Keeping the doctor away

A survey from the 1930s could reveal important links between diet and disease, says Steve Farrar

The cardboard boxes, crammed with yellowing papers, looked anything but precious. Yet when the Bristol University scientists glimpsed their contents for the first time, they knew they had found something of immense importance.

"Pure gold" is how one of the team described the information packed onto the flimsy sheets.

Here was a remarkable time capsule, a snapshot of the lives of 4,973 British children in the 1930s. The Carnegie survey of diet, poverty and physical wellbeing in prewar Britain was meticulous and detailed, the most thorough assessment yet made of the state of a nation's health.

Yet the Bristol team were not social historians. They were epidemiologists who recognised that these documents, which had sat for decades in the Rowett Institute near Aberdeen, could prove to be a very powerful tool.

The youngsters whose childhoods were recorded therein were now reaching the end of their lives. If the scientists could track them down today, they would be able to match the background of a child to the fate of an adult and do so in sufficient numbers to start answering some fundamental questions about the influence of our childhood diet on later life.

They could quite literally watch our past catching up with us. Other large scale surveys have been carried out in the past, but none could look back to the 1930s in this way.

Concern that poor diet among the most impoverished was having a serious impact on the health and development of some children prompted the Carnegie UK Trust to fund Sir John Boyd Orr and a team of health experts to travel the length and breadth of the country between 1937 and 1939.

They recorded what each family in 16 different neighbourhoods, from Dundee to Bethnal Green, ate in the course of a week, checking individual larders as well as quizzing the head of the household. They physically examined every child's health and painstakingly set the details down on paper.

The outbreak of conflict in Europe shortly after the work was completed delayed its publication until 1955, although it is believed to have played a role in informing legislators drawing up recommendations for the wartime food policy.

Recovering the original data from the Rowett was just the start for the Bristol team, who then had to set about tracking down and contacting the children six decades later.

Ten years on, the scientists have found 3,500 of those children - 1,000 of those surveyed had died in the intervening years and 500 could not be traced.

The survivors have been written to and the fates of the deceased checked with National Health Service records. The experts then started to put the two sets of data together and watched surprising patterns emerge.

The work has attracted funds from the Medical Research Council, the British Heart Foundation and the World Cancer Research Fund, and a stream of scientific papers has followed.

David Gunnell, senior lecturer in epidemiology and public health at Bristol, who is leading the work, said that a decade after this research began there is a great deal of valuable information yet to come out. "It's a gold mine of data. By trying to map out what has happened to the children since the survey, we have already been able to find associations between nutritional status in childhood and both cardiovascular and cancer mortality in adulthood," he said.

"This is evidence that childhood is an important time in terms of developing risk of disease in later life." The analysis has shown that childhood height was linked to diet and living conditions, the poorest children having their growth stunted by their poverty. The links between this and later health, however, are not always obvious.

Those youngsters who were comparatively short or overweight appear to be at greater risk of dying of coronary heart disease in later life - even if they were slim as adults. "It is a consistent link, and this is quite worrying, especially given that the proportion of overweight children is increasing nowadays," Dr Gunnell said.

Conversely, taller children, particularly those with long legs, appear to suffer a significantly greater risk of cancer in later life, a fact that may be linked to increased levels of growth factors in the larger youths.

Nevertheless, it seems taller is best as the death rate from heart disease outweighs that from cancer.

To take the work further, the team has just completed a postal questionnaire survey of current health and diet among surviving study members. They also plan to approach a sample of them for detailed clinical examination soon.

"We hope this will allow us to explore the biological mechanisms underlying these epidemiological associations," Dr Gunnell said.

This should reveal new facts about the influences on childhood development and subsequent impact. It will also allow the team to make practical recommendations as to diets that will make not only the children healthy but also the later adult as well.