range (Appendix C, Tables C-1A and C-1B). For example, the pattern of GWG by maternal BMI category was examined in a large cohort of women visiting the University of California, San Francisco clinics (Abrams and Selvin, 1995; Carmichael et al., 1997). Mean rate of gain was 0.169 kg per week in the first trimester. Mean weight gains were higher in the second (0.563 kg per week) than the third trimester (0.518 kg per week) in all groups except for obese women; and mean gains in the second and third trimester were higher in underweight and normal weight women than in overweight and obese women. Birth weight was correlated most strongly with gain in the second trimester (32.8 g/kg GWG versus 18 and 17 g/kg in the first and third trimesters, respectively).
In another study, mean rates of GWG in non-obese, low-income black and white women were 2.48 kg in the first trimester and 0.49 and 0.45 kg per week in the second and third trimesters, respectively (Hickey et al., 1995). In contrast, GWG rates among predominantly Hispanic women (n = 7,589) participating in the Prematurity Prevention Project were similar in the second (0.52 kg per week) and third trimesters (0.53 kg per week) (Siega-Riz et al., 1996); although the third-trimester gain was slightly lower in women who delivered preterm (0.50 vs. 0.53 kg per week). A similar GWG pattern has been observed in adolescents, although the median gain and rate of gain were higher throughout gestation; from mid-pregnancy to term, the rate of gain was 0.51 kg per week (Hediger et al., 1990).
In summary, the pattern of GWG is generally higher in the second trimester and is related to maternal pregravid BMI. However, pattern of GWG can vary depending on maternal ethnicity and age.
Luke and colleagues (1992) conducted a series of observational studies on outcomes associated with the rate of GWG in women with twin pregnancies who delivered infants at 37-42 weeks’ gestation and with mean birth weights exceeding 2,500 g. They (1992) found that low rate of GWG, defined as < 1.0 pound/week, was associated with a significant decrease in mean birth weight for twins compared to singletons (β, −0.137; p = 0.001). Significantly higher rates of GWG in the third trimester were observed among women whose mean birth weights for twins were ≥ 2,500 g compared to women with birth weights for twins that were < 2,500 g, regardless of BMI category; and no significant differences were seen for first and second trimester GWG rates.
Among a large multiethnic population of 646 twin pregnancies at ≥ 28 weeks’ gestation, birth weight increased by 14, 20, and 17 g for each pound of weight gained between 0 and 20 weeks’ gestation, 20 and 28 weeks’ gestation, and 28 weeks to birth, respectively (Luke et al., 1997). Mean total GWG