Excessive accumulation of fat in women of childbearing age is a concern, since obesity is an important cause of morbidity and mortality. Deposition of fat during pregnancy, which is not metabolized during lactation, may contribute. However, the individual effects of age and gravidity on fat accumulation have not been disentangled. Based on multiple linear regression analysis of anthropometric data from 1113 pregnant women from Zimbabwe, we found evidence to suggest that fat deposition is an effect of age rather than gravidity that is precipitated by the first pregnancy. International Journal of Obesity (2002) 26, 1274 -1276. doi:10.1038=sj.ijo.0802074Keywords: pregnancy; age; fat deposition; obesity Understanding the role of pregnancy in the etiology of obesity is important, since obesity is an important cause of morbidity and mortality from chronic disease. 1 Despite colinearity of age and gravidity, it is generally assumed that fat accumulates with age due to fat retention after pregnancies. We attempted to disentangle the effects of age and gravidity based on anthropometric data from HIV-uninfected women registering for antenatal care in Harare, Zimbabwe. Permission was obtained from Medical Research Council, Zimbabwe. The study population has previously been described. 2,3 Anthropometric measurements were taken with the women barefoot and wearing light clothing. Height and weight were measured to the nearest 0.1 cm and 0.1 kg, respectively. Triceps skinfold thickness (TSF) was measured to the nearest 0.1 mm in the midline of the posterior aspect of the left mid-upper arm. At the same level, the arm circumference (AC) was measured to the nearest 0.1 cm without compressing the tissue. Body mass index (BMI), arm muscle area (AMA) and arm fat area (AFA) were calculated, using the formulae: BMI ¼ weight=height 2 (kg=m 2 ), AMA ¼ (AC 7 (TSFÁ p)) 2 =(4 Á p) (cm 2 ) and AFA ¼ [AC 2 =(4 Á p)] 7 AMA (cm 2 ), as previously described. 4 ANOVA was used to compare means between groups. Multiple linear regression analysis was used to test for interactions between age and gravidity, and to estimate the statistical effects of age and gravidity on body weight, AFA and AMA, while controlling for height and gestational age. When interactions were identified, then interaction terms were computed to estimate the effect of age for each level of gravidity.Of 1669 women examined, 1113 were found to be HIVuninfected. The mean gestational age was 29.2 (22 -35) weeks. The mean age was 24 (range 14 -45) y: 46.5% were primigravidae (mean age 20.4 (range 14 -40) y), 23.8% secundigravidae (23.6 (16 -35) y), 14.6% tertiogravidae (26.9 (20 -40) y), and 15.2% multigravidae (4 þ 32.2 (20 -45) y). Compared to data from a study among pregnant US women, 5 the mean AMA was similar (38 vs 38 cm 2 ), whereas the mean AFA was lower (21 vs 27 cm 2 ). Anthropometric variables were compared over categories of gestational age, age and gravidity (Table 1). AFA declined with gestational age. All anthropometric variables, except height, increased with age a...