Fat, Fate, and Disease : Why we are losing the war against obesity and chronic disease (19 page)

In the English aristocracy, life expectancy was greater for women who had fewer children and lower for those who had more. It was also high for those who had their first child at a later age. So having more children (and starting to have them earlier) is associated with a shorter life. Of course, these data relate to only one family, but the trends are in the direction predicted by evolutionary biology.

Nobody would claim that the lives of the English aristocracy are representative of human existence across the globe, or even in the UK itself, especially over a historical timescale. More recent attempts to address this problem have come from the work of the evolutionary psychologist Daniel Nettle, who also works in Newcastle. Nettle used data from the Millennium Cohort Study which comprised 8,660 families in the UK, in which social and other factors were measured across the population, from the most deprived to the most affluent neighbourhoods. He was able to show that, compared with affluent families, in more deprived neighbourhoods the age at first birth is younger, birth weights are lower, and breastfeeding duration is shorter. He also found some evidence that reproductive rates are higher—giving larger family size. He interprets his work as a modern-day human equivalent of the trade-offs we discussed for animals. But here the challenge is poverty and poorer social circumstances leading to earlier, and perhaps more frequent, reproduction.

The broader social consequences of such processes, if they can be shown to operate more widely across society, are extremely important. The children of younger mothers, especially in low-income settings, are less likely to do as well at school or to have the same level of lifetime earnings as those of the older mothers of affluent families. They are also less likely to have their father resident in the home or even to have as much contact with their maternal grandmothers. These effects have the potential to be extremely costly for society in humanitarian as well as financial terms, and the fact that they still operate in the UK today is very worrying. Poverty and social deprivation beget poverty and social deprivation, and breaking this cycle is extremely difficult, as every recent UK government has experienced. To a degree every recent UK government has also failed to meet this challenge because the gap between the rich and the poor, in terms of access to education, healthcare, and financial stability, is greater today than it was when the welfare state was founded just after the Second World War.

Dan Nettle’s provocative study does not give us any insight into the underlying unconscious mechanisms which are operating, and in some respects this makes it quite difficult to predict which individuals or population groups will be at particular risk. And the data are subject to alternative interpretations. This is the difficulty of such research—studying humans is much more complex than studying experimental animals where the conditions can be controlled. But on the other hand, humans are animals too, and if we can obtain strong data in other species and suggestive data in humans then we can be moderately certain that the conclusions reached apply to us too.

If life history trade-offs are operative in modern Western societies, then we would expect to see that girls who had a poorer start to life, and thus whose biology was tuned by prediction of a shorter lifespan, would be younger at menarche. Furthermore, as there is no reproductive value in reaching menarche early unless one is in a good enough physical condition to support an early pregnancy, we believe that the earliest age at menarche would be seen in those who were
born smallest—that is, had a poor start to life—and then developed relative fatness in childhood. We developed this concept from a series of studies made in rats and from trying to make sense of a rather disparate group of studies in humans. But more recently this concept was validated in a study in Western Australia led by John Newnham. Deborah Sloboda and her fellow researchers divided a group of almost 1,000 girls into subgroups based on their weight at birth and their weight at seven years of age. They found that the earliest menarche occurred in the girls of lowest birth weight and greatest weight at seven years. The latest menarche was in the opposite subgroup—girls of highest birth weight and lowest childhood weight gain. The difference in the timing of menarche between these two extremes was more than a year, which is a very substantial effect in evolutionary—and in modern social—terms when looking across a population.

Imagine a class of 11-year-olds just starting secondary school, where perhaps a third of the girls have already started to have regular monthly periods, but another third show no signs of menarche at all. Their relationships and reactions to the boys in the class will be substantially different, as of course might be the consequences. Needless to say, many studies have confirmed that girls who have menarche earlier are engaged in sexual activities at a younger age. Other studies from Europe have shown that both boys and girls who have earlier puberty are at greater risk of behavioural and mental health problems: in girls, eating disorders, and in boys, even suicide. These are serious matters, and while they are not all related to puberty and sexual activity, the fact that a high proportion of them are raises concerns.

There is something else which is reflected in the findings from Western Australia: size at birth is not everything—just being bigger is not
necessarily better than being smaller. As we will soon discuss in detail, many population studies around the world have shown that the risk of cardiovascular disease and diabetes is greater in people who were smaller than average at birth. Equally the risk of some forms of cancer in later life is greater in those who were larger babies at birth. And, as we will discuss in the next chapters, babies who are very big are likely to have mothers who had diabetes in pregnancy, and this in turn puts them at greater risk of developing diabetes themselves. Developmental processes act as a continuum. Even if we take a simple measure of development such as weight at birth, we can see that there is a spectrum of consequences which follow. The Western Australia study shows that this is so even across the normal birth weight range—a point of major importance.

Add to this idea the fact that life itself is a continuum, and we can see that decisions and strategies established at one point in life, say in prenatal development, can have consequences a long way down the track. The types of chronic disease which develop, their nature, and the time in life at which they occur vary enormously between individuals. We all travel on different trajectories through our lives and what happens to us at one point in time, from our early development onwards, will have consequences much later.

