Good as the practical reasons for trusting in early intervention might be and sound the grounds for saying that problems emerging during infancy often hindered later progress. But in their efforts to promote the benefits, advocates of early years services had exaggerated the stability of pre-school impairments, Professor Jay Belsky told Sieff Foundation conference delegates

“It’s true that the period from conception to reception is an important and influential time, but it’s misleading to talk of a critical period,” he said.

Claims about brain development had also been adrift, he said. As the highly influential review of the evidence in From Neurons to Neighborhoods demonstrated, it was true that early experiences influenced brain development, but research did not support the disproportionate focus on first three years of life.

This lens was too narrow; it was as important to examine what happened before and afterwards. “Just because we can now see the brain function thanks to MRI does not change this situation.”

Belsky went on to give stern warnings about unhelpful claims as to the effectiveness of early years programs. “There’s no single way to do early intervention well”, he said. “It happens that some people advocate center-based programs, others high quality child care. Some claim that family involvement is crucial, others that the most progress can be made by keeping families at bay.”

In reality, all of these approaches, if well designed and well delivered, could be effective. And all tended to show similar effects on children’s health and development. Typical effect sizes for cognitive and social emotional development were a modest 0.3.

There were dangers, he warned, that the same problems would occur among early years programs being introduced to the UK.

Family Nurse Partnership, the UK variant of David Olds’s Nurse Family Partnership. “It works in the US”, he acknowledged, “but will the results be replicated here? And will we know which sub-populations it works with? And will we know how to get the program to those sub-groups in the future?”

Belsky is well placed to comment on the successes and failures of Sure Start, UK’s flagship prevention program. He helped to lead the national evaluation. The early results from the quasi-experimental evaluation were disappointing, reflecting, he suggested, government misjudging practitioners’ willingness to embrace a menu of manualized programs.  

But these negative results had now been published and acted upon. He drew a comparison with the first reports on the US Head Start program which were also negative but did not emerge for many years. In the UK, the early results were used to re-assimilate the approach in the new Children's Centre provision.

As an alternative to zealously promoting a single approach, Belsky encouraged the audience of leading policy makers and practitioners to consider what they wanted to achieve. Different prevention strategies produced different results. Some programs were better at improving cognition, reading and math scores; others majored on social and emotional development. Claims to do both were probably unrealistic, as were policy makers’ expectations of dual impact. 

Thinking about which children to target was another challenge. Treatment effects for some interventions seemed to be enhanced when the focus was on higher risk child, such as those lacking executive function. Emerging evidence on the way genetic predisposition could interact with an intervention to alter the effects upon sub populations further complicated this task. 

“Distinguishing between efficacy and effectiveness trials should be high on the agenda of research funders. We do not know at present how to maintain the same kinds of effects generated in the design phase when proven models are taken to scale,” he said.

“We do know that the quality of program delivery is key. Ensuring fidelity to a well specified model will make the difference between success and failure. But we do not yet know how to do this in all contexts.”