The effect of temperature on BIA measurements

Bioelectrical Impedance Analysis (BIA) was created from the need for accurate, reliable, and convenient methods of assessing body composition. Various methods are used for this purpose, including Skinfold Caliper calculations, Dual-Energy X-Ray Absorptiometry, Underwater Weighing, and of course BIA itself. Each has its own benefits and limitations, and no matter what method is used, it is always important to understand how the technology works, to help improve accuracy.
As the name suggests, BIA utilizes bioelectrical impedance, which is measured by sending a safe electrical current through the subject's body (oftentimes using multiple frequencies). It's a well-established technology used by Charder in its body composition analyzers due to ease of use, low cost, reliability, lack of side effects, and overall convenience.
However, because the body's impedance is something that can be temporarily affected by various factors, there are some things that users need to take into account to improve validity and consistency of measurement results.
If you have conducted a BIA scan before, you may be familiar with various rules, such as the need to remove shoes and socks and gloves, so the electrical signal can be measured properly.  You may have also heard that you shouldn't measure shortly after eating and drinking, and the reason for this is because of the additional weight, as well as the temporary change in body fluid distribution from the digestive process.
Today, we'll be explaining the effect of temperature on BIA measurements, using a paper by researchers from the University of California: "Body composition analysis by bioelectrical impedance: effect of skin temperature"
Bioelectrical impedance analysis (BIA) was used to estimate body water and composition under both cool (14.4 degrees C, dry bulb) and warm (35.0 degrees C) ambient conditions
The corresponding BIA resistances were 461 +/- 48 omega and 426 +/- 47 omega, respectively. (Mean reduction was 35.0 +/- 9.8 omega, t = 10.13, P less than 0.01). 
Consequently, predicted fat mass was significantly lower in the warm than in the cool condition (8.8 +/- 3.2 kg vs. 11.0 +/- 3.7 kg; mean difference 2.23 +/- 0.69 kg, t = 9.22, P less than 0.01). These findings indicate that varying skin temperature by altering ambient temperature significantly changes resistance measurements and the estimation of total body water and percent fat by BIA. The observed changes in resistance are consistent with an apparent expansion of conductor volume in the warm environment and a reduction in the cooler condition. In this regard, the temperature-induced change in resistance could be due to alterations in cutaneous blood flow and/or compartmental distribution of body water. Thus, BIA measurements should be taken only under well-standardized ambient conditions.
As can be seen from the research results, temperature has an effect on impedance, which then has an affect on the calculated body composition results. Similar results were replicated and confirmed by other researchers as well.
This is why Charder's standard recommendation for measurement is to conduct them in "room temperature", between 24-28°C / 75-82°F.
Can you use BIA Body Composition Analyzers in colder or hotter environments? Yes, but maintaining consistency is key. If you always measure in a room that's 21°C then the fact that it's colder than the recommended range is not a problem, since change in temperature has been removed/controlled for as a variable affecting impedance.
If you are using a BIA device in a place where it gets particularly hot or cold outdoors, then we would recommend that subjects entering the measurement location from outside should be given some time to cool down/heat up to a more "normal" temperature before measurement!
(NOTE: This is also why BIA measurements are affected by fever, and you should wait until you've recovered before conducting a measurement)

Recommended reading