The study, based on dietary recall data, got national attention, but experts believe it’s a highly confounded connection.
Time-restricted eating, a type of intermittent fasting, appeared to significantly up the risk of death from cardiovascular causes in an observational analysis, and the finding, which generated a flurry of headlines, has led to no shortage of pushback from cardiologists, researchers, and others who have looked into the popular dietary trend.
Some even questioned whether the research, presented as a meeting abstract, should have been highlighted as it was in a press release from the American Heart Association (AHA).
In an analysis from the National Health and Nutrition Examination Survey (NHANES), eating only during an 8-hour or smaller window throughout the day was associated with a 91% higher risk of death from cardiovascular causes, with even greater risks seen in those with cardiovascular disease or cancer, according to research presented this week at the AHA’s Epidemiology and Prevention|Lifestyle and Cardiometabolic Health Scientific Sessions 2024.
“[The] hazard ratios are not small,” senior investigator Victor W. Zhong, PhD (Shanghai Jiao Tong University School of Medicine, China), told TCTMD in an email. “We had expected that long-term adoption of 8-hour time-restricted eating would be associated with lower risk of cardiovascular death and even all-cause death. We were surprised to find that compared to people whose eating duration was 12-16 hours—the mean duration of the study sample—those who restricted their eating time within 8 hours per day had higher risk of cardiovascular death and did not live longer.”
The ensuing media coverage, with stories in USA Today, The Washington Post, NBC News, and others, led cardiologists to say they were fielding calls from concerned patients. Some argued that the study was too flawed to draw real conclusions, adding that the media fell hard for observational data—published in abstract form, no less—and unnecessarily pushed the findings into a national spotlight.
“The fact that an abstract at a meeting got this much press is just absurd,” Pam Taub, MD (University of California, San Diego), who has studied time-restricted eating patterns, told TCTMD.
In fact, Taub, along with others, took issue with how the findings were communicated to the media and public. She shared a letter signed by 34 other researchers who have studied time-restricted eating that was sent to AHA President Joseph Wu, MD, PhD (Stanford Medicine, CA), stating that “unpublished and unvetted data presented in posters are meant to be discussed at scientific conferences, not used for press releases to alarm the public.”
I always say let the data speak for themselves. Victor Zhong
The study, along with the media coverage, will make it much harder to recruit patients and secure funding for future studies, she argued. “I’m getting emails from people internationally asking, ‘What's going on?’” said Taub. “Everybody believes the AHA, so when the AHA says it’s linked to a 91% higher risk of cardiovascular death in their headline, that’s all people look at. They don’t read the details.”
Donald Lloyd-Jones, MD (Northwestern University Feinberg School of Medicine, Chicago, IL), past president of the AHA, is also among the many surprised by how things played out.
“We come to meetings like this to look at abstracts, which are preliminary data, in order to debate the quality of the science and how meaningful we think the outcomes might be,” he told TCTMD. “This is a great example of that—this is not ready for prime-time consumption. Consumers and the general public should not be alarmed by these findings.”
Taub echoed that sentiment, noting that abstracts get a lot of leeway at meetings because the purpose is to generate discussion and to help researchers prepare studies for publication. Unfortunately, the headline from the press release was all the media picked up on, she said. “That’s not right—they could have worded it a better way,” said Taub.
Both Taub and Lloyd-Jones also argued that the data are likely highly confounded. “It wasn't their eating time that was causing their higher risk [of cardiovascular mortality],” he said. “It was what they brought to the table.”
Lloyd-Jones further suspects the study identified a group of people who weren’t necessarily practicing intermittent fasting, but who followed a restricted eating pattern because their lives demanded it. They may have been working two jobs or doing shift work. “There are many other explanations for why this might be the case and that could increase the risk for cardiovascular disease without implicating timing of their eating,” said Lloyd-Jones.
Intermittent Fasting
Time-restricted eating or intermittent fasting, in which people limit eating to a 4- to 12-hour window each day and fast for the remainder, have become increasingly popular weight loss methods. Short-term intervention studies—those ranging from a month to a year—have shown that following this type of eating pattern can promote weight loss and improve cardiometabolic health through metrics such as healthier glucose levels and lower blood pressures. There are no outcomes studies or data tracking longer-term follow-up, though.
