Thread by @JohnSpeakman4

Very happy to distribute here the latest paper from the @IAEANA DLW database published in @NatMetabolism which concerns the decline in energy expenditure over the last 3-4 decades in the USA and Europe. 1/
nature.com/articles/s4225…

Obesity is caused by an imbalance of expenditure and intake. There has been previous debate about the relative contributions falling expenditure due to declining physical activity, and increased food intake have made to the epidemic. 2/

Previous work showed that increasing levels of food supply was probably enough to explain rising levels of obesity nature.com/articles/s4301…
but an issue with these data is separating increased consumption from increased waste @KevinH_PhD 3/
pubmed.ncbi.nlm.nih.gov/19946359/

A study of direct measures of TEE using doubly-labelled water (DLW) measurements over time in the Netherlands suggested energy expenditure had not fallen between 1988 and 2006. 4/
pubmed.ncbi.nlm.nih.gov/18504442/

However, that study was based on a relatively small sample (<300) in a single Dutch city where obesity rates were relatively low. Moreover, direct measures of physical activity during work time do suggest a progressive decrease over time in the USA. 5/
pubmed.ncbi.nlm.nih.gov/21647427/

We now studied the Total energy expenditure (TEE) of 4799 individuals in Europe and the USA between the late 1980s and 2018 using the IAEA DLW database. We show there has been a significant decline in adjusted TEE over this interval of about 7.7% in males and 5.6% in females. 6/

We are currently expending about 220 kcal/d less for males and 122 kcal/d less for females than people of our age and body composition were in the late 1980s. These changes are sufficient to explain the obesity epidemic in the USA. @KevinH_PhD 7/
pubmed.ncbi.nlm.nih.gov/21872751/

For 1472 individuals we also had measures of basal energy expenditure (BEE) and could therefore calculate the activity expenditure (AEE) both adjusted for body composition. To our surprise adjusted Activity Energy expenditure has actually INCREASED over time. 8/

This means the decline in TEE is driven by a decline in BEE. This was confirmed directly in the data for males, but in females marginally failed to reach significance (p = 0.07). This lack of significance was completely due to inclusion of data from a single anomalous study. 9/

If data for that single study was removed the trend was also highly significant in females. The surprising conclusion is we spend less energy when resting now than individuals did 30-40 years ago! The magnitude of the effect is sufficient to explain the obesity epidemic. 10/

To validate this result we performed a meta analysis of data on BMR from the literature also in Europe and the USA (n = 9912 individuals in 163 studies) and showed that this trend of falling BMR is replicated (with a shallower gradient) back to the early 1900s. 11/

Interestingly this parallels the previously reported decline in body temperature over the same interval. 12/
pubmed.ncbi.nlm.nih.gov/31908267/

How is it possible that AEE has increased when worktime activity has fallen? The most obvious answer an increase in leisure time activity. Several studies have shown this directly. E.g. 13/
pubmed.ncbi.nlm.nih.gov/28749561/
pubmed.ncbi.nlm.nih.gov/27659584/

What caused the decline in the BEE? One possibility is a reduction in infectious diseases reducing the cost of the immune system. Amazonian subsistence agriculturalists have higher BEE than predicted linked to a high pathogen burden. @HermanPontzer 14/
pubmed.ncbi.nlm.nih.gov/27375044/

But if its important in the US/Europe over recent times is unclear. Note all the data included in the analysis were pre-covid 15/

Another potentially important factor is diet. Diets have changed enormously over the past 100 years. Notably there has been a large reduction in consumption of saturated fats and changes in the balance of n-3 and n-6 fatty acids.. 16/
pubmed.ncbi.nlm.nih.gov/21367944/

We varied the fatty acid composition of mouse diets and showed that their energy expenditure (adjusted for body composition) increased as their consumption of saturated fat increased. (@YaleCMed) 17/

This raises the interesting possibility that reducing saturated fat consumption may have driven reductions in BEE. RCT studies in humans however are required before that conclusion could be drawn. 18/

Some previous work in humans does not suggest the balance of n-3 to n-6 is important driver of resting metabolism 19/
…ritionandmetabolism.biomedcentral.com/articles/10.11…

There are many other dietary factors that could be important drivers of BEE. For example the carbohydrate-insulin model postulates that increased intake of high glycaemic load diets should drive a reduction in energy expenditure. 20/

The change in BEE observed here might then be consistent with the CIM if consumption of high GL diets has increased in the same interval. 21/ @davidludwigmd @garytaubes

Even if the CIM doesn’t explain all of the effect on TEE it may still have contributed something to the overall secular trends in obesity prevalence at some points during the epidemic. 22/

If declining BEE has a simple dietary basis then reversing it should be possible by dietary change. Impacts of diets on BEE have been studied, but it isn’t yet clear how long exposure to a diet would be necessary to generate a significant effect. 23/

Thanks to everyone who submitted their data into the DLW database to make this paper possible. 24/24
@HermanPontzer @YosukeYamada831 @NIBIOHN @StJohns_Blr @akurpad @HSagayama @UNIV_TSUKUBA_EN @PBRCNews @roodjc62 @maasuniversity @YaleCMed @iaeaorg @CorneliaLoechl @lucpubhealth