I first posted “The Inefficiency of Humans” in January 2017. It’s about the reasons that our energy metabolism provides only about 20-25% of the energy we consume as food for muscular work. I referred to this as our “inefficiency”, although it isn’t really that. The explanation and conclusions remain largely unchanged, but since then I’ve learned more about how energy metabolism is measured, so I’ve rewritten the post as this 2.0 version. Much of the original is unchanged.
The human body converts the energy contained in food to mechanical output with an efficiency that may surprise you — it’s that low. Exercise machines at the gym, the ellipticals, stationary bikes, and rowing machines, usually provide readouts that show how much work the user is doing, and how many Calories she is burning. Checking the numbers now and then helps pass the time, and may feel rewarding at the end. But if you look at the work done, the power that went into driving the machine, it turns out to be only a fraction, between 18 and 26%, of the total caloric energy burned during the session. (The way to do this comparison is described below.) What’s going on?
Measuring the workout
Most people going to the gym have at least a passing interest in how many Calories they have burned. Knowing that may provide at least a faint hope that weight is being lost, even though for most of us, the number of Calories consumed during gym workouts is pretty small compared to total food consumption. (As in other posts on this site, I use the term “Calories” to measure energy. The Calorie is the same as the scientist’s kilocalories – the kcal – but is more familiar to most people. Food packages provide information in Calories.) The Calories burned by the average adult working out fairly strenuously for three hours a week is about 10% of the total energy content of their food. Regular exercise will usually improve health, and may help psychologically with maintaining a healthy life, but real weight loss requires dietary control as well.
Serious athletes such as competitive rowers or cyclists look at a different measurement of their workout’s intensity: they are interested in their “work rate”, which measures how much power they are delivering to the exercise machine. They know that the ability to deliver a high level of power for a longer time is the key to improved performance in almost any sport, but particularly in events like bicycle racing or speed skating. Actually, there’s an even more informative measure for high level participants: the power delivered per kilogram of body weight.
Work rate, or power, is measured in watts, and most workout machines provide that number. A power output of 150 watts for one hour equals 150 watt-hours, or 0.150 kilowatt hours (kWh). This is the measure of how much energy was delivered to the machine. (The kWh is also how the electrical utility measures the energy consumption at your house.) Calories are also a measure of energy, and the units are interconvertible: 1 kWh is the same as 860 Calories.
A workout machine like a stationary bicycle will usually provide both the power being delivered to it (in watts), and the total amount of energy expended by you (in Calories). It may also give you the average power over the time of the training session. But if you compare the power output with the calories burned, there will be a big difference. The number of Calories burned will probably be at least four times higher than the power delivered to the machine. What’s going on?
Looking at the numbers
Let’s say, as a reasonably fit adult, one who finds time to work out two or three times a week but has a life, you are comfortable pedaling an exercise bike at a rate of 150 watts for 30 minutes. 150 watts for half an hour is 75 watt-hours, or 0.075 kWh. 150 watts is enough power to keep a few light bulbs going, but your hair dryer may consume energy at ten times that rate. Professional athletes can do far more.
The Belgian bike racer Eddy Merckx is considered by many to be the greatest cyclist who ever lived. He used to annihilate his opponents; his nicknames included “The cannibal” and “One man forest fire”. One day in 1975 he was put on a stationary bicycle and told to go hard. That day, Merckx produced 455 watts of power for one hour, about twice what a fit young amateur athlete can do. Bicycle racers have a phenomanal power to weight ratio. Riders in the peloton can produce 800 watts for 15 seconds, and 600 for one minute. The strongest cyclists can put out 1200 watts for 15 seconds.
Back to your session at the gym. OK, you produced 150 watts for 30 minutes. Now you look at another part of the display and see that it reads 375 Calories burned. Feels good. But wait a minute. From the conversion factor (1 kWh equals 860 Calories) 150 watts for 30 minutes (0.075 kWh) is equal to only 64.5 Calories (.075 x 860). To repeat, what’s going on?
The difference between work done, and the total Calories burned, reflects the fact that our bodies don’t convert 100% of the energy released by the consumption of carbohydrate, fat and protein into mechanical work. No machine is 100% efficient, and we are no exception. Cars with internal combustion engines are about 20% efficient at best. The efficiency of the human body, the rate we can do work compared to the calories we burn, is generally between 18 and 26%. (There are many web sites dedicated to explaining the difference. I don’t recommend them. They often get it wrong, and almost always make it even more confusing.)
Human energy metabolism
Energy metabolism is complex, and most people would rather work on their income tax than have someone try to explain it to them. But in simple terms, it helps to think of energy metabolism, the conversion of the energy contained in sugar, fat, and protein into energy usable by the body, as a stream flowing downhill. There are little turbines along the way which harness the energy of the water flowing through them and use it to synthesize adenosine triphosphate, ATP. ATP is the chemical form of energy that the body can use to power its muscles or think its thoughts, among other things (the brain has a high level of energy metabolism; in a body at rest, about 20% of the total).
The products of energy metabolism are carbon dioxide (yes, the greenhouse gas), water, heat, and ATP. The amount of ATP formed is a function of the design of the energy metabolism pathways. As found, in organisms ranging from humans to fruit flies to bacteria, it represents only about 55 – 60% of the energy made available by the path from foodstuff to CO2. If it was designed to extract more energy as ATP, it would be more efficient, but it would run slower, limiting the rate of ATP production. If it was designed to deliver less ATP, less energy captured, the pathway would be faster, but the efficiency would be less. What evolution has provided is a Goldilocks solution: just the right amount of ATP produced, about 57% of the energy of food oxidation captured, so that our normal energy needs are met, without having to pour a lot more fuel into the pathway. (If we were, say, only 10% efficient, we’d have to eat like pandas to provide the energy for our daily lives. Either that, or move like sloths.)
