Primary Care Corner with Geoffrey Modest MD: physical activity energy expenditure, a new paradigm

by  Dr Geoffrey Modest

​​The Scientific American magazine just had an article which made the point that in cross-cultural studies, human beings expend the same number of calories whether they are doing intense exercise or sitting around (see ). These anthropologists studied several human populations, though this report was largely on the Hadza people who lived in the dry savanna wilderness of northern Tanzania. In particular they found that Hazda men who spent days hunting and tracking game, ate and burned 2600 cal a day. Hadza women, who also did a lot of physical work, ate and burned 1900 cal a day. This is pretty much the same as adults in the US or Europe, and was independent of body size, fat percentage, or age. Similarly, research has shown that rural Nigerian women and African-American women in Chicago have similar energy expenditure, despite large differences in activity level. And, a large collaborative effort on non-human primates found that captive primates living in labs or zoos expended the same number of calories as those in the wild.

They posit that there might be differences in calories spent on different activities, and those who are sedentary may spend more energy on other things. For example exercise reduces inflammation, so sedentary individuals may have to spend more energy reducing inflammation. In addition women who exercise a lot may have decreased estrogen levels and fewer ovulatory cycles (ie, less energy spent there), perhaps to compensate for their increased exercise-related energy expenditure. Other studies have found that those who do long-term exercise have reduced basal metabolic rates [and, my addition, have slower heart rates, which may itself be associated with longer life], expending less energy in this manner

A more scientific study published by the author of the above article looked at total energy expenditure over a range of physical activity, finding that there was a positive relationship between total energy expenditure and physical activity but only at the lower ranges of physical activity. Energy expenditure plateaued at the upper ranges of physical activity (see Pontzer H. Current Biology 2016: 26, 410.)


— 332 adults age 25 to 45, 55% female, from 5 populations across Africa North America (Ghana, South Africa, Seychelles, Jamaica, and the US)

— total energy expenditure was measured using doubly labeled water, considered the most accurate measurement (where each subject would ingest radioactively labeled water and assess urinary excretion of the isotopes)

— physical activity was measured by wearable accelerometers, measured in counts/minute


–there was a small increase in the total energy expenditure from a baseline of about 600 kcal/d with 0 counts/min (cpm) on the accelerometer to about 800 kcal/min as this increased to 230 cpm, with no significant further increase up to >700 cpm


— these articles are consistent with the repeated finding that exercise by itself does not lead to weight loss, though is an important factor in maintaining weight loss in those on a diet; these data undercut an Additive model (where total energy expenditure is a simple linear function of physical activity and that this determines variations in total energy expenditure) to more of a Constrained total energy expenditure model (where the body adapts to changes in physical activity to maintain total energy expenditure within a narrow range). And perhaps there is an evolutionary aspect as to why those who exercise have to conserve energy in other ways, such as lowering basal metabolic rate, decreasing population (and the decrease in estrogen/ovulatory cycles in women also decreases energy expenditure on growth, perhaps with the added issue that those who exercise lots to get food may need to have fewer mouths to feed to survive),  etc.

— But, of course, even if exercise does not lead to weight loss, it is important to maintain the perspective that physical activity has a multitude of important healthy benefits, both mentally and physically, and I believe we as clinicians should be encouraging it regularly with our patients. Increasingly, we are learning of the benefits of exercise even where physicians initially felt it was potentially harmful, including in patients with severe heart failure for example. (and in the old days, clinicians prescribed prolonged strict bedrest post-MI, or back pain…)

— A recent meta-analysis for example found that lower levels of exercise was associated with excess mortality (see ), and, as opposed to other studies, this one looked at total physical activity and not just leisure-time activity:

— a review of articles from 1981 to 2016, prospective international cohort studies examining association between total physical activity and least one of the 5 diseases studied: 35 for breast cancer, 19 for colon cancer, 55 for diabetes, 43 for ischemic heart disease, and 26 for ischemic stroke, yielding 174 identified articles.

— Higher levels of physical activity are associated with low risk for all of these outcomes, at 3000-4000 metabolic equivalent minutes per week, or METs/wk.

— overall, comparing those with < 600 METs/wk to those with > 8000 METs/wk was associated with:

— 14% decreased risk of breast cancer

— 12% decreased risk of colon cancer

— 28% decreased risk for diabetes

— 25% decrease risk for ischemic heart disease

— 26% decreased risk for stroke

–but for all of these, the major risk reductions occurred at lower levels of activity, with diminishing returns after 3000-4000 MET minutes/wk.

–and, another recent study (see Ekelund U. Lancet 2016; 388: 1302–10) did a meta-analysis of 13 studies with over 1 million individuals followed for 2 to 18 years, assessing sitting time, TV time and physical activity. They found that, as compared to those in the most-active quartile (35.5 MET-h/week):

— those performing < 16 MET-h/week had a 12% higher mortality rate

— those with the lowest quartile of physical activity (<2.5 MET-h/week and sitting > 8 h/ day) had mortality rates 59% higher

— And, subgroup analysis revealed that it was the exercise level determined risk, independent of the sitting time (comparing <4 h/d and up to >8 h/d of sitting time).

— However, watching TV for >5 hours per day was associated with increased mortality regardless of physical activity, with noticeable increases in the group watching 3-5 hours TV/d.

–of course, these are observational studies, though large and consistent in their conclusions. but those doing less exercise may be very different from those doing more (poorer health, different social determinants of health, etc)

So, what does all of this mean? A few points:

–exercise is really important in maintaining a healthy life (as above, but also as shown in smaller randomized controlled trials: see here ​ for an array of blogs)

–the first studies suggest that there is a conservation of total energy expenditure, largely independent of how much exercise is done. This is likely evolutionarily determined. And perhaps the energy expenditure on exercise is in part diverting energy away from what would be unnecessary bodily functions needed for the sedentary (eg decreasing inflammation, which exercise itself helps). But this finding of conservation of energy expenditure may help explain in part why increasing exercise is not associated by itself with weight loss.

–and it is interesting in the last study how TV watching seems to be a particularly bad sedentary activity, much worse than just sitting time (which would include such things as sitting at work, which might involve more thought, fidgeting, other activities than the more typically passive “vegging out” in front of the tube).