dickinson, et. al., “Health habits and other characteristics of dietary supplement users: a review,Nutrition Journal 2014, 13:14

Photo courtesy of Shutterstock.