What’s going on…?
The widely-held view is that we are eating too much and exercising too little, which results in an excess calorie balance that leads to weight-gain. This is the energy-balance (calories in — calories out) model of weight gain, and it is so pervasive that many treat it as an axiom (something that doesn’t require proof), or at least as something that has already been proven beyond reasonable doubt.
However, there is little data to support this model at the population level. At a minimum, the model predicts that as the percentage of obese individuals in a population goes up over time, there should be at least some increase in the caloric intake of that population. However, a recent paper has shown that this is not the case — for the past 20 years, calorie intake by US adults has been relatively constant (even trending slightly down), while obesity rates have increased alarmingly.
The paper drew on data from the United States National Health and Nutrition Examination Survey (NHANES), which was conducted sporadically from the 1960s, but that became an annual survey from 1999 onward (thus we currently have ~20 years of detailed data collected under standardised conditions). Annually, around 5,000 representative members of the US population undergo medical, dental, physiological and laboratory tests as well as providing dietary, socioeconomic and health-related information. Trained health professionals travel across the US in purpose-designed and equipped mobile caravans to gather these data.
Dietary data was collected with a standardised, interviewer-administered 24-hour diet recall questionnaire from which caloric intake was calculated. Weight and body-mass index (BMI) were routinely collected. These data allow the graph below to be generated (see source).
The blue line represents total calorie intake over the past ~20 years (scaled on left vertical axis). The line is sitting consistently between 2,000 and 2,300 Calories over those decades, and if anything there was a slight decline over the period (the decline is statistically significant, even though it is small). The red line is the percentage of the population that were obese (BMI > 30, right vertical scale). Percentage obesity increased in a way that didn’t have any relationship to caloric intake. The green line is an estimate (using US data from the Food and Agriculture Organisation) of the availability of calories, and it is also constant over time, indicating that we have not been inundated with more food driving gluttony.
Taken together, these data indicate that the increasing rate of obesity over the last 2 decades has had next to nothing to do with food calories — either their consumption or their availability. Which means that the energy-balance equation that drives current thinking does not add up. This is important to realise, because it will never be possible to tackle obesity if the model being used to explain it is wrong.
What about the other side of the energy-balance model, calories-out? Has physical activity decreased enough over the last 20 years to explain the obesity curve?
No, this is not the case either, multiple studies point to an increase in physical activity over that time. This doesn’t mean physical activity levels are as good as authorities would like (not even close), but overall, physical activity hasn’t been going down in the population, it’s been going up.
If energy-balance is getting us nowhere, where else to look?
Perhaps the most obvious explanation is food quality. For example, the energy balance model does not distinguish between the biological effects of 100 Cal coming from sugar vs. 100 Cal from protein — they are both treated the same when it comes to weight-gain, thereby ignoring our biology. That alone should discredit the model.
With the steady industrialisation of our food supply — particularly easily-digestible shelf-stable refined starches and sugars, we may not be consuming more calories in total, but we might be digesting and absorbing more of those calories. If this is the case, the solution is not to eat fewer calories, but rather to switch the calorie source to favour less-refined whole foods.
As well, ultra-processed and refined foods tend to be carbohydrate-based and carbohydrates can trigger fat-storage. Lowering total carbohydrates, while maintaining total caloric input by increasing protein and fat, is an option that has worked for many.
A further worrying possibiity is that if a pregnant mother’s diet is obesogenic, then epigenetic cues may be delivered to the developing foetus that predisposes the offspring to obesity — obesity may be being passed on.
Finally, government dietary guidelines haven’t helped either — they have consistently favoured grains and carbohydrates over nutrient dense animal foods, and have been developed with biased scientific processes and conflicts of interest. Unfortunately, as they are influential they can have wide-ranging obesogenic implications in their own right. They are part of the problem.
In summary, the factors that are contributing to increased population obesity are less certain than commonly thought. Ironically, it may well be that progress is being held up by the erroneous belief that we already know the factors responsible (i.e. calories in — calories out). Meanwhile, there is little in mainstream thinking that is likely to stop obesity rates becoming an ever-increasing and compounding problem.
Primary source: Mozaffarian, D. Perspective: Obesity — an unexplained epidemic, The American Journal of Clinical Nutrition, 2022.
Physical activity in the US (1988–2018)