The prevalence of female obesity has increased steadily over the last 40 years in both the UK and the USA (1,2). Even during the last decade, the prevalence of female obesity (BMI > 30) in Britain rose from 8% in 1980 to 12% in 1987 (3). This increase led the British government to identify 'the reduction of obesity' as one of its main targets for health improvement (4), yet in spite of this initiative female obesity continues to rise and currently stands at 20% (5). Paradoxically, the increasing prevalence of obesity has occurred in the face of remarkable social pressure to be thin, and data from weight surveys consistently show more women than men to be affected (Figure 20.1). If pregnancy is a determinant of obesity, then this might explain the higher prevalence of obesity in women. The accelerated rate of weight gain observed among young women might also be a reflection of the current trend towards encouraging women to gain more weight during pregnancy (6).
PREGNANCY AS A CAUSE OF OBESITY: THE EVIDENCE
Women in developing countries often become progressively malnourished with successive pregnancies and this has been shown to lead to maternal depletion rather than maternal obesity (8,9). However, in developed countries most investigators report a net increase in body weight with pregnancy that may persist and even increase with successive pregnancies (10,11).
Anecdotal Evidence and Evidence from Case Series Studies
Pregnancy has long been thought to be aetiologi-cally related to obesity (12) and as early as 1939, Greene (13) observed that some of his obese female patients: 'gained 15 to 25 pounds with each pregnancy, maintained the added weight, and thus became obese after three to six pregnancies.' By 1949, Sheldon (14) suggested that it was: 'a matter of common observation that women may at times develop a severe obesity after having a baby', and he coined the term 'maternal obesity' to describe this phenomenon. There is certainly a wealth of anecdotal evidence that pregnancy can lead to obesity, and evidence from case series studies suggests that obese women often cite pregnancy as a triggering life event for the development of their obesity (15-17). The results of such studies should, however, be interpreted with caution as women may retrospectively report pregnancy as a socially acceptable cause for an obesity that resulted from excessive energy intake. The validity of these studies can also depend on the accuracy with which obese women report their body weight (18). Although self-reported body weight is highly correlated with measured weight (19), there is considerable interindividual variation in the accuracy with which it is reported. For example, Stevens-Simon et al. (19),
International Textbook of Obesity. Edited by Per Bjorntorp. © 2001 John Wiley & Sons, Ltd.
found that women underestimated their body weight by an average of 1.3 kg. This figure, however, ranged from a mean over-report of 1.4 kg for women who were initially underweight (less than 90% of ideal weight-for-height) to a mean under-report of 5.0 kg for women who were initially overweight (more than 120% of ideal weight-for-height). Therefore, women's desire to report a body weight that conforms to their own ideal or to perceived norms represents an important methodological limitation for studies that rely on self-reported information. Nevertheless, case series studies provide the first level of evidence that pregnancy may trigger obesity, even if only in a small percentage of women.
Cross-sectional Studies that Examine the Effect of Pregnancy on Body Weight
More compelling evidence comes from cross-sectional surveys that examine the effect of childbear-ing on body weight (Table 20.1). Some of the very first systematic studies were based on demographic and industrial surveys of women's body weight (20-24). These surveys found that women with children had higher body weights than those without, and that maternal body weight increased with increasing parity (parity approximates to 'number of live births'). Since this time, there have been numerous other studies demonstrating the same association (25-28). Parity is, however, associated with a number of sociodemographic characteristics, such as higher maternal age, lower social class and marriage, all of which are independently associated with an increased risk of weight gain (23,29). It is therefore possible that some, or all, of the observed effect of parity on body weight results from failure to control for the confounding effects of factors that are inherently associated with pregnancy but independently associated with weight gain.
A number of population-based cross-sectional studies (25,28), have employed stratification or multivariate statistical techniques to adjust for the confounding effects of age and other factors. In so doing, they reduce the possibility that any remaining differences in body weight between women of differing parity are simply the result of confounding. Williamson et al. (28) undertook one such analysis using data that had been collected during the first US National Health and Nutrition Examination Survey. This analysis showed that for women with children, body weight increases with each additional live birth, even after accounting for a large number of known confounding factors (Figure 20.2). Women with three or more live births had higher mean body weight than women with fewer live births, with the greatest increases in mean body weight being associated with five or more live births. Such cross-sectional studies provide strong evidence that pregnancy is independently associated with persistent weight gain. However, it is probably not possible to control for all of the factors that might be responsible for differences in energy balance between women. There might also be inherent, immeasurable, and therefore uncontrollable, differences between women who choose to have children and those who have none.
Many cross-sectional studies are essentially opportunistic analyses of large, readily available, epi-demiological data sets. They therefore often fail to control for many of the psychosocial and behavioural confounders associated with pregnancy, simply because these data are not routinely collected on a national basis. As such, cross-sectional studies can rarely differentiate between the effects of child bearing and child rearing on body weight.
Longitudinal Studies that Examine the Effect of Pregnancy on Body Weight
Perhaps the best evidence comes from longitudinal studies that effectively use each woman as her own
Date of data collection
Control for potential confounders Sample size (n)
Was this article helpful?
Tired of Trying To Loose Weight And It Never Works or You Have To Starve Yourself Well Here's A Weight Loss Plan That takes Care of Your Weight Problem And You Can Still Eat. In This Book, You’ll Learn How To Lose Weight And Not Feel Hungry! In An Easy Step-By-Step Process That Enables You To Feel Good About Loosing Weight As Well As Feeling Good Because Your Stomach Is Still Full.