This idea does not just apply to developed, high-income societies. One of the best examples, which demonstrates this concept dramatically, is the study from The Gambia which we discussed at the start of this chapter. The seasonal effect on mortality does not become evident until after the time of reproduction, when the trade-off produced by a poor start to life—developing and being born in the hungry season—leads to greater mortality later.

So what happens to us at the very start of life can impact on our destiny through the rest of our lives. These effects are not matters of extremes—of disruptions to our developmental programme such as those induced by measles infection or poisons during pregnancy. Rather, they are the echo of our evolutionary past, and utilize the
processes of developmental plasticity, whereby every developing organism tries to take information from its developmental environment to predict its future, and to prepare itself accordingly. Now we need to explore what happens when a fetus predicts a good environment and what happens when it predicts a bad environment—for within this may lie the clues to the missing factor that explains why some of us get fat more easily, and some of us suffer from diabetes and cardiovascular disease, while others do not.

For too long, the role of development in biomedical science and clinical medicine has been played down. For many years scientific research on development has been seen as the poor cousin, or perhaps we should say the poor son or daughter, of studies of the biology of adult organisms, including humans. One of the authors remembers overhearing a remark made by a very eminent professor during a break in a scientific meeting. It coincided with the time at which the author was deciding to switch to research in developmental physiology himself. His illustrious senior must have been discussing with a colleague a paper or a research funding application which they had both been sent to review. ‘Oh well, of course you don’t have to be very clever to work on development,’ he said. ‘After all, it’s much simpler than adult biology.’ The other author was told by his distinguished Professor of Medicine when he decided to be a paediatrician that he was
throwing his career as a medical scientist away—‘Nothing fundamental or important ever comes out of paediatrics,’ he said.

Such disparaging remarks were familiar to those of us who have built careers in developmental physiology and medicine—the dominant idea was that development was merely a process of assembly of simple components, uninteresting in themselves, which merited the attention of the greatest intellects in science only when they were put together to form the adult organism. Bizarre as it might seem now, babies were seen as simpler versions of adults—and so the medical problems of children were regarded as simpler than the medical problems of adults.

To those biomedical scientists who insisted on looking at development through the wrong end of the telescope, it has come as an enormous surprise that there are highly complex processes operating during this period of our lives, some of which are unique to development itself. They were shocked to discover that the fetus has a good deal of control over its destiny and even the very processes of its development, while it is undergoing them. Development is not like a car being assembled on a production line from bins of identical parts, which has no role or function until the assembly is complete and it can be driven.

Even more surprising to those solely focused on the adult has been the revelation that many developmental processes play a critical role in evolution. Not only have they evolved to fulfil specific fitness functions, but they also play a major part in influencing Darwinian fitness itself. This realization has caused a major revolution in evolutionary thought.

We are relieved that such blinkered views are now very rare, and developmental biology is recognized as one of the most exciting areas of modern science.

We have now come to the point in our story where we need to be specific about the ways in which early human development contributes to the increased risk of obesity and the chronic diseases.
In view of what we have just said, it should not be a surprise that this developmental aspect too was ignored for so long. Is it still being ignored? Let’s look at the evidence and then return to that question.

There are really two lines of evidence which have demonstrated beyond a shadow of a doubt that early life processes play a major role in influencing obesity and non-communicable disease risk. Let us consider studies of human populations first. The first studies were made in England in the 1930s, and suggested that children who had poor early development had a high later risk of heart attack. The significance of these observations seems to have been overlooked or forgotten. Then in the 1980s three groups, two in England and one in Sweden, led by Michael Wadsworth, David Barker, and Gerhard Gennser, respectively, reported at about the same time that children who were born at the small end of the normal range had a higher risk of developing high blood pressure as adults.

While the other two groups made the observation and passed on, David Barker, a Southampton physician and epidemiologist, was not content with just making the observation only once. He conducted a series of compelling observations in many communities and started on an important mission: trying to persuade the medical community that the problems their patients faced started before they were born. His observations were dismissed as flukes, as confused by uncontrolled factors, and as scientifically implausible. How could factors operating before we were born influence our risk of disease 50–60 years later? However, gradually more and more population scientists began to confirm Barker’s findings. They extended the observations from cardiovascular disease to diabetes and then to insulin resistance in children, an early sign of adult diabetes.

Even after these concepts had been accepted they were still considered to be of relatively minor importance. It was not clear how a pathway to disease could start with low birth weight. And in any case the fraction of infants in any population born small was low, whereas diabetes and heart disease were becoming so common that in some societies up to 50 per cent of the population suffered from them.

These criticisms were based on fundamental misunderstandings which, it seems to us, would have been quickly addressed in any other field. These epidemiological observations were, in time, clarified so that they made no claim for a link between low birth weight itself and later disease—in fact, what they showed was that processes operating
before
birth were linked to later disease risk. Birth weight is a marker of the prenatal environment, made at the earliest time when it is possible to obtain some direct measurements of the developing baby. Even more important is the fact that the observations did not only concern low birth weight babies. The original observations, along with those of many other studies conducted by scientists in many countries, some of which involved over 100,000 subjects, showed that the risk of later disease was graded across the entire range of normal birth weights. For example, in Barker’s studies of men and women born before the Second World War, someone who had been a perfectly normal baby weighing 7.5 pounds at birth was at greater risk than someone who had been an 8-pound baby, whose risk in turn was less than that of an 8.5-pound baby.