Science is constantly shaping how we attribute risk to certain activities, our food consumption, our use of medications, and much more. The science around human health is constantly evolving as our understanding of research and science changes.
It’s important to keep up with the many areas of science that are evolving, especially those that relate to human health. We want to understand how the findings of the latest research can impact human health, but our changing information landscape may lead consumers to misconstrue the facts.
It may seem like every type of food is related to a health risk or a negative health outcome. Every so often, new studies, articles, and the media claim that there is a new risk with eating a certain type of food.
This can lead to confusion and fear about what eating certain foods may do to people’s health. Some ‘scientific studies’ have made people fear that the overconsumption of coffee might lead to a heart attack, eating bacon is as dangerous as cigarettes, dairy products will clog arteries and lead to heart disease, and so on. While some of these individual claims are untrue, speaking about food-related risks is much more complicated than any one statement.
The information ecosystem
The likelihood of experiencing negative health effects from eating foods regularly associated with health risks tends to be relatively low. For example, the World Health Organization announced in 2015 that evidence strongly indicates that eating bacon and other processed meats can contribute to colorectal cancer.
However, Cancer Research UK wrote that the risk of developing colorectal cancer is relatively rare. For people that rarely eat meat, there is a 5.6 per cent risk of developing the disease, and for frequent meat eaters, the risk rises to about 6.6 per cent. Therefore, while eating processed meats has risks associated with it, giving up these foods completely and opting for ‘healthier’ options may not lead to much change in your health either.
Breaking down bits of data from nutritional studies and accurately contextualizing them when they’re used in media can provide more insight into the relative risks of consuming certain foods. So why do we see a lot of hype about food-related health risks in the media? In an era when many online media publications are driven by clicks for articles, website page traffic, and social media engagement, attention-grabbing and misleading headlines are unfortunately too common.
In the age of the ‘infodemic’ — a rapid and far-reaching spread of accurate and inaccurate information — it is vital to responsibly report information and statistical data. Information can easily become convoluted, intentionally or unintentionally.
A major risk with improper representation of and reporting on food-related risks arises when information enters the realm of the internet and social media. Research has found that false information can spread faster than accurate information on Twitter. When misinformation is so prevalent, especially as it relates to health information, reporting accurate statistics and information related to human health is paramount.
Reporting on the findings of nutritional studies responsibly requires looking at how strong a given correlation is. However, there is not always a strong incentive to do so — if the media did report statistics with appropriate context and report the statistical data meaningfully, would these be understood by the average reader?
Reporting statistical data
Many foods are associated with health risks, but statistical literacy plays a vital role in reporting these risks meaningfully and ensuring that they are understood accurately by readers.
Many publications run into the pitfall of assuming that the correlation between certain foods and negative health outcomes is causal. Conflating correlation and causation with nutritional research findings can take findings about health risks out of context. When nutritional research finds that a certain food is associated with a negative health outcome, it means that people might be at an elevated risk of developing that outcome, not that eating that food will necessarily lead to a given individual developing that negative outcome.
There is also the grey area of statistical significance. Statistical significance indicates that there is evidence to conclude that the observed relationship is not simply chance and that it exists significantly in the population. However, this still does not address causality. While statistical significance is deeply ingrained in scientific research, many researchers and statisticians are calling for scientists to abandon its widespread use.
The calls to drop the use of statistical significance are not asking that this tool be done away with completely, but saying that it should not be used as a catchall measurement of significance. Using an arbitrary numerical threshold to decide whether hypotheses are accepted or rejected also determines which studies are published and how information is marketed, especially as it relates to products and consuming foods. Using an arbitrary value as a determinant of what is true can lead to research being biased and important observations being overlooked.
It is also important to consider the accuracy of the source of the nutritional facts reported in the media. Unfortunately, nutrition research and the methodologies used to explore related questions are not immune to flaws.
While randomized controlled trials (RCTs) are considered the gold standard in scientific research, they can be unrealistic for exploring nutrition questions. First and foremost, carrying out long-term RCTs is expensive. It is also difficult for researchers to ensure that participants in the experiment or control group are following their assigned diets. RCTs might be more feasible for assessing short-term outcomes related to specific foods or diets, but this can lead to gaps in knowledge on the long-term health effects. At this point, researchers are relegated to inferring these impacts based on the short-term data.
Observational studies are another common study design in nutrition research and have been valuable for learning about health-related risks for things like smoking. They can run for many years and track a large number of people, but, unlike RCTs, they are not controlled, which can lead to observations being influenced by confounding factors or external variables.
Conflicts of interest are also a major issue in nutrition research, as food manufacturers can be involved in funding nutrition research studies. Have you heard of the conspiracy theory that the sugar industry wanted to implicate saturated fats as the cause for heart disease instead of sugars so they paid scientists?
Well, this is actually a true story. In 1965, the Sugar Research Foundation funded research on coronary heart disease. This led to a literature review published in the New England Journal of Medicine that focused on fat and cholesterol as the dietary causes for coronary heart disease and downplayed the consumption of sugar as a risk factor.
This is unfortunately not an isolated incident in nutrition research. This pattern has led to many studies investigating these conflicts of interest, which have found that research funded by food manufacturers often produces more favourable study results. Marion Nestle, a nutrition professor at New York University, tracked studies funded by food and beverage companies and trade groups in 2015 and found that 168 nutrition studies were funded by these groups. Nestle found that 156 of those studies contained biased results that favoured the interests of the food manufacturers funding those studies.
The role of science communication
It may seem like a challenge to know what to believe about food-related risks, but there are ways to accurately talk about food-related health risks meaningfully and without sensationalizing new discoveries. Good science communication is important in ensuring nutrition research findings are reported accurately and appropriately.
When reporting on research findings from nutrition studies, it is important to evaluate who is funding the study to discern potential conflicts of interest. Consumers of information also need to be comfortable with the changing information landscape.
Science is always evolving — which is a good thing, because it means that we are learning and hopefully adapting to new information accordingly. However, we should exercise caution before latching on to new discoveries on food-related health risks. When it comes to emerging findings, it can be helpful to look at the accumulated information and what is already known in that area.
Rather than reporting a food-related health risk based on a single study, looking at multiple studies or systematic reviews and meta-analyses can help to talk about risks meaningfully.
When reporting on food-related health risks, writers should include information on the study population and the context that these risks were extrapolated from. Context is important because food and diets will not impact everyone in the same way. If research is conducted on a non-representative sample, the reporting of these findings needs to reflect this. Contextualizing who these findings may apply to and being careful with how they are generalized are important to ensure nutritional studies are not taken out of context.
Science and research is complex, but this complexity shouldn’t detract people from pursuing complex questions and talking about exciting new findings. This should encourage the people involved in conducting research and communicating scientific findings to be empowered to critically analyze research methods and sources of information, and strengthen how we conduct research and communicate science.