Positive association between resting energy expenditure and weight gain in a lean adult population.

BACKGROUND: Weight gain in adulthood is common, from modest gains in developing countries to substantial increases in Western societies. Evidence of the importance of energy expenditure in adult weight change has been limited to studies conducted in Pima Indians, in whom resting energy expenditure (REE) was found to be inversely associated with weight gain.
OBJECTIVE: The aim was to determine whether REE was predictive of weight change in lean Nigerian adults.
DESIGN: Weight was measured in 744 adults on 2-4 occasions over 5.5 y. REE was measured in the second follow-up examination. Sex-specific, mixed-effects models with REE, fat-free mass, and age as fixed effects were used to test the association between REE and weight change.
RESULTS: Adults aged >19 y (n = 352 men and 392 women) were included in these analyses. At baseline, the mean (+/-SD) age was 45.9 +/- 16.1 y for the whole population; the mean weight was 61.4 +/- 10.7 and 58.1 +/- 12.1 kg and body mass index (in kg/m(2)) was 21.4 +/- 3.2 and 23.1 +/- 4.0 for men and women, respectively. Over a mean 5.5 y of follow-up, the age-adjusted weight gain was 0.42 kg/y for the men and 0.59 kg/y for the women. In mixed-effects models, REE was positively associated with weight gain in both men and women (P < 0.001). No significant association was observed in participants who lost weight.
CONCLUSIONS: In contrast with observations in overweight Pima Indians, REE adjusted for body size and composition was positively associated with weight gain in lean Nigerian adults. This suggests either that the potential for differential regulation of body weight in lean compared with overweight populations exists or that the increased REE in this population was the result, rather than cause, of weight gain.