target="_blank" rel="nofollow" href="#ulink_04ef263f-9dd6-5d19-9d0d-4db27f8f1adf">Figure 1.2 The prevalences of obesity in men and women in cohorts with over 263,000 adults either from Asia or from Australasia and Iran (depicted as Caucasians) [30]. Superimposed on this graph are data from the Mexican national survey in 2000 [35]. The Mexican study compared the national survey data with nationally representative data for US non‐Hispanic Whites, but these data are not shown in this graph as they were almost identical to those of the Australasian/Iranian data from the Asian study.
Furthermore, the seemingly genetically obesity‐prone Pima Indians from Mexico and Arizona in the United States show that with similar genetic profiles, there are big national differences in BMI and diabetes prevalences which are largely environmentally determined [39]. Very low obesity and diabetes rates occur in the hard‐working, home‐farming Pima Mexicans consuming a 25% fat, high fiber diet with a negligible sugar content [40]. So it is difficult to be sure what constitutes a greater genetic propensity to diabetes in different ethnic groups and how this relates to abdominal obesity without taking account of their prevailing diet and physical activity [41]. Indeed, the propensity to develop type 2 diabetes has been related to the duration of being overweight/obese, as well as the magnitude of excess weight [42].
The impression that Caucasians are relatively protected from diabetes associated with weight gain was further amplified by our findings in the Middle East. Carefully conducted randomly selected populations in each of the 21 countries covered by the WHO Eastern Mediterranean Region (EMR), i.e. including not only the Gulf countries, Lebanon, Syria, and the North African countries abutting the Mediterranean but also Djibouti, Somalia, Sudan, Afghanistan, and Pakistan showed (Fig. 1.3) that, despite the wide‐ranging levels of average BMI in both men and women, the national average diabetes rates are appreciably higher in the EMR than in the 28 countries of the Europe Union throughout the range of average BMIs.
Figure 1.3 The relationship between the prevalences of obesity and diabetes in each of the 21 WHO Eastern Mediterranean countries compared with equivalent data from the 28 countries of the European Union. Diabetes prevalence rates are about twice as high in the Eastern Mediterranean countries as in the European Union at equivalent obesity rates. These data are based on randomly selected adults with measured anthropometry and fasting blood glucose levels in each country undertaken according to a standard protocol (the WHO STEPS program).
(Source: Reproduced from Alwan et al. [43].)
If these differences are not genetic, then what might be the basis for the increased susceptibility to abdominal obesity and diabetes? Barker et al. [44], in his original study, claimed that lower birth weight in babies of pre‐Second World War women in England seemed to determine an increased propensity of the offspring to develop the metabolic syndrome and diabetes 64 years later. The conditions amongst the working class in England at that time had already been documented as nutritionally totally inadequate [45], so Barker ascribes the propensity to metabolic syndrome and diabetes as reflecting the impact of maternal malnutrition with induced poor fetal growth. Yajnik in Pune, India, then showed with Barker startling differences between the offspring of mothers in Pune and those of now well‐fed mothers in Southampton, England [46]. The Indian children were smaller, with less lean tissue, and relatively fatter with distortions of the organ sizes which already related to increased insulin resistance in the newborn Indian babies. This seemed to be proportional to the degree of vitamin B12 deficiency and the discordant folate/B12 nutritional status of the mother [47], which has been shown to influence fetal growth. These vegetarian women were often being given inappropriately high doses of folic acid without any B12 and most likely would have been eating very modest amounts of food providing nucleic acids. Thus the generation of nucleotides in the crucial phase of early fetal development would have been compromised due to the lack of vitamin B12 required to generate methyl groups through folate remethylation.
These early nutritional effects relating to organ metabolism are probably amplified if there is some subsequent childhood malnutrition, because although they can recover their intestinal absorptive capacity on refeeding, their ability to mobilize insulin after a standard glucose test remains markedly reduced despite being extremely well‐fed for months and with body weights that had returned to the normal weight for their heights [48]. Further analyses have shown that this effect is long‐standing because adults in Jamaica who had been malnourished as children had a persisting impairment of insulin secretion [49], and this defect was also seen in survivors of early fetal deprivation during the Dutch famine [50].
The Millennium report for the United Nations on the global prevalence of persisting childhood malnutrition [51] highlighted the fact that almost all non‐Western countries after the Second World War had high prevalences of childhood malnutrition leading to long‐standing global, intergenerational malnutrition, which persisted throughout life. So a lifespan approach [52] to considering the problem of adult obesity begins with the nutritional state of the mothers before and during pregnancy with all its pathological and epigenetic implications for the offspring. We already know that the increase in body weight in previously malnourished women as they enter pregnancy leads to a much greater propensity to gestational diabetes [53], and they are then more likely to develop diabetes later themselves as well as having bigger and fatter children.
These global states of malnutrition almost certainly have lasted for millennia, so the early descriptions of obesity associated with ill health described by Bray are in line with the observed debilitating disease such as diabetes resulting from the impact of excess weight gain in people with lifelong malnutrition. In contrast, the Roman description of relatively healthy obese may then have reflected the better overall nutrition of the Romans in their Mediterranean environment.
Historical analyses of contributors to obesity
Weight gain leading to obesity occurs when energy intake from food exceeds energy expenditure from physical activity and metabolic processes over a considerable period. There has been much speculation as to the main driver of the global obesity epidemic, and this has often led to intense debate (see the analysis by Swinburn et al. [54] and the responses this generated). A complex and diverse range of factors can give rise to a positive energy balance, but it is the interaction between a number of these influences, rather than any single factor acting alone, that is thought to be responsible. The genesis of obesity at an individual level often focuses on a lack of cognitive control over personal behaviors that directly influence energy intake and ignores the critical role of physiological processes in driving or attenuating these behaviors. In addition, it is now recognized that powerful societal and environmental forces influence energy intake and expenditure through effects on dietary factors and physical activity patterns and may overwhelm the physiological control of body weight. It is the emergence of these environmental forces and the adjustments brought about by rapid changes in society which are the focus of most attempts to explain the emergence of the global obesity epidemic. The key elements of this transition and the emergence of our understanding around them are now set out below.
Marked declines in society’s physical activity
Given how sudden the development of obesity has been on a population basis, one has to ask what happened in the early 1980s to make such a difference to the prevalence of obesity in our environment? One clear societal change is the introduction of computers for a myriad of tasks in the early 1980s but in particular a complete transformation of many people’s office hours so now they spend their time sitting while working on their computers for hours on end. Even in such physically active jobs as construction or repairs, the introduction of a myriad of mechanical aids has also drastically reduced the need for sheer physical effort. The major development of the internet has also brought multiple opportunities