Eat Stop Eat

Best Weight Loss Programs That Work

Get Instant Access

0 1 2 3 4 5 6 7+ Parity (total number of reported live births)

Figure 20.2 The relationship between parity and body weight among women examined by Williamson et al. (28). Adjusted for the confounding effects of age, education, smoking status, drinking status, physical activity, health status, marital status, and dieting control by comparing body weight after pregnancy with that recorded before conception. In this way, any differences in weight can be directly attributed to events that occurred during the intervening per correct for ageing-related weight gain in the pregnant (case) women. In practice it is actually very difficult to recruit an appropriate group of control women since any differences in the amount of weight gained by pregnant and non-pregnant women might simply be the result of confounding. For example, non-pregnant women without children have been shown to gain more weight over time than non-pregnant women with children (28). To avoid these problems other researchers have corrected for ageing-related gains by employing multivariate statistical techniques to account for the weight gain that is independently associated with the study interval (49,50). In a pioneering study by McKeown and Record (51), this problem was overcome by using each woman's rate of weight gain during the second year after delivery to correct for the ageing-related weight gain during the period of study.

There are a variety of longitudinal studies that have examined the development of body weight following pregnancy, and the results of those studies whose primary aim was to assess the impact of pregnancy on body weight are summarized in Table 20.2. A more exhaustive survey of studies published up to 1995 is given by Harris and Ellison (12). A critical review of these longitudinal studies (12) revealed three studies to be particularly robust (28,46,48). The first of these studies formed part of an extensive 4-year follow-up study monitoring the development of obesity in young Dutch adults (46). This study compared the changes in body mass index (BMI) of a group of women who became pregnant (cases) with the changes in BMI of a group of women who did not fall pregnant (controls). All pre-pregnant body weights were measured before conception, and these measurements were taken at least 9 months after any preceding pregnancy. The follow-up measurements of body weight were taken, on average, 12 months after delivery so that all of the women in the study had sufficient time for their body weight to return to its pre-pregnant level before follow-up measurements of body weight were taken. Rookus et al. (46) showed that the 40 cases gained, on average, 0.61 + 0.15 kg/m2 during the study period, whereas the 400 non-pregnant controls gained only 0.27 + 0.05 kg/m2, the crude difference (0.34 + 0.16 kg/m2) being statistically significant. However, after accounting for differences in age, level of education, parity, giving up work, smoking status and change in smoking status be tween the cases and the controls, the differential change in BMI between the two groups was no longer statistically significant (0.15 + 0.21 kg/m2).

The second study by Smith et al. (48) examined the relationship between pregnancy and persistent changes in weight among young US women in a prospective investigation that used data from the CARDIA (Coronary Artery Risk Development in Young Adults) study. In this study, measurements of body weight were taken at a baseline examination in 1985-1986, at a second examination in 1987-1988, and again in 1990-1991. Women were included in the study if they remained childless throughout the three examinations (nulliparous controls) or if they had a single pregnancy between examinations 1 and 3 (primiparous or multiparous cases, depending on their baseline parity). The measurement of body weight taken at baseline served as the pre-pregnant weight, and all multiparous cases were at least 12 months postpartum at baseline to exclude the effects of any previous pregnancies. Similarly, cases were excluded if they were not at least 12 months postpartum at the time of examination 3, so that any expected return to pre-preg-nancy weight should have occurred prior to the follow-up examination. Their results showed that black mothers having their first child gained 3.3 kg more weight than black nulliparous control women (P = 0.02). Likewise, white mothers having their first baby gained 2.3 kg more weight than white nulliparous control women (P = 0.03). In contrast, all those mothers who delivered a second or higher order child during the period of study did not gain significantly more weight as a result of their (later) pregnancies than did the nulliparous controls. Overall, mothers who experienced pregnancy gained 0.92kg more weight than the controls (2.76 kg more for primiparous mothers, and 0.52 kg less for multiparous mothers), which reduced to 0.80 kg after adjusting for a variety of confounders (2.35 kg more for primiparous mothers, and 0.42 kg less for multiparous mothers).

The third study by Williamson et al. (28) compared the weight development of mothers with both parous and nulliparous controls. In this study Williamson et al. (28) presented results from the analysis of a nationally representative cohort of US white women aged 25-45 years, who were followed for approximately 10 years from 1971-1975 to 19811984. All pre-pregnant weights were measured before conception during the baseline examin-

Table 20.2 Longitudinal studies whose primary aim was to investigate the effect of pregnancy on maternal body weight


Date of data collection

Type of pre-pregnant weight measurement

Postpartum follow-up

Control for ageing (kg/year)

Long-term weight gain x + SEM (kg)"




Was this article helpful?

0 0
Losing Weight Natures Way

Losing Weight Natures Way

Feel Healthier And Look Better Without Resorting To Drugs! Would you like to get your hands on a this report that can teach you everything you need to know about losing weight naturally?

Get My Free Ebook

Post a comment