Headlines blare that a food will cure cancer, a dietary pattern will help you live longer, or a nutrient will cause disease. But what does the science say?
Rather than worrying about whether a headline might be misinformation or disinformation, first we have to know what the information is. The original research article (or at a minimum its abstract) is the source of this information and as clinicians, you can help your patients and clients understand the findings. To evaluate news articles or research studies, I often start by figuring out a study’s PICOS (see below) elements to help me draw my own conclusions.
PICOS elements describe an idea in evidence-based medicine to help find scientific evidence that can help with a particular patient or client. However, the approach can be used more broadly in understanding what a study can conclude. The acronym stands for:
- Patient, Population, or Problem – Who or what is the study trying to address? Factors may include age, sex, disease status, region, and others. Ideally, the evidence will match who you are interested in (including yourself!). Studies involving animals, cells, or computer simulations can be very useful in trying to learn what happens in people, but they do not directly show effects in humans.
- Intervention (or exposure) – What is the treatment or exposure being looked at? In the realm of food and nutrition, is it a dietary pattern, a nutrient, or a particular food? Is it a medicine or other treatment? Maybe it is even a social determinant of health, like being ‘exposed’ to poverty? Another facet to consider is how it is measured. Is there an objective measurement of a food being consumed, or were people asked to record what they ate, or were they asked to remember what they ate over the past year? These factors affect how much you can trust that the exposure being studied represents what you are actually interested in.
- Control or Comparator – What are you comparing the intervention (or exposure) to? When it comes to nutrition, there is rarely a ‘placebo’ unless someone is studying a nutraceutical or a supplement. Scientists compare the intervention/exposure diets, foods, or nutrients to other diets, foods, or nutrients. If they did not, they would compare against nothing. In many cases, comparing against nothing means that there is a difference in calories. Consider a study that adds a food; that food has calories and a variety of nutrients. The control could be equal in energy, or matched for protein content. But if the research group did not give the control group anything, then the control may be lower in energy. Whenever someone makes a claim about nutrition, always ask, “Compared to what?”
- Outcome – What is it that we expect to change? The outcome is what is actually measured in the study. Yet, study outcomes do not always match the real outcome we are interested in. We hope that blood sugar helps us understand diabetes, but unless diabetes is actually diagnosed it is a leap or an extrapolation to conclude about changes in diabetes from short term changes in blood sugar. Acute food intake may not predict long term weight change. Changes in blood lipids may not predict long term cardiovascular disease. Changes in cognitive function tests may not predict Alzheimer’s disease. Self-reported values are not necessarily accurate. For example, asking someone their weight does not always result in ascertaining their actual weight. Always consider how the outcome was measured and whether it matches the outcome you are interested in.
- Setting (and Study Design and Timing and, sometimes, everything else about a study) – Where was the study done: in a lab, in a clinic, or with a free-living population? How long was it: a few hours, days, weeks, years? Was the study randomized in which interventions were directly given to participants? Was it observational, in which people were just watched over time? All of these factors influence the relevance of the evidence to the question you want to know the answer to.
Other tools and concepts, like hierarchies of study designs and grading of bodies of evidence, can help you evaluate the totality of evidence rather than just any single headline. This can be important when determining whether there is a cause-and-effect relationship by asking about the strength of the study design. At a minimum, PICOS elements can help you consider these important things:
- What do the researchers believe is a cause (and compared to what)?
- On what outcome is it thought to have an effect?
- To whom does this apply?
Next time your patient or client shares an article claiming, “New diet will cause you to lose weight”, you can help them translate that headline into PICOS elements to understand what the research actually indicates. When you look at the actual study, you may find out that “A diet that has been around for a long time (not a new diet) was associated with (did not necessarily cause) rodents (not you) to eat less in a single setting (no weight measured).”
This blog is sponsored by SNI Global and U.S. Soy.