Celebrating the National Survey of Health and Development: 1946-2026
Eighty years ago this month, the United Kingdom pioneered a novel form of social science research, the life-long cohort study. The tool proved so powerful that it has been copied internationally, and repeated every decade or so in the UK. It grew out of discussions about fertility decline that had been taking place in many developed countries during the 1930s. Why were women having fewer babies, and what were the likely societal consequences?
The study was launched by a grant from the Nuffield Foundation to the Population Investigation Committee, an ad hoc group based at the London School of Economics since 1936. Its secretary was David Glass, a demographer, who recruited James Douglas, a disenchanted doctor, to lead the project. Some 15,130 babies were born in the week of 3-9 March of 1946, and the study attempted to interview every mother. In order to achieve this, they got co-operation from almost every public health department (424/458) in the country, making use of their health visitors to interview mothers, using a standard questionnaire, between April and June 1946.
Two things now stand out. The first is the speed and relative informality of the launch. Douglas wrote to public health departments in February 1946. They agreed (or not), and the survey was launched. Any comparable project today would be bogged down for months getting ethics and other clearances. Nevertheless, the research team has been obsessive over multiple generations in protecting the anonymity and confidentiality of the participants.
The second is the availability of health visitors (visiting nurses) to carry out the work. They had played a crucial role as the foot soldiers of public health for women and children since the early 20th century. Their work in World War II had been particularly valued, in helping to promote the new diphtheria vaccine and in managing the consequences of vaccination and bombing. Virtually every new mother in the country, except perhaps the highest income groups, had contact with them and they were uniquely trusted as sources of advice and support. Their efforts achieved a 91 percent response rate (13,687). Like many public health services, successful health visiting reduces problems, which then becomes an excuse for reducing investments, without recognizing that the consequent re-emergence of problems will be far more costly. There is now generally considered to be a crisis in health visitor provision, which is likely to be contributing to declining vaccination rates for childhood infections. The survey work simply could not be replicated today.
From the start, the study reflected the combination of social and medical influences in a way that contributed to its enduring value. Early analyses pointed to the strong influence of social class on child health and development. Would this change as the children grew older? Were there any factors that promoted resilience or recovery from initial disadvantage? Even as the first publications were being prepared, Douglas and Glass were looking to follow-up studies. To reduce costs, the original sample was reduced to 5,362 members, by eliminating children born to unmarried mothers (who were mostly adopted and untraceable), multiple births and including one in four of those born into households headed by manual workers.
As Michael Wadsworth, the second director, pointed out, it is important to remember that the experiences documented by the study occurred in a context of rapid social and economic change, which necessarily affects the interpretation of the findings. Being a single parent in 2026 is a very different experience from being a single parent in 1946. Industrial manual workers are a much smaller proportion of the workforce and reflect a way of life that has largely vanished with the decline of coal, steel and similar industries in the 1980s.
Douglas eventually ran the study for 33 years, following the cohort into adult life. It has retained more than 2,000 participants and added new types of data, including genetic testing and neuroimaging, as technologies have evolved. Current studies are looking at brain health and possible associations with early and mid-life experiences. The initial focus on social and educational data collection has, for example, pointed to a potential link between childhood cognitive development and the likelihood of Alzheimer’s developing by the mid-70s. Data on exercise may help with the current debates about contact sports and long-term neurological conditions.
The National Survey of Health and Development, to give it the full current title, has proved a unique resource for health and social policy. In its early years, funding was often precarious but its contribution has more than justified the vision of the founders. Remarkably, it has had just four directors in its lifetime, which has undoubtedly contributed to the preservation of organizational memory. It is now accommodated alongside the 1958, 1970 and Millenium cohorts, with a 2026 cohort being recruited. The original punched cards have given way to a fully computerized archive, available for wider use under appropriate regulatory conditions. The website also includes the birthday cards that have been sent out every year since 1962 as a way for the team to keep in touch with the participants. It has been a great adventure for everyone – and there is still another wave and a birthday party to come!