Given the lack of long-term data, the researchers turned to NHANES (2003-2018), linking survey data about eating patterns with the National Death Index. Eating duration—the time between the first and last meal—was calculated based on two recall days in 20,078 adults (mean age 48 years; 50.0% men) and categorized as less than 8 hours, 8 to less than 10 hours, 10 to less than 12 hours, 12 to 16 hours, and more than 16 hours.
The fact that an abstract at a meeting got this much press is just absurd. Pam Taub
There was a significant association between an eating duration of less than 8 hours with increased cardiovascular mortality during a median follow-up of 8 years (HR 1.91; 95% CI 1.20-3.03), but no association with all-cause mortality or cancer mortality. Among patients with cancer, the risk of cancer mortality was lower among those who ate during a much larger window of more than 16 hours per day (HR 0.47; 95% CI 0.23-0.95).
With an observational study, Zhong said they are unable to uncover the underlying mechanism driving the heightened risk seen with restricted eating.
“However, we did observe that people who restricted eating to a period less than 8 hours per day had less lean muscle mass compared with those with a typical eating duration of 12-16 hours,” he said. “Loss of lean body mass has been linked to higher risk of cardiovascular mortality. Other mechanisms linking 8-hour time-restricted eating with cardiovascular morality require further investigations. This will be an exciting research question, but we really do not know much right now.”
Results Confounded by Big Group Differences
To TCTMD, Taub noted that her team’s initial pilot study of limiting food take to 10 hours per day in patients with metabolic syndrome reduced body weight, blood pressure, LDL cholesterol, and hemoglobin A1c levels over 12 weeks. They have just completed a National Institutes of Health (NIH)-funded study of intermittent fasting in 120 patients that will be published soon. Other studies, as well as meta-analyses, show a clear benefit to time-restricted eating, she said.
Acknowledging that present-day research is mostly short-term, typically around 3 months, Taub emphasized that all the important biomarkers move in the right direction.
“I haven't found a study, any study, where people improve blood pressure, where the LDL gets better, and that's associated with increased mortality,” she said. “There is a big gap in the field, and we do need long-term data, and we do need data on hard cardiovascular outcomes. Right now, there is none, but this study didn’t offer any insights into that data gap.”
Citing numerous limitations with the study, Lloyd-Jones pointed out that the assessment of eating patterns in NHANES relies on dietary recall in the past 24 hours. While this is a validated tool for collecting this information, it’s imperfect because researchers are relying on memory and not feeding people as would happen in a randomized, controlled trial. More critically, the study assumes the 2-day recall represents their usual eating pattern over the entirety of follow-up.
It wasn't their eating time that was causing their higher risk [of cardiovascular mortality]. It was what they brought to the table. Donald Lloyd-Jones
“It’s a sample and there’s a lot of imprecision in that assumption,” he said. “That creates some uncertainty in the results, which is pretty substantial.”
Additionally, Taub highlighted the very small sample size, noting that just 2% of the entire study population—414 patients—reported eating during an 8-hour or less window, of whom 31 died from cardiovascular causes.
Despite the researchers’ conclusions, Lloyd-Jones and Taub both said there are clues to suggest the data are highly confounded by significant baseline differences in those who ate during an 8-hour window and those who were eating 12-16 hours per day.
“One of the tip-offs here is that there were three to four times more Black people in the short-duration eating group than in the longer-duration eating group,” said Lloyd-Jones. “That tells you that they're going to be at higher baseline risk because we know that Black people in this country are at higher risk than white people.”
Additionally, those who restricted their eating to less than 8 hours were more likely to be smokers and have a higher body mass index. While blood pressure and lipids weren’t documented in the abstract, Lloyd-Jones suspects these risk factors, based on the other variables, would also be higher in those who fasted 16 hours per day.
‘Let the Data Speak for Themselves’
Zhong acknowledged that the study findings need replication and warned that no specific dietary advice around restricted eating/intermittent fasting should be given based on their study alone. Nonetheless, he added, “I always say: let the data speak for themselves.”
Incorporating intermittent fasting for 3 to 6 months could lead to weight loss and improved cardiometabolic health “based on what we have known so far,” he added. But he stressed that people should “be extremely cautious” when considering practicing intermittent fasting, particularly 8-hour time-restricted eating, for a longer period of time.
Lloyd-Jones said these results won’t change his practice, adding that he is awaiting more data. If people are under the care of a physician and having weight-loss success with intermittent fasting and their blood pressure, lipids, and other measures look good, there’s no need to change what they’re doing based on this study, he said. Taub agreed, noting she has many patients in clinical practice following this eating pattern with good results.
