effect of food composition,
meal size and exercise
The TEF can last for several hours after a meal
and will depend on the amount of energy con-
sumed and the composition of the meal. In
general, the thermic effect of a mixed meal is esti-
mated to be 6–10% of total daily energy intake;
however, the total TEF will also depend on
the macronutrient composition of the diet. For
example, the thermogenic effect of glucose is
5–10%, fat is 3–5% and protein is 20–30% (Flatt
1992). Carbohydrate and fat have a lower
thermic effect than protein because less energy is
required to process, transport and convert carbo-
hydrate and fat into their respective storage
forms. Conversely, protein synthesis and metab-
olism are more energy demanding. Thus, diets
higher in fat will have a lower TEF than diets that
contain more carbohydrate or protein. In addi-
tion, a diet high in energy will have a higher TEF
than a diet lower in energy because there is more
food to be digested, transported and stored. For
example, the TEF of an individual who consumes
12.6 MJ (3000 kcal) daily would be approximately
756–1260 kJ · day–1(180–300 kcal · day–1), while an
individual consuming only 6.3 MJ (1500 kcal)
daily would have a TEF of 378–630 kJ · day–1
(90–150 kcal · day–1). The total TEF for a day does
not appear to be influenced by meal size or
number, as long as the same amount of energy is
consumed throughout the day (Belko & Barbieri
1987). Thus, the TEF will depend both on the
amount of energy consumed each day and the
composition of this energy.
Although exercise may influence the TEF,
there are few data available on the effect of exer-
cise before and after a meal in trained athletes.
One study in trained swimmers reported that 45
min of swimming significantly increased the
metabolic response to a meal when the meal
preceded the exercise (104.2 kJ · h–1, 24.8 kcal · h–1)
compared with no exercise (84.8 kJ · h–1, 20.2 kcal ·
h–1) (Nichols et al. 1988). However, this difference
is so small that its long-term significance on
energy regulation is negligible, especially con-
sidering the high variability in the termic effect of
a meal (TEM) measurement between individu-
als. Similar results are reported by Bahr (1992),
who exercised physically active males for 80 min
at 75% V.
o2max.and measured oxygen consump-
tion after exercise. In the treatment condition
subjects were fed a meal 2 h after exercise, while
subjects fasted in the control condition. They
found only a 42-kJ (10-kcal) difference between
the two conditions over a 5-h postexercise
period.Measurement of energy expenditure
Energy expenditure can be measured in the lab-
oratory or estimated using prediction equations.
Since access to a laboratory for the measurement
of energy expenditure (calorimetry or doubly
labelled water) may be limited, this review will
focus on the prediction methods used to estimate
energy expenditure.predicting energy expenditure
One of the most commonly used methods for
estimating TDEE is to predict RMR using a pre-
diction equation and then multiply RMR by an
appropriate activity factor (Food and Nutrition
Board 1989; Montoye et al. 1996). A number of
prediction equations have been developed to
estimate RMR, but most have been developed
using sedentary populations. To date, no equa-
tion has been developed to predict the RMR of
athletes who may spend hours in training each
week. Some of the commonly used RMR predic-
tion equations and the population from which
these equations were derived are discussed
below. To determine which of these equations
work best for athletes, Thompson and Manore
(1996) compared the measured RMR values from
indirect calorimetry with predicted RMR values
using the following equations. In all these equa-
tions, weight (wt) was measured in kilograms,
height (ht) in centimetres and age in years; LBM
stands for lean body mass.- Harris and Benedict (1919); based on 239
subjects, 136 men (mean age, 27±9 years; mean
weight, 64±10 kg) and 103 women (mean age,