Nutrition lab: coffee experiment

coffee, coffee beans, roasted-5447420.jpg

The alarm goes off. Its 5am. My mind immediately goes to the kitchen- get a pot of coffee going. Except, not today. I’m taking part in an experiment, testing the effect of caffeine on bioimpedance analysis measurements- you know the machine in the gym that measures your body fat percentage- and I have to come to the lab fasted, and without my morning dose of caffeine. An interesting thing happens, now that I am awake I am not entirely sure what do with myself next. Making coffee is such a part of the autopilot that I spend the next five minutes still in bed, in a sort of half in-half out position. I rationalise that I don’t need to get up yet anyway, because normally this five minutes would be spent waiting for the kettle to boil. Soon however the dog has realised that I am awake and we’re on our way out of the door for the morning walk around the block. I get back and enter the kitchen. Still can’t have coffee. Or breakfast. Or water. Well, nothing for it, time to get ready and onto the tube to get to the lab.

The reason we are doing this experiment is exactly to do with the disorientating experience I’ve just had in not having any coffee that morning. Bioimpedence analysis works by sending a series of weak electric impulses through the body, meeting resistance and capacity along the way through the cell membranes, and by extension, the water found within and around the cells. It follows that hydration makes a big difference to the accuracy of the measurements, and other things that affect fluid levels in the body including food in the belly, recent intensive exercise and the menstrual cycle will affect the measurements taken from the machine. It is recommended that coffee and caffeine are avoided as well, because of the potential fluid changes associated with its diuretic effects. But we’ve since established that coffee isn’t a dehydrator- the actual amount of water present in a typical cup of coffee is enough to overcome the diuretic effect of the caffeine. So the experiment aims to show that we can drop the need to avoid caffeine for accurate measurements.

I get to the lab, say hi to my colleagues and we’re ready to begin the measurements. I’m following the protocol of the study to get a first hand idea of what its like to deprive yourself of caffeine and food for a couple of hours as the rest of the study participants would, and to get an idea of what this would be like for people visiting clinics for their body composition data. The coffees are presented in white paper cups with coded numbers written on them in permanent marker. I start drinking the black, unsweetened instant coffee being used for the experiment and wonder if I am getting the decaf treatment this morning. Within a shockingly short space of time, I start to feel (or imagine I do) the blood moving through my veins more quickly, heat radiating from my face. Unless this is a powerful placebo effect, it appears that I have just ingested 300 mg of caffeine (we had the chemists test the samples) and its starting to take its effect. I am still fasted however, and expected to take the measurements of the other people in the lab taking part in the experiment this morning. A headache starts to form behind my eyes, or do I imagine it is? The numbers on the machine printout swim a little bit as I take them down on the clipboard. Perhaps it was the decaf treatment after all…

Among the insights I gain from the experience, is that even with my fairly moderate coffee drinking habit, caffeine dependence is surprisingly powerful. This isn’t actually a problem, as long as your caffeine intake is moderate (about 300 mg/day, or 2-3 cups of coffee). As the results start to come out of the experiment, we notice some interesting findings about the hydrating effect of caffeine as well- we’ve been taking urine samples and measuring the amount of solute in them over the course of the experiment- the more solute present, the more concentrated the urine, the more water is being reabsorbed by the kidney. On average, there was slightly more solute in the coffee drinkers over the course of an hour (+20) and slightly less in the decaf drinkers (-80)

What we have also found is that there appears to be a difference in the body fat percentage depending on the treatment- while the statistical tests haven’t been done yet, it looks like drinking decaf coffee is associated with a reduction in body fat percentage of around 0.6%, before returning towards baseline at the end of the hour, whereas drinking coffee containing caffeine is associated with a slight increase in the body fat measurements (around 0.4%). 

This seems promising, only a 0.1% change in body fat percentage is not exactly a large, or even noticeable amount. It isn’t even as large the natural variation in accuracy we would expect from the machine itself (and it is a very precise medical-research grade machine).

Coffee is a big part of life in London, is hydrating, doesn’t appear to mess with body composition measurements, and is used by many athletes as an ergogenic aide. Perhaps more on that in another blog post. For now, we’ll wait for the rest of the results, and I will cure this headache with a nice big americano.