Both High- and Low-Carb Diets Increase Long-term Risk of Mortality: ARIC Analysis

More research points to moderate carbohydrate consumption as well as plant-based proteins and fats as the best dietary approach.

Both High- and Low-Carb Diets Increase Long-term Risk of Mortality: ARIC Analysis

Study findings have wavered in their verdicts on high- and low-carbohydrate diets, but new research suggests that moderate carbohydrate consumption may be better than either extreme. The study is one of several in recent weeks to take aim at the once-popular idea that low carb should supplant low fat as the most heart-healthy diet.

In a new cohort study following 15,000 US adults for a median of 25 years, high- and low-carbohydrate diets each were tied to an increased risk of mortality. By contrast, diets based in plant-based proteins and fats had a lower risk of death.

Though low-carbohydrate diets are associated with short-term weight loss, the long-term health impacts of these diets are not as well known, write Sara Seidelmann, MD, PhD (Brigham and Women’s Hospital, Boston, MA), and colleagues.

“Although many randomized controlled trials of low-carbohydrate diets suggest beneficial short-term weight loss and improvements in cardiometabolic risk, morality risk has typically not been investigated in light of the practical challenges posed by studies involving very long durations of follow-up,” they say.

The study was published online in Lancet Public Health earlier this month.


Researchers selected 15,428 adults from the Atherosclerosis Risk of Communities (ARIC) study who were enrolled between 1987 and 1989. All included participants completed a questionnaire documenting food intake that researchers could then use to calculate the percent of carbohydrates, proteins, and fats that a subject regularly ate.

The study followed participants until 2013, looking carbohydrate intake and all-cause mortality. A moderate carbohydrate intake of 50-55% of daily energy intake was used as the reference point. There were 6,283 deaths in ARIC during follow-up.

After for adjusting for potentially confounding factors such age, race, gender, diabetes, cigarette smoking, and physical activity, the study found that ARIC individuals who fell into the lowest range of carbohydrate intake (< 30% of total energy) were 37% more likely to suffer from all-cause mortality than the group of moderate carbohydrate intake (HR 1.37; 95% CI 1.13-1.63). Individuals with the highest carbohydrate intake (> 70%) were 16% more likely to die than the reference group (HR 1.16; 95% CI 1.02-1.33).

The graph of these hazard ratios formed a U-shape, with moderate carbohydrate consumption associated with the lowest risk.

Seidelmann et al later combined the ARIC data with seven multinational prospective studies for a meta-analysis of 432,179 patients and a total of 40,181 deaths. The pooled analysis had similar findings. Low-carbohydrate diets were associated with a 20% greater risk of death compared to diets with moderate carbohydrate consumption (pooled HR 1.20; 1.09-1.32), and high-carb diets increased risk of death by 23% (pooled HR 1.23; 95% CI 1.11-1.36).

Eat Carbs in Moderation

This U-shape, where either end of the spectrum resulted in higher risk, is reasonable, says Andrew Mente, PhD (Population Health Research Institute, Ontario, Canada). “It is really not all that surprising to say that there is sweet spot–no pun intended–in the middle for carbohydrates,” he said, “because that would suggest that a moderate diet is optimal. Which does make sense, because as you go to the extremes, then you're eating a more extreme diet and that is much less proven in the literature.”

Mente and Salim Yusuf, MD (McMaster University, Ontario, Canada), co-authored an editorial that accompanied the study; they are also the leads on a new dietary analysis from the massive PURE study that came out earlier this week at the European Society of Cardiology Congress 2018.

Previous hypotheses pose that either extreme carbohydrate consumption pattern could trigger negative biological reactions, the authors of the paper say. “Long-term effects of a low-carbohydrate diet with typically low plant and increased animal protein and fat consumption have been hypothesized to stimulate inflammatory pathways, biological ageing, and oxidative stress,” they write. “On the other end of spectrum, high-carbohydrate diets, which are common in Asian and less economically advantaged nations, tend to be high in refined carbohydrates, such as white rice; these types of diets might reflect poor food quality and confer a chronically high glycemic load that can lead to negative metabolic consequences.”

Individuals in ARIC who substituted carbs for higher amounts of animal-based proteins and fat also had a higher risk of mortality (HR 1.18; 95% CI 1.08-1.29); conversely, participants who substituted carbohydrates with plant-based fat and protein sources had an 18% lower risk of death (HR 0.82; 95% CI 0.78-0.87). This finding is consistent with previous research, the authors say.

The next step, Mente says, is to go beyond macronutrients like carbohydrates, proteins, and fats and to look at food groups when advising healthy dietary habits. “The problem with looking at nutrients is that nutrients are complicated. Looking at carbohydrates: there are good carbohydrates and there are not-so-good carbohydrates,” he said. “For instance, a bagel is not the same as eating fruits, vegetables, legumes. If you are looking at animal foods, ice cream is not the same as yogurt. Salami is not the same as steak.”

Since the authors did not distinguish between sources of carbohydrates or sources of animal proteins and fats, it is difficult to use these findings to inform what an optimal diet for health should be, Mente said, but ultimately moderation is the best bet.

“It's not a good idea to make recommendations for extreme diets until we better understand their long-term effects,” he advised.

Lucy Hicks is the 2018 recipient of the Jason Kahn Fellowship in Medical Journalism. She is currently a master’s student…

Read Full Bio
  • Mente reports not relevant conflicts of interest.