Chloe Harris prefers reading to sport and would rather travel to school by car than walk. Sean Bowden does at least 13 hours of exercise a week, including basketball, football, swimming and PE, and would take on more if his parents let him. Yet analysis indicates that Sean does just one minute more moderate or vigorous physical activity a week than Chloe.

These two 11-year-olds from Plymouth — and 300 others like them — may force us to rethink our ideas on children and exercise, as well as on the origins of obesity.

For the past seven years the EarlyBird Diabetes Study, based at Derriford Hospital, Plymouth, has monitored the children’s activity levels and health in an attempt to track the childhood roots of diabetes. It has found, to almost universal astonishment, that children’s activity levels are governed not by the number of PE lessons in the school time-table, or even by the sport they do in their own time, but by an internal mechanism that may be preset before birth. In other words, how much energy children expend may be determined by their genes.

“If Chloe can get out of PE she will, and if I suggest that we go for a walk she will always say that she would rather read her book,” says her mother, Sue Frost, a paediatric nurse. “But she’s forever dashing up and down the stairs, running after her baby brother or dancing to music.”

Sean Bowden, meanwhile, “will do any kind of activity”, says his mother Karen, a local government officer. “Sometimes he wants to do too much. He used to play tennis twice a week, but when he got into the basketball team and needed to go to practice, I told him that he could do one or the other. We also said no to judo: he needs to go to bed reasonably early because he gets tired from all the sport.”

Terence Wilkin, professor of endocrinology and metabolism at Peninsula Medical School and director of the EarlyBird programme, says: “Our research has led us to suspect that it is not environment that determines activity but some central biological mechanism within each child that we might call an ‘activitystat’.

“Just as a central heating thermostat maintains a preset temperature and switches off the boiler when that temperature is reached, so the amount of activity of each child is set — probably genetically in the hypothalamus — and the control system ensures that the child’s activity meets that setting.”

The EarlyBird work is carried out on children from a broad social spectrum, monitoring their activity levels and metabolism. If the research continues to receive funding — now in doubt — it will follow them until the age of 16, providing an insight into the causes and origins of diabetes.

The activitystat hypothesis emerged after trials suggested that no matter how much or how little exercise children were offered, they found their own level. “Like horses brought to water,” says Professor Wilkin, “children with low-activity settings may simply not participate.”

One study monitored a week of physical activity during waking hours among 215 children aged 7 to 10 at three schools with different sports facilities and timetables. A private preparatory school with playing fields offered nine hours of PE a week; a village school 2.2 hours; and an inner-city school with a small playground 1.8 hours.

All the children were fitted with accelerometers, tiny electronic boxes that sample activity 600 times a minute and record every movement. Professor Wilkin says: “As expected, pupils in the first school recorded the most activity in school time. Yet the total physical activity between the schools was similar because those who had little provision in school compensated with large amounts of activity when they got home.

“There was no relationship between the school attended and the amount of activity undertaken. Clearly, the total amount of activity done by primary school children does not depend on how much PE they do at school.”

The activitystat hypothesis effectively slays some sacred cows. For example: girls, constantly berated for their reluctance to exercise, may take less activity not for social reasons but because their activitystat settings are lower.

“There is a wide range of activity among children and those who do more seem generally healthier,” says Professor Wilkin. “The big question is whether it is possible to get inactive children to do more. So far the evidence is bleak.”

Not everyone agrees. Andy Ness, professor of epidemiology at the University of Bristol, says: “There are clearly balance mechanisms that stop us running until we drop dead, but the debate is about the extent to which these hard-wired mechanisms control the exercise we do.

“I think there are some people who are naturally active and some who aren’t, but there is a bit in all of us that is modifiable and that is the part that we can deal with and change, though it may be harder than we would like to think. I do think the environment affects the amount of activity we do and perhaps the activitystat can be overridden or rewired through learnt behaviour as children move into adulthood.”

If, as Professor Wilkin believes, the exercise that children do is not influenced by opportunity, then what determines it? “When habitually inactive mice are forced to do more wheel-running, they rest and even reduce their metabolic rate to compensate during the rest of the day. By contrast, if you stop the active mice from running on the wheel too early, they will race round the cage until they have reached their ‘setting’.”