The efficiency of muscular work, driven by ATP, is itself not 100%. Muscle fibers sliding over each other experience friction, like any moving machinery. But the main reason for the “inefficiency” of the human body (not really inefficiency at all) is that many other processes have to keep going to sustain life, and they use energy. Even at rest, when the skeletal muscles are using almost no energy, the human body consumes about 80 watts of energy (65-70 Calories per hour). This keeps the heart going, the blood flowing, and the lungs inhaling and exhaling. A lot of energy is required to keep the right level of ions like sodium in cells. And then there’s the repair. Damaged molecules and cells are constantly being replaced. The cells of the gut, for example, only last on average 4 days, and by then new cell have to be available.
Nothing runs without energy, and during intense work, everything works harder. (Even repair. Dark color in the urine after a marathon is probably due to red blood cells that were destroyed by the pounding of the feet on pavement.) At full tilt, the heart rate may be three times the resting rate, and the lungs puff faster and harder. All of these things are machines that require ATP as a power source. So, not only is the production of ATP by metabolic pathways only about 55-60% efficient, that ATP is shared between the muscles driving the athletic activity and other activities necessary to keep us alive. The result is that only about 18-26% of the energy content of that chocolate bar you ate shows up as work output powering the stationary bike or rowing machine. The rest is heat. (When Eddy Merckx performed his epic work rate on a stationary bike, he had to have fans blowing air over him to keep from overheating — he was probably generating about 1,900 watts, or 1,600 Calories per hour, of heat. You could sear a steak. Seriously.)
Making sense of the numbers
At the gym, the power output of your exercise machine is a hard number, measured directly by the force applied to the pedals. It’s the work actually done. The total Calories burned, on the other hand, is a calculated number, obtained by dividing the power output by your estimated efficiency. This is necessary because measuring the Calories expended directly is a bit of a science project, and can’t be done on your exercise bike. The non-intrusive method used today, called the “Doubly Labeled Water” (DLW) method, involves doping the body with water containing the heavy isotopes deuterium (2H) and 18O. The rate of excretion of this heavy water, together with some other data, is then used to calculate the total energy consumption. The DLW technique measures the total energy burned over a period of days or weeks, not minutes. It has been used extensively to monitor energy intake and expenditure in bicycle races such as the Tour de France, to see whether riders are taking in enough nutrients to keep them going. The numbers are eye-popping: a Tour rider may take in as much as 8,000 Calories a day and still be losing weight during several highly challenging stages of the race.
The DLW technique involves some estimations: for example, the total Calories expended are measured over a period of several days, and corrections have to be made for energy burned during the non-race hours. But the number obtained for total energy consumed while racing is undoubtedly very close to accurate.
Bikes can be equipped with special hubs or pedals to allow them to measure the power delivered, so the efficiency, the ratio of power delivered to total energy consumed, can be determined. The results for high-intensity bicycle road races have led to the number 24%; in a Tour de France racer, 24% of the energy consumed is delivered to the pedals. This is the efficiency. The net efficiencies of competitive rowers has been reported to be 27.5% (Fukunaga et al., Eur. J. Appl. Physiol. Occup. Physiol. 55(5):471, 1986).
The calculation of Calories expended on your stationary bike depends on what value the machine’s computer uses for the efficiency factor: how much of the total metabolic energy output that you produced actually drove the pedals? In the present example (150 watts for 30 minutes, equal to 64.5 Calories of actual work done), if we use an efficiency factor of 22%, which is a reasonable value for a healthy young adult, the total energy would be 64.5/.22 + 50 (50 is a half-hour’s worth of basal energy consumption while you’re awake), or 343 Calories. Your readout should be somewhere in that neighbourhood.
Heavier people are generally less efficient than lighter ones, and some machines ask for your weight and adjust the efficiency in calculating Calorie output. Some don’t. But there are no absolute numbers for a given weight and work rate, so the total energy output (Calories burned) is an estimate. There’s not much you can do about your energy efficiency, beyond losing some pounds if you’re overweight.
I determined the Calories burned and the power expended on an elliptical machine in my gym recently. What I observed was an apparent efficiency that varied with body weight and exercise intensity (from 100 to 300 watts). When I dialed in a very low body weight, the machine used a higher efficiency factor than if I dialed in a much higher body weight. That’s reasonable: an overweight person has more body to service than a lean person. And if I worked at a low power, the efficiency was also lower, compared to a high rate of power. That’s also entirely reasonable; at rest, very little of the total energy consumed is going to drive the pedals, most of it is being used to keep me alive. Judging by the ratios between power produced (watts) and total energy consumed (Calories), the efficiency values used by the machine ranged between 18 and 23%.
According to online exercise geeks, some makers of workout machines inflate the numbers produced by their machines. A sales gimmick: wouldn’t you rather work out on a machine that says you expended 400 Calories, rather than 300, for the level of effort reflected by heart rate and the feeling of exhaustion? If you are buying a machine for home use, you might prefer a machine that, based on your experience at the gym, indicated a higher expenditure of calories for a given level of effort on your part. Caveat emptor.