Every week there is a new headline claiming that a certain food is either really good for you or really bad for you.
And since these headlines are published to the general public who for the most part have no experience in critically analysing scientific data, I think it is important to share the following problems with scientific nutrition studies and why you should take headlines with a massive pinch of salt.
1. Most nutritional science is based on observational studies.
What that means is that scientists take a bunch of people with a certain condition – say being overweight – and take a look at what they eat, then try to connect the dots.
The problem with this type of research is that we are either trusting people to remember exactly what they have eaten, or stick perfectly to a specific diet plan set out by the researchers.
Do you remember how many times you ate red meat over the last 12 months? Or even what you had every day for lunch last week?
Unfortunately both of these methods are notoriously inaccurate.
2. Most nutritional studies demonstrate correlation, not causation.
Say we were asked to figure out the cause of car crashes.
We start investigating by going out to examine the scene of every car crash in our area. We start to notice a common theme…
At every single car crash one thing is exactly the same. There is always an ambulance there! Solved! Ambulances cause car crashes!!!
Of course that isn’t the case, what our study has found is that the presence of one thing (car crash) correlates with the presence of another thing (ambulance).
They tend to be seen at the same time.
This is not proof that the ambulance is the CAUSE of the car crash. Further to this, in our minds it is pretty obvious that that isn’t the case.
However, try to imagine that you’ve been told your entire life that ambulances are bad and will inevitably cause car crashes. I bet this study now PROVES what you knew all along right?!
Now say you’ve been taught all your life that eating saturated fat is bad and will inevitably cause heart disease?
How open are you to reading a study or science report that suggests otherwise? How likely are you to read a study or scientific report that confirms what you think you know?
We are only human, and we like to be right.
But good science is about asking questions.
I remember hearing this from my veterinary anatomy lecturer my first year of university when I was desperate to be included on a research trip.
He happened to be one of the world’s leading humpback whale researchers and I love whales.
I mean 10 whale watching trips, teary-eyed, can’t speak, love them.
He said to me, “You can’t just love whales to be a researcher. You have to have questions that you want answered”… or something along those lines.
I remember emailing him with my application literally writing “I’m trying so hard not to say “I love whales!”. (P.S It was a glorious two weeks counting whales from the headland on North Stradbroke Island.)
The fundamental flaw in all science is human nature.
Scientists develop a hypothesis, an idea, and then set about proving it to be true.
Of course any scrap of evidence that may support our theory is like finding a golden nugget, whilst evidence against our belief is a bitter pill to swallow – that’s assuming we even bother to look at the opposing evidence.
3. The third problem, is that there are variables and confounding factors.
In one study (reference below), long story short, the aim was to show how poly-unsaturated fats such as in margarine were healthy and saturated fats from animals cause heart disease.
Basically, on the surface, it seemed to demonstrate just that. However there were two major issues.
The first was that if you looked at the data up until 3 or 4 years in, the “healthy fats” group seemed to do fine. But if you kept looking to 6 and 7 years in, the rates of cancer in the “healthy fat” group leapt up.
The second flaw was that although the “bad fats” group did indeed succumb to heart disease, there was a major difference in the adults put into these two groups – a variable that wasn’t accounted for (or initially mentioned in the reports).
The group that ate “healthy fats” – like margarine – were mostly light or non-smokers. The group that ate the “bad fats” were moderate to heavy smokers – in fact twice the number of heavy smokers! Smoking is a known risk factor for heart disease!
Without knowing those two things, i.e. reading a headline or news report, you could easily accept that indeed saturated fats inevitably cause heart disease, and polyunsaturated fats are perfectly healthy.
Nutritional studies can be very mis-leading, let alone the reports that are published second hand by the media.
4. Lastly, there is the healthy user bias.
This is where the groups of people studied may have other traits or habits that make them more likely to be healthy and give false support to the scientific theory being tested – hence the results are biased.
One example is red meat.
Some studies have indeed suggested that red meat can cause cancer.
When we compare two groups of people – red meat eaters and vegetarians – it is very likely that the vegetarian group are on average healthier since we already know that they have made a conscious decision not to eat red meat.
They are therefore also more likely to have made other conscious health decisions – regular exercise, yoga, avoiding smoking and drinking heavily, eating more vegetables, choosing organic, even shopping at health food stores more regularly.
Where as red meat eaters are more likely to consume fast food such as burgers (with chips and soft drink), eat less fresh produce in total, do less exercise, drink more alcohol and smoke.
So if the group that don’t eat red meat are less likely to get cancer – the real question is – what percentage of that risk is associated solely with not eating red meat?
Not to mention that there would also be differences between processed meats, muscle meats, organ meats, grass-fed and organic meats, and the amount of meat eaten in total. All very difficult to measure.
Unfortunately, the headline you’ll read remains the same: “Red meat causes cancer.”
To top it off, there is no ‘one diet fits all’.
A person who responds well to eating a vegetarian diet long term is likely better able to convert the necessary nutrients from plant sources to meet the body’s needs than someone who develops deficiencies and fails to thrive on the same diet.
Similarly for people able to tolerate dairy foods well they can be a significant contributor of nutrients whereas for others who react poorly to dairy proteins or sugars, the ensuing inflammation is not worth the trouble.
You don’t need to be a scientist to benefit from reading nutritional information, but now that you know more about this science, I simply hope that you’ll ask more questions.
—–
Do you struggle with choosing healthy snacks?
Yep, my clients did too.
That’s why I created my Ultimate Snack Guide – and you download it right now – for free!
Ultimate Snack Guide FREE Download
Kelly Moriarty is an online health & nutrition coach and founder of The Green Body Plan.
After 5 years as a veterinarian Kelly was unhealthy, stressed out and generally dissatisfied.
She now helps professional women make the same changes she did to beat burn out, lose weight, recover their energy and take back control of their health naturally.
REFERENCES
A Controlled Clinical Trial of a Diet High in Unsaturated Fat in Preventing Complications of Atherosclerosis
SEYMOUR DAYTON, MORTON LEE PEARCE, SAM HASHIMOTO, WILFRID J. DIXON, and UWAMIE TOMIYASUOriginally published1 Jul 1969https://doi.org/10.1161/01.CIR.40.1S2.II-1Circulation. 1969;40:II-1–II-63
kellyannmoriarty.com 