Statistics from Altmetric.com
“Joined up thinking” and “joined up government” are now familiar phrases. Joined up real life is rarer, with services for children and families frequently fragmented. Sure Start is therefore particularly encouraging as an initiative which will potentially pull together health, education, and welfare services for 0–3 year olds in a coordinated way.
Sure Start is part of the current government's policy to prevent social exclusion, and as such, it is targeted at preschool children and their families in disadvantaged areas. The initiative was the result of a cross cutting review of services for young children chaired by the Treasury.1 The review's conclusions focused on the importance of the early years for child development, and highlighted the problems of multiple disadvantage for young children, the variation in quality of services for children and families, and the need for community based programmes of early intervention.
The cash behind Sure Start is substantial: £540m is to be spent between 1999 and 2002 of which £452m will be spent in England. The policy is being developed in different ways in Scotland, Wales, and Northern Ireland.2 Starting with 60 “trailblazer” schemes selected on the basis of deprivation, geographical spread, and links with other government initiatives to tackle deprivation, there will eventually be 250 local programmes covering up to 150 000 children.3
The range of activities included in Sure Start
The aim of Sure Start is to work with parents and preschool children to promote the physical, intellectual, social, and emotional development of children—particularly those who are disadvantaged. The idea is to ensure that the children who have been in Sure Start programmes are ready to thrive when they get to school.4
In each area where there is a Sure Start project, locally based programmes are encouraged to build on what already exists to ensure a range of core services including:
Outreach services and home visiting
Support for families and parents
Good quality play, learning, and child care
Primary and community healthcare and advice about family health and child development
Support for those with special needs.
In addition, local communities may provide extra services according to local needs, such as skills training for parents, personal development courses, and practical advice and support such as debt counselling, and language or literacy training.2
Many of the attempts to alleviate harm for children in the past, as an independent report commissioned by the Department of Health has pointed out, were based on a moral confidence, and a belief in the extreme pliability of children once exposed to new environments. “If what was being done in the interests of the child was self evidently right, the question of whether it actually led to desirable outcomes was hardly likely to be asked”.5 Until recently, a keen interest in the outcomes of our well meaning interventions has not characterised services in the early years in the UK.6
Sure Start programmes, in common with a number of other government interventions, are highly—some feel too—outcome focused. There are a set of objectives that relate to children's social and emotional development, health, and ability to learn, and the strengthening of communities. The objectives are linked to targets: some rather “hard” measures, such as a 5% reduction in low birth weight babies; some more process driven, such as parenting support and information available for all parents. Delivering good outcomes will be a tall order when we know that modest interventions normally have modest effects, and although the funds are considerable in relation to past funding for early years work, the gains currently hoped for are substantial.
The objectives are:
To improve social and emotional development
To improve health
To improve the ability to learn
To strengthen families and communities
To increase productivity of operations.
Some of the targets are described above. The targets for the final, rather mysteriously described objective are to make sure that 250 programmes get off the ground in England, and 100% of families in the areas targeted are in contact with Sure Start within two months of a baby's birth. The final target in the operational objective is to have an evaluation strategy in place by 2000–01.
This last is key if we are to have an understanding of whether Sure Start is achieving its goals, and will be a considerable task. As well as looking at outcomes, processes need exploration. What actually works in getting projects up and running, and parents and children engaged? In order to explore what the evaluation might look like, a development project coordinated by the Centre for Longitudinal Studies at the Institute of Education, is in progress.7
Why might we expect Sure Start to have beneficial effects?
Sure Start is unusual as a UK policy initiative in taking as a starting point robustly evaluated interventions.1 The guidance for the first wave of projects suggests that: “Decisions about the services provided ... should be based on existing best practice and, where approaches have been rigorously evaluated, on what works in promoting the development of young children”.3
An extensive literature links early disadvantage with later effects on health and wellbeing.8 9 The cohort studies indicate that parental interest in, and enthusiasm for, education, offer the best protection from disadvantage in the long term.10 11Children fortunate enough to have this help tend strongly to do better in educational attainment12 and in due course, such children as adults are more likely to be enthusiastic about their own children's education.13 Those who go on to gain qualifications have much better chances in health14 as well as in occupation and income.10 The most important protective factors for children in terms of health appear to be those which optimise growth and development before birth and in early childhood.15 All of this suggests that early investment “works” both for current wellbeing and later outcomes.16 17
The UK has a less distinguished record than North America of robustly evaluated interventions or policies. There is some evidence, however, that components of Sure Start are likely to be beneficial.
