Table 7a. Infant growth in weight (g) by mother's post-partum body mass index (BMI), at 4-7 weeks

 Mother's BMI < 17 17-18.4 ³18.5 n Mean SD n Mean SD n Mean SD Birthweight <2800 Weight: at birth 17 2577 185 63 2419 217 106 2558 173 3 months 17 4866 532 63 5051 631 106 5235 677 6 months 17 6094 535 63 6285 759 106 6481 713 12 months 17 7699 568 63 7792 1104 106 7902 935 Weight gain 0-6 months 3579 452 3802 668 3928 676 6-12 months 1520 444 1505 617 1453 562 Birthweight ³2800 Weight: at birth 27 2891 168 124 3105 226 369 3145 263 3 months 27 5186 772 124 5301 642 369 5515 716 6 months 27 6260 681 124 6560 704 369 6660 829 12 months 27 7540 831 124 8012 820 369 8082 936 Weight gain 0-6 months 3278 646 3426 732 3515 829 6-12 months 1240 500 1528 569 1415 516

Numbers at birth adapted to numbers at 3 months.

Table 7b. Statistical analysis

 Test: < 2800, < 17 vs 17-18.4: attained weight + weight gains NS < 17 vs ³17: attained weight + weight gains NS < 18.5 vs ³18.5: birth weight t = -1.98 P = 0.049 weight at 3 months t = -2.35 P = 0.020 weight at 6 months t = -2.24 P = 0.026 weight at 12 months NS weight gains at 0-6 months/6-12 months NS ³2800, <17 vs 17-18.4: birth weight t = -2.68 P = 0.008 weight at 3 months/6 months NS weight at 12 months t = -2.64 P = 0.009 weight gain at 0-6 months NS weight gain at 6-12 months t = -2.38 P = 0.019 < 17 vs ³17: birth weight t = -3.10 P = 0.002 weight at 3 months t = -1.97 P = 0.050 weight at 6 months t = -2.34 P = 0.020 weight at 12 months t = -2.88 P = 0.004 weight gains at 0-6 months/6-12 months NS <18.5 vs ³18.5: birth weight t = -2.76 P = 0.006 weight at 3 months t = -3.56 P < 0.001 weight at 6 months t = -2.48 P = 0.014 weight at 12 months NS weight gains at 0-6 months/6-12 months NS < 2800 vs ³2800, BMI < 17: birth weight t = -7.37 P < 0.001 weight at 3, 6 and 12 months NS weight gains at 0-6 months/6-12 months NS BMI 17-18.4: birth weight t = -19.47 P < 0.001 weight at 3 months t = -3.01 P = 0.004 weight at 6 months t = -2.27 P = 0.024 weight at 12 months NS weight gains at 0-6 months/6-12 months NS BMI ³18.5: birth weight t = -25.88 P < 0.001 weight at 3 months t = -4.14 P < 0.001 weight at 6 months t = -2.01 P = 0.046 weight at 12 months NS weight gain at 0-6 months t = 4.98 P < 0.001 weight gain at 6-12 months NS

As commonly observed in undernourished communities, Madurese women were able to produce about 700 g breast milk per day in the first year (Steenbergen et al., 1989). However, these women happen to have a 4-7 week postpartum BMI of ³18.5. To assess lactation performance of CED women, infant growth can be used as a proxy for breast milk intake.

Although it is customary to force-feed infants from as early as the first week after birth, the semi-solid additional food (mashed rice and banana) did not influence milk production in the study population (Steenbergen et al., 1991). To adjust for birth weight, infant growth is presented in two categories with a birth weight of 2800 g as cut-off. The infant growth was obviously determined by birth weight (Table 7). The data also reveal the expected upward gradient in weight and weight gain in the first 6 months by increasing post-partum BMI of women in both birth weight categories but only attained weight was significantly lower in the group with a BMI <18.5. Hence, infants remained in their growth channel and there was no catch-up growth. At 12 months the differences by BMI categories disappeared, indicating that the heavier babies at birth started to falter in growth in the second half year of life.

Table 8. Maternal weight (kg) changes in the first year post partum by 4-7 week post-partum (BMI) categories, mean and SD

 4-7 week post-partum BMI < 16 16.0-16.9 17.0-18.4 18.5-22.4 ³22.5 Months n Mean SD n Mean SD n Mean SD n Mean SD n Mean SD Post partum 1 16 35.2 3.1 52 37.5 2.2 230 40.7 2.7 512 44.9 3.6 52 54.8 5.4 3 16 35.7 3.0 52 38.0 2.7 230 40.9 3.0 512 44.7 3.8 52 54.6 5.4 6 11 36.5 4.3 39 38.3 3.1 210 40.9 3.3 502 44.2 4.0 43 54. 6.0 12 11 36.9 4.2 39 38.4 3.6 210 41.1 3.5 502 43.9 4.1 43 52.7 5.6 Change at 1-6 11 +1.3 2.3 39 +0.9 2.6 210 +0.1 2.2 502 -0.7 1.9 52 -0.7 2.6 at 1-12 11 +2.5 3.1 39 +0.8 3.3 210 +0.4 2.9 502 -0.9 2.3 43 -2.1 3.3

Weight changes between

< 16 vs 16-16.9: NS.
< 17 vs 17-18.4: t = -2.40 P = 0.018.
< 17 vs ³17: t = -4.56 P <0.001.
< 18.5 vs ³18.5: t = -6.57 P <0.001.

Interestingly, in the first 6 months the weight gains of infants with a birth weight <2800 g were consistently and significantly higher than those of the heavier ones at birth. It is tempting to attribute this difference to a substrate limitation to milk production, being insufficient for the bigger babies. One can at least conclude that infants, born to CED women, are not 'in the driving seat', in determining energy intake according to their needs.

The repercussion of lactation on the women is illustrated in Table 8. Contrary to the appreciable weight loss commonly observed among well-nourished lactating women (NAS, 1991), as a group Madurese women did not lose much weight in the first year post partum (0.2 kg, not shown in Table 8). Broken down by post-partum BMI categories at 4-7 weeks, the same pattern of weight changes can be observed as during pregnancy. The severe CED women gained weight (1.3 kg) in the first 6 months and 2.5 kg in the whole year. So did the moderate CED group but only in the first 6 months and to a lesser extent (0.9 kg). Mild CED women had only a slight weight increment of 0.4 kg over the 12 months. On the other hand, the non-CED group lost weight and the fatter women more so (2.1 kg) at BMI (22.5 and 0.9 kg at BMI 18.5-22.4 over the year. There was no difference between primiparae and multiparae. As for the weight changes over pregnancy, the differences in the first 6 months and over the whole period of infancy were highly significant at BMI cut-offs of 17.0 and 18.5 (P < 0.0001).

Trends over two consecutive reproduction cycles

There were 217 women who had given birth twice in the study period and from whom complete data were available on pre-pregnant weight and post-partum weight at 6 and 12 months for the consecutive reproduction cycles. Table 9 shows a remarkable similarity in weight trends by BMI categories as compared with the cross-sectional data. The most severe CED women gained weight while among non-CED weight did not change much.