Low-fat, high-carb: new data from the PURE cohort study
The internets are abuzz with a new study from the journal Lancet surrounding an age-old controversy in the nutrition community: low-fat versus high-carbohydrate diets. This macronutrient debate has been going on since long before millennials were even born, with each side having distinct opinions regarding amounts and sources. Since we are self-proclaimed food nerds who jump on the chance at taking a deeper look at new science, here’s our overall take on this study.
What’s the 411 on the background and design of the study?
The research from Dehghan et al. is a large prospective cohort study that examined the associations between consumption of macronutrients (specifically focusing on carbohydrate, total fat, and different types of fat) with health outcomes like major cardiovascular events and total mortality. This study included 135,335 individuals ranging from 35-70 years of age in 18 different countries.
Intake of macronutrients was assessed by baseline food frequency questionnaires (FFQs), which we have talked about before. Participants were grouped into quintiles (five groups) based on macronutrient intake. Hazard ratios were calculated to determine the strength of associations between macronutrient intake and health outcomes.
Hazard ratios sounds scary. What are they?
Hazard ratios measure the relationship between two variables and are usually expressed as a single value, with a range of values placed next to the single value. A hazard ratio of one means that there is no relationship between the two variables. A hazard ratio of greater than one means that the variable is more likely to be associated to a specific outcome. A hazard ratio of less than one means that the variable is less likely to associated to a specific outcome. In the PURE study, hazard ratios assessed the link between macronutrient intake and health outcomes, including major cardiovascular events and total mortality.
What were the main findings of the study, and how might they be interpreted?
Let’s take a look at the macronutrient-specific findings:
- Carbohydrate-specific findings: The link between carbohydrate intake and total mortality compared individuals with the highest reported carbohydrate intake group (quintile 5, 77.2 percent of calories) to the lowest reported carbohydrate intake group (quintile 1, 46.4 percent of calories). The hazard ratio for this link was 1.28, meaning there was a 28 percent increase in total mortality for those who were in the largest carbohydrate intake group compared to those in the lowest intake group. Although this link was statistically significant, the hazard ratio is still pretty low.
- Fat-specific findings: The relationship between fat intake and total mortality compared individuals with the highest reported total fat intake group (quintile 5, 35.3 percent of calories) to the lowest reported total fat intake group (quintile 1, 10.6 percent of calories). The hazard ratio for this relationship was 0.77, meaning there was a 23 percent decrease in total mortality for those that were in the largest total fat intake group compared the lowest intake group.
Were there any general limitations to the study?
There are several limitations found in this new research:
- The most significant limitation is tied to its design and methodology. Because the PURE study is an observational study, it cannot establish a cause and effect.
- In addition, dietary intakes via FFQs were only measured at the beginning of the study, so it is quite possible that dietary habits changed over the follow-up period, which was about seven years.
- While the study did adjust for a variety of possible variables including age, sex, education, total energy intake, waist-to-hip ratio, and urban vs. rural location, there are still a number of possible confounding variables.
- These variables could include alcohol intake, medication use and socioeconomic status, all of which could greatly impact the results.
- The research did not drill down into specific foods.
- Although the analysis examined the associations between the health outcomes and different types of fats, including saturated fats, polyunsaturated fats and monounsaturated fats, the research did not examine the associations between different types of carbohydrates, such as enriched carbohydrates, fortified carbohydrates, whole grains and fiber.
- In addition, the study did not measure trans fat intake. Since there is strong scientific evidence linking trans fatty acid intake and the risk of developing cardiovascular disease, partially hydrogenated oils (a major source of trans fats) are required by FDA to be removed from the U.S. food supply by June 2018.
- The study’s commentary called out (seven times, if you are counting!) the need to revise current dietary recommendations.
- This is a bit off-putting for me, given that it’s just one study, and an observational one at that. However, when you take a look at actual dietary recommendations, you see that there’s already some significant alignment between the study’s results and current dietary recommendations, which leads us to our final question.
How do these new findings stack up to current dietary recommendations?
Let’s start with fats. The 2015 Dietary Guidelines Advisory Committee (DGAC) flirted with the idea of eliminating long-standing advice to limit total dietary fat intake. Official guidance in the 2015-2020 DGAs comes from the Acceptable Macronutrient Distribution Range originally established by the Institute of Medicine, which says adults 19 years and older should consume 20-35 percent of their total calories from fats and less than 10 percent from saturated fat. Both the 2015 DGAC and the 2015-2020 DGAs agree on the importance of fat in our diet, encouraging us to be more aware of the types of fats we eat (i.e. replacing saturated with mono- and polyunsaturated) and less about the total amount of fat we eat. That’s what the best available scientific evidence currently shows, and evidence put forward in the PURE study also agrees.
Carbohydrates, on the other hand, are where the debate gets more contentious. Or does it? Depending on activity level, the DGAs recommend eating between 45 percent and 65 percent of total calories from carbohydrates, with less active people being advised to aim for the lower end of the range (45-55 percent) and active individuals toward the higher end of the range (55-65 percent). Basically, if you don’t do a lot of rigorous exercise but keep busy (i.e. the majority of us), then aiming for a little more than half of your calories from carbohydrates is a reasonable goal. This DGA advice isn’t so different from what the authors highlight that “a certain amount of carbohydrate is necessary to meet short-term energy demands during physical activity and so moderate intakes (e.g., 50–55 percent of energy) are likely to be more appropriate than either very high or very low carbohydrate intakes.”
If these latest findings have you scratching your head and wondering if you should follow the current DGAs, look at it this way: The guidelines are not perfect. They are developed using the highest quality scientific evidence available at the time, which means that they evolve over time based on what the totality of science shows. One study such as this shouldn’t override current recommendations from the DGAs, but it could help inform the next set of guidelines in 2020.
This blog includes contributions from Kris Sollid, RD.