In the prenatal period for instance,18 a randomised controlled trial directed at women at risk of a low birth weight baby has indicated the effectiveness, appropriateness, and safety of a social support intervention provided by midwives. Interestingly, as follow up has continued, differences between the intervention and control groups have been maintained. At seven years, there were fewer behavioural problems among the children and less anxiety among the mothers in the intervention group.19
In the immediate postnatal period, an Edinburgh study, in which women identified as depressed were randomly allocated either to counselling by health visitors or to standard treatment, found that health visitor counselling is helpful in managing non-psychotic depression.20
A promising social support intervention is the Child Development Programme whose fundamental goal is to help and encourage parents.21 This programme offers monthly support visits to new mothers, antenatally and for the first year of life. Most of the visits are undertaken by health visitors. A radical development of this was the Community Mothers Programme in which mothers were recruited to provide support. A randomised controlled trial showed that children in the intervention group were more likely to receive all of their primary immunisations, and to be read to daily. They were less likely to begin cows' milk before 26 weeks. Mothers as well as children in the intervention group had a better diet than the controls. At the end of the study, intervention mothers were less likely to be tired or feel miserable.22
Perhaps the best known evaluation ofpreschool interventions is Highscope. This involves an active learning curriculum, trained staff, and parent participation. The UK has a particularly strong history in preschool work and Highscope shares many of the elements of other good quality preschool interventions. A randomised controlled trial with long term follow up indicates significant benefits.23
A further component of Sure Start is preventive work and family support. A number of reviews24-29have explored the effectiveness of interventions designed either to protect children and/or provide support to parents. Home visiting has been shown to reduce negative outcomes for parents, children, and families.30 31 The Acheson report32concluded that while the research evidence suggests parent support is helpful, our understanding of the nature of that support, the point in time, and for what length of time this is helpful remain unresolved. The strongest research evidence comes from a 15 year follow up of a randomised controlled study of directive home visiting support for socially disadvantaged mothers. This found significant effects on maternal and child functioning, including child abuse and neglect, maternal welfare dependence, adult and child alcohol and drug misuse problems, and child antisocial behaviour and criminality.33
Community based interventions are central to Sure Start. While community development has been less susceptible to robust evaluation than some of the other aspects of Sure Start, a review provides some pointers towards those issues which appear promising in relation to interventions relating to children and families.34
Lessons to be learnt from large scale evaluations elsewhere
The majority of robustly evaluated early childhood initiatives were launched in the USA.35 In the 1960s and 1970s at least 30 separate federal educational and training programmes for low income populations in America were funded.36 A major objective was to increase the basic cognitive skills of disadvantaged children, so there was a good deal of emphasis on measured IQ as an outcome. However, some programmes took a broader view of the positive outcomes which might be encouraged by early intervention.
Many other large scale evaluations of social programmes were launched in the USA, covering such fields as income maintenance, and training and employment initiatives for socially disadvantaged groups.37 These raised a number of issues likely to be pertinent in understanding whether Sure Start delivers the expected outcomes:
The need to take into account the complexity of social settings
Involving the “targets” of intervention in designing appropriate initiatives and outcomes
Distinguishing between individuals, families, households, and communities as targets of intervention
The importance of a multidisciplinary approach to evaluation.
Design problems characteristic of the early intervention field38-40 included:
Lack of standardisation of the intervention, leading to unmanageable and unanalysable variability between sites
Narrow outcome measures
Lack of comparability between intervention and control groups, leading to probable underestimates of social programme effectiveness.
Will Sure Start work ?
There is a good deal of evidence to suggest that intervening in the early years is key to sound later development. If intervening in this period is so powerful, then we need to be as sure as we can be that there is convincing evidence that interventions work, and where there is not, unproven interventions will be scrutinised through good evaluations.
There are aspects of the programme which will need further careful thought if it develops beyond its current period of operation. Not all disadvantaged children live in communities which are identified as poor, so more work will be needed on the relative effectiveness of targeted and universal benefits.
The whole question of whether at a local level, the best of what works will be used in providing services is another issue which will need to be addressed. Just as evidence based medicine has its enthusiasts and detractors, evidence based social care is not universally acclaimed.
Crucially, it will be important to focus on the immediate lived experiences of children receiving interventions. The enjoyment, the challenges, and the fun now are important in their own right as well as longer term outcomes. Early childhood is not just a period in training for schooling and adulthood.
A programme such as Sure Start which both draws on good early interventions, and which proposes a first class evaluation, is a step in the right direction.
I am grateful to a number of colleagues with whom I have discussed Sure Start, in particular Ann Oakley, Trevor Sheldon, Ian Roberts and David Gough.
Sure Start is an exciting experiment in child public health, but three important issues have yet to be fully addressed. The first is the extent to which the content of the Sure Start intervention should be standardised. The importance of local community “ownership” is rightly stressed, but as a result trailblazer projects seem to differ widely in content. We now know a good deal about what might work with regard to social support, parents' mental health and child rearing styles, language acquisition and pre-reading skills, and behavioural management. Perhaps the government's Sure Start team should be rather more prescriptive about the essential elements of the individual projects.
The second issue concerns outcome measures. The primary aims of Sure Start are educational—“enabling children to benefit from their schooling”. A rigorous experimental evaluation is needed in a small sample of the projects, but will not be possible for the whole programme. If Sure Start is achieving its goals, teachers should be able to see the difference in key baseline skills, such as ability to communicate and readiness to learn. The success of each project might therefore be measured by an improvement in the average performance of children starting school, or by a fall in the number of children entering school with baseline skills that are so poor as to need intensive additional teaching support. The projects are focussed on very small individual communities and there are no direct comparison groups, so “medical” outcomes (such as “reduction in the proportion of low birth weight babies”) are very unlikely ever to approach statistical significance—indeed random chance might easily drive them in the opposite direction.
The third concern is about generalisation. The investment in Sure Start is very substantial. The author's figures suggest that over the three years of the project £3000 will be spent on each child. By no means all the children potentially able to benefit from intensive early intervention will be included. Is this government (or the next) prepared to sustain that level of funding and perhaps even extend it to cover all children living in deprived circumstances?
It would be very sad if a fundamentally sound and potentially evidence-based programme were to founder because of insufficient quality controls on content, failure to define appropriate outcome measures, or uncertainty about which elements of the programme give best value for money. We hope sound evaluation proposals will emerge and be funded very soon.
I am grateful to Linda Fox, health visitor in Sheffield; P Hannon, Professor of Education, University of Sheffield; and J Law, reader in speech sciences, City University, London, for helpful discussions about Sure Start.