The idea that genes govern our exercise behaviour is not new — a 1971 study of twins indicated that fitness is an inherited trait — but it remains unpopular. Dr Stephen Phinney, a metabolic specialist at the University of California, says: “We are prone to resist evidence indicating that we are not all fundamentally the same in choosing what we eat, how much activity we do and to whom we are attracted.”

The EarlyBird research suggests not only that it may be impossible to make children exercise more, but that it may also be pointless fighting obesity. “We have looked at the relationship between fatness and activity and when analysed over time — which only a study such as EarlyBird can do — the data suggests that activity responds to weight rather than weight to activity. In other words, heavier people exercise less,” says Professor Wilkin.

Separate research, carried out on 545 Glasgow nursery school children and published in the BMJ last November, supports the EarlyBird conclusion that increasing physical activity has no effect on BMI.

“We are not saying that children should give up exercise, which has many benefits, social and physical,” insists Professor Wilkin. “For example, those who are more active have better metabolic health — their blood pressure is lower, among other things — but they are not necessarily slimmer.”

His contention appears to be at odds with the results of a study published last month using data from the Children of the 90s project. This research, involving 5,500 children from the Avon Longitudinal Study of Parents and Children (Alspac), suggested that an extra 15 minutes a day of moderate or vigorous physical activity halved the risk of obesity in 12-year-old children.

“Most studies, including ours and a big European study, find an association between obesity and activity,” says Andy Ness, co-director of Alspac. “Which way the association works is less clear. We cannot rule out the possibility that obesity leads to a reduction in physical activity rather than the other way round, but the fact that these associations were observed across the range of fat mass rather than just in obese children makes this explanation less likely.

“But even if the associations are due to reverse causality and obesity leads to reduced activity, this is an important observation as reduced physical activity in obese people may be harmful to health.”

The true picture may never emerge because the EarlyBird Study is running out of funds and the Government, which has helped to finance the project in the past, shows no sign of doing so again.

Caroline Flint, Minister for Public Health, has said: “The department has never made a commitment to long-term funding of the EarlyBird study into the link between childhood obesity and diabetes.”

Richard Morgan, chairman of the EarlyBird Diabetes Trust, responds with regret: “EarlyBird is the only longitudinal study to trace the issue of diet, exercise and metabolic health all the way from childhood to adolescence and this is the only way to answer these vexing questions.”

Fat is a natural state. . . for some people

For an issue that is at heart so simple, obesity is beginning to look very complicated. We know that consuming more calories than we expend makes us fatter, and doing the opposite makes us lose weight. That’s hardly Wittgenstein. We also know that the appetite is a sophisticated mechanism brilliantly calibrated to keep intake and output in balance. Yet more and more people are ballooning, for reasons that aren’t wholly obvious.

The role of exercise is particularly paradoxical. Everybody knows (a phrase that generally precedes a well-rehearsed prejudice) that children are taking less exercise.

Yet if Professor Terence Wilkin is right, our propensity to exercise is hard-wired. Denied sports opportunities at school, we will find them elsewhere, if we happen to be exercise-prone. And if we are exercise-averse, we will make sure we are excused games whenever possible.

What is new in today’s “obesogenic” environment is that we have options, and they cut both ways. We can choose to eat too much, or to exercise too little, or both at once. As our options are widened, so it becomes easier for our inherited characteristics to express themselves.

We might, for example, have inherited a gene that enables us to store fat especially well. Or we might have a gene that predisposes us towards slumping in a couch.

Like abundant food, the slumping option is now available to millions. By being true to their natures, they become tubby, then gross, then impossible to contemplate with equanimity.

The message seems to be that we are not designed to remain thin in today’s world. Fat is the natural state, at least for some. They will remain thin only by denying themselves the instincts conferred on them by their genes.

There are lessons in this. We must either stop being censorious about the overweight, or recognise that only compulsion will change their life-course.

The alternatives are recognising fat people as fully-formed human beings who just happen to be an awful lot bigger, or going back to compulsory PT and teaching nutrition properly from an early age. Personally I’d go for the latter — but that’s another prejudice.

Nigel Hawkes, Health Editor