Slow eaters were 42 per cent less likely to be obese than fast eaters, according to a 2018 study of 59,717 Japanese adults who had been diagnosed with type 2 diabetes.
The paper, by Hurst and Fukuda, appeared in BMJ Open as a secondary analysis of longitudinal health check-up data. It is observational. Nobody was assigned to chew for longer and then followed up. The record on PubMed lists it as BMJ Open 2018, volume 8, article e019589, PMID 29440054.
What the researchers had to work with
The authors built patient-level panel data covering 2008 to 2013, six years in total, from commercially available insurance claims data and health check-up data. The subjects were Japanese men and women enrolled in health insurance societies who had been diagnosed with type 2 diabetes during the study period.
Body mass index was measured, and obesity was defined as a BMI of 25 or more, the threshold the paper uses.
Eating speed was not timed. It came from the subjects' own answers to questions asked during the health check-ups, with three options, fast, normal and slow. That self-reported scale is the main exposure the analysis rests on.
Other habits were recorded at the same time, including eating dinner within two hours of sleeping, after-dinner snacking, skipping breakfast, alcohol consumption frequency, sleep adequacy and tobacco consumption. The published abstract reports no separate results for any of them, so nothing can be said here about what those habits did.
Two statistical approaches were used. A generalised estimating equation model examined the effects of these habits on obesity. Fixed-effects models then assessed the effects on BMI and waist circumference, tracking the same individuals across the six years.
The numbers the paper reports
In the first model, the odds ratio for obesity was 0.58 for slow eaters and 0.71 for normal-speed eaters, both measured against fast eaters, at P less than 0.001. The 42 per cent in the headline is one minus 0.58.
The abstract puts it plainly: "The generalised estimating equation model showed that eating slower inhibited the development of obesity." It adds that the two odds ratios "indicated that these groups were less likely to be obese than fast eaters".
The fixed-effects results sit behind the second half of the headline. Relative to fast eaters, the coefficients of the BMI model were minus 0.11 for slow eaters and minus 0.07 for normal-speed eaters, again at P less than 0.001.
Waist circumference also fell. The abstract states that the fixed-effects models "showed that eating slower reduced BMI and waist circumference", but it gives no coefficient for waist, so there is no figure to quote.
The two sets of numbers describe different things. The odds ratios describe how likely each group was to sit above a BMI of 25. The BMI coefficients, at minus 0.11 and minus 0.07, are movements in BMI itself.
What the study cannot settle
This was not a randomised trial. The researchers did not assign anyone to an eating-speed group, so the design cannot rule out the possibility that something else about slower eaters, or about the circumstances in which people slow down, is doing part of the work. The authors open the paper by saying that few studies have examined the causal relationships between lifestyle habits and obesity, the gap they set out to address with panel data.
The population matters too. Everyone in the sample was a Japanese adult enrolled in a health insurance society and diagnosed with type 2 diabetes during the study period. The results describe that group, over those six years, using that three-point self-reported scale.
The authors' framing in the discussion is measured. They write that "Changes in eating speed can affect changes in obesity, BMI and waist circumference" and that "Interventions aimed at reducing eating speed may be effective in preventing obesity and lowering the associated health risks". Both sentences are discussion rather than a trial result. An intervention of that kind is raised as a possibility, not something this paper tested.
One point about the video above. It is titled "Why People Who Eat Slowly Age Differently", and ageing is not something Hurst and Fukuda measured. The outcomes in this paper are obesity, BMI and waist circumference.
What the data support is narrower than the habit in that title. Slower reported eating went with lower odds of obesity in a large observational data set, and changes in reported speed tracked with changes in BMI and waist circumference in the same people over six years.
For other research on meal timing and dietary patterns, see what three trials on early eating windows show and how Okinawan vegetable intake shifted between 1988 and 1998.
Photo: People eating at a spice market (illustrative stand-in). 玄史生, CC0, via Wikimedia Commons.




