After one recent study hyperbolically linked an increase in carbohydrates to an early death, I thought it best to add some salt to the situation.
Recently, a major study came out apparently calling into question current advice about carbohydrate and fat recommendations. The Internet and media organisations went into meltdown with claims that too many carbohydrates could be sending you to an early grave. Scrolling past the headlines (and actually reading the paper) gave a healthy dose of boring because what makes for a healthy diet has changed little.
Called the PURE (Prospective Urban Rural Epidemiology) study, this looked at the link between fat and carbohydrate consumption and longevity in 18 countries across five continents. Running for seven years and involving over 135,000 people, this was a very large observational study so its conclusions are right to take notice of.
The key finding from the work that attracted the most media attention was that a high carbohydrate diet was linked with a higher risk of earlier mortality, whereas total fat and individual types of fat were related to a lower risk of earlier mortality.
Digging deeper into the study, the research team found that global diets consisted of 61% energy coming from carbohydrates and 24% energy from fats. And while those in the highest carbohydrate consumption group (a whopping 77% of energy) had a higher risk of earlier death, it wasn’t a cardiovascular disease they were dying from.
A paradigm shift? Not quite
Does this study turn on its head “everything we knew about nutrition”? Not quite. And here’s why.
Before the PURE study, there were many studies showing the opposite link between carbohydrates and longevity. So, when a conflicting study comes along, this grabs the media spotlight for the day.
Here is just one example – a major systematic review and meta-analysis from 2013 involving 17 individual studies and over 242,000 people showing a higher risk of earlier mortality as carbohydrate intake decreased. And this is the problem at times with observational research in that two studies can give polar opposite results, so the findings of the PURE study should be seen through this filter.
I’m not going to pick apart the PURE study for its flaws. Such issues are consistent across all observational studies no matter if the conclusions support consensus views or not. What is of value to look at is the positive messages the study gave and how when you look at the full research field, it takes you back to some pretty sensible advice.
The PURE study was not so much an endorsement of low-carbohydrate diets, only a question mark on very high carbohydrate diets at a population level. Even the research team made this conclusion and, when quoted in the press release for the study, said:
“The best diets will include a balance of carbohydrates and fats – approximately 50-55 per cent carbohydrates and around 35 per cent total fat, including both saturated and unsaturated fats.”
A key shift in dietary habits over the last few decades has been towards an increase in consumption of highly refined carbohydrates and sugar which is at odds with what dietary guidelines actually recommend. It is possible that the PURE study was detecting some of that change. Without knowing the overall diet quality of the participants, it is difficult to make firm conclusions.
In Australia, we sit at 43.5% energy from carbohydrate. We’re hardly killing ourselves slowly with a high carbohydrate diet at that level. But unfortunately, a lot of those carbohydrates are highly refined and we have too much added sugar in our diet. To add to this, only 6% of us eat enough vegetables each day.
The PURE study is interesting though as it shows how complex it is determining the role of nutrition over many decades in improving health, and how simple measures of fat or carbohydrate quantity lose sight of the foods that are part of this. We eat foods, not nutrients, and as nutrition science advances, research is now turning to dietary patterns as being key predictors of health, rather than fixating on percentage targets for macronutrients.
Nutrition is a messy science
Nutrition science is messy. It is a young science and we have to deal with making conclusions from often imprecise research using well below par tools to measure dietary intake.
Here’s an example following of individual research studies that have been linked to either increasing or decreasing the risk of cancer. Green dots are good and show that the food lowers the risk; red dots are bad. Being selective in what dots you looked at and the studies you cite, red meat can lower your risk of cancer, but if you have it with onions, up goes your cancer risk!
Nutrition science is messy because it is impractical to run long-term, high-compliance randomised controlled trials (considered the “gold standard”) to answer big health questions. Good luck trying to get thousands of people to adhere to a particular type of diet for years on end – most people struggle for a week or two following a prescribed diet. So, we turn to the next best source of evidence and that’s observational studies. The PURE study is an example of an observational study.
Conflict of interest is also a problem. And I do not just mean “big food” funding research. Researchers tied to a particular narrative with their research program (and maybe a large public and media following and the odd best-selling book or two) may find it difficult to do an about face when conflicting evidence presents itself. As humans, we are all susceptible to this. The scientific method helps reduce it, but it cannot eliminate it entirely.
Selective use of evidence (called cherry picking) can also be a problem. It has been said that you can build a case for any particular food or nutrient being good or bad for you by quoting a study or two that agrees with you. Just look at what happened when I did just that intentionally in making the case of why broccoli is toxic.
But all is not lost. There are recurring themes that we can learn from in nutrition research and these are themes I have seen repeated year after year.
There are many paths to good health by eating a range of diets, with the biggest predictor of this being adhering to a dietary pattern that connects with a person, rather than following prescriptive advice. Low fat, high fat and somewhere in between like the Mediterranean diet can only be healthy if food quality of mostly plant-based, minimally-processed foods is put first.
Dietary patterns for the win
So what are the key nutrition themes seen in healthy dietary patterns around the world?
A major scientific review has taken things back to basics to reinforce where the best health gains are to be found with diet. No small undertaking, the review looked at the diet and chronic disease links from 304 meta-analyses and systematic reviews published in the last 63 years. Type 2 diabetes, overweight and obesity, cancer and cardiovascular disease together accounted for most of the chronic disease links found. And here’s the summary:
- Plant-based foods are more protective against the risk of chronic disease compared to animal-based foods;
- Wholegrain-based foods are slightly more protective than fruits and vegetables;
- Highly refined grains are deleterious to health;
- Dairy foods are neutral;
- Red and processed meat increases risk;
- Tea is the most protective beverage while soft-drinks the least.
Eat more plant-based foods than animal foods, choose whole grains over refined grains, limit red and processed meat and choose other beverages in preference to soft drink. And watch how much added sugar is creeping into your diet. Such recommendations may not get media attention like the PURE study, or help sell books in numbers like the latest popular diet, but they are the cornerstone of long-term good health.