Abstract
First of all, I would like to pay tribute to Miranda Davies who retired in March after 27 years as our Journal Manager. Her publishing acumen, sensitivity to readers’ needs and intellectual insights were major factors in maintaining the standard of the journal and hundreds of people benefitted from her help and advice. She has been replaced by Victoria Walker who brings to the journal considerable experience of producing research publications.
My time as Commissioning Editor is also coming to a close after 19 years and the advertisement for my successor appeared in the March edition (2022). Having been involved in child welfare research for nearly 60 years, I have been thinking about suitable swansongs for the final editorials.
In the past, I tried to use editorials to comment on current affairs and draw out the relevance of articles in each edition. But as we move to thinking about what a post-Covid, post-Brexit welfare state should look like, it seems most fruitful to focus on fundamental principles rather than service details. So, following St Paul's advice, my final editorials will dwell on the things eternal that are not seen rather than the things temporal that are.
In the March edition, this process started with a framework seeking to help fashion a more effective service for adolescents who now form a large proportion of looked after children but seem to have been overlooked in recent years. This time, the topic is the enduring concept of ‘prevention’ and what its role might be in future children’s services.
In the following discussion, several references are cited that are quite old. This is deliberate as I wish to show that pioneers were seeking answers to difficult questions half a century ago and there is a body of high-quality research available that has passed into history. It would be a pity if the new generation simply reinvented the wheel.
The concept of prevention
The idea of prevention has much appeal to professionals who often feel they are working too far down the line. Could children’s problems not have been prevented by action earlier on? Are benefits similar to those emanating from immunisation and clean water in public health possible with social needs?
There are dozens of examples in the social work literature of the late-20th century that distinguish levels and types of prevention. They often use the concepts of primary, secondary and tertiary prevention; the first stage is to stop a problem happening, the second is to stop it getting worse if it occurs and the third to improve the condition of those who have succumbed.
A review of this thinking in the 1999 publication Prevention and early intervention with children in need by Little and Mount expanded this typology and specified four categories of prevention activity. These are:
prevention to intervene with an entire population to stop potential problems from emerging. Universal pre- and post-natal care to reduce infant mortality is an example; early intervention with people who show the first indications of an identified problem and who are known to be at risk of succumbing to it. Special classroom help with disruptive children in primary school is an example; treatment or intervention with individuals who have developed the symptoms of an identified problem. Most personal social services for children and families fall into this category; social prevention to minimise the damage that those who have developed the identified condition can do to others with whom they come into contact. The community supervision of sex offenders is an example.
Early intervention, of course, does not mean early in a child’s life; it refers to early in the development of a problem, so it applies just as much to an older teenager showing incipient mental illness as to a baby at risk of neglect.
Ideas about prevention have naturally developed since this publication and are now firmly integrated into policy and practice. The Early Intervention Foundation was established in 2013, and MP Graham Allen has been a forceful advocate for innovation in this area. But some of the more recent references in guidance and policy documents seem to have been made more in hope than expectation as cuts to welfare services have decimated resources for relational social work and family support.
So, what can we suggest that is feasible in the current social, economic and political context?
Preventing a need not a service
Initially, I would suggest that the Little–Mount classification is more useful than the primary/secondary/tertiary model because it avoids the idea of preventing a service. In the 1970s, for instance, new approaches were developed to prevent children entering residential care, but this view was based on perceptions of a hierarchy of services and the prevention of escalation. This was challenged by the thinking underpinning the Children Act 1989 which proposed a continuum of services from which a selection can be made. Thus, the idea of services in opposition to one another is dropped. A package of interventions selected from a comprehensive menu should be provided whenever there is evidence that they will benefit a child at some point in her or his life.
Principles of effective prevention and early intervention with children in need
This shift in thinking to preventing a need rather than a service sounds simple but is quite radical as it demands a paradigm shift. So, what are the principles underpinning this view?
Prevention and early intervention tend to be more effective when they are a response to clear evidence on the needs of children in any location. It is better to start from empirical evidence about the nature of the problem than from the perceptions of need by professionals and politicians. Prevention and early intervention are likely to be more effective when they are designed in response to clear evidence about the likely causes of children’s problems. The research literature expresses these potential causal links as ‘chains of effects’ in children’s lives.
However, a difficulty complicating the implementation of this model is that we are not usually certain about the precise causes of the problems we are trying to prevent. We may know the factors that are associated with them but this is not the same as understanding the causal process. For example, we know that children in care who need warm and supportive relationships are more likely to have grown up in poverty despite the fact that most poor parents provide a loving environment for their children.
As a result, there is a danger in being over-confident about causes. I am old enough to remember the notion of schizophrenia-inducing mothers whose parenting was alleged to produce mentally ill offspring. There have also been classic disasters resulting from well-intended preventative actions based on wrong beliefs. One occurred in 1848 when public health reformers piped seemingly clean water from the sanitised countryside into central London. As they thought that cholera was caused by bad air, they did not perceive any risk and undertook no tests. Unfortunately, the water turned out to be infected with cholera and killed half of Lambeth.
A second difficulty is the bug bear of all preventative work: what researchers call ‘false positives’. We may think we know the risk factors from daily work with children who display a particular problem, concluding, for example, that they nearly all come from disrupted families or have witnessed violence. Although such observations are useful for understanding the backgrounds of the children, they almost certainly have limited predictive value when applied to the general population.
An example of this difficulty was provided almost 35 years ago by Browne and Saqi (1988) who tried to predict which families would abuse their children. To do this, they conducted a follow-up study of all 14,238 children born in a selected geographical area in 1984. They applied an abuse risk schedule informed by research to all of the families at the time of their child’s birth and found that 949 of them displayed factors associated with a high risk of child abuse. Two years later, they found that 57 of the 14,238 children had been abused. Of these, 47 came from the families identified previously as showing a high risk and 10 came from families displaying no or little risk. The fact that 47 (82%) of the 57 families that abused their children had been identified early on is encouraging because it seems that they can be picked out. However, the problem was that the schedule also identified 902 high-risk families that did not. Thus, for every correct prediction among the high-risk group, there were just under 20 (19.2) incorrect ones or ‘false positives’. If a preventative service had been given to all 949 high-risk families, only one in five would have actually needed it and a lot of time and money would have been wasted.
However, despite these difficulties, researchers like Rutter and Farrington pressed on and were among the first to show that in certain areas, such as offending and some emotional and behavioural disorders, chains of effects could be identified. These chart the interplay between risk factors and protective factors operating in a child’s life, thus helping to explain not only the reasons for the high levels of risk but also why some children succumb and others do not.
Several messages emerged from this tranche of work:
Since chains of effects intersect all areas of children’s lives, prevention and early intervention require the co-operation of many statutory and voluntary agencies. Better diagnosis is a prerequisite of improved understanding of the role of preventative services in children’s lives. The medical term ‘diagnosis’ is preferrable to the social work term ‘assessment’ because the latter tends to describe a situation rather than analyse it. In a diagnosis, the evidence is reviewed in the context of a validated taxonomy of conditions about which a great deal is known with regard to causes and prognosis. This may not yet be possible in children’s services, but the foundations can be laid by seeking consistency among professionals about the nature of children’s needs, their seriousness, the likely prognosis and interventions that work. Since children’s problems are frequently a manifestation of difficulty in several areas of their lives, accurate diagnosis requires information about all aspects of the child’s situation and previous history. A prevention strategy should combine early intervention, treatment and social prevention in a complementary way and not view them as in competition.
But a frustrating reality that will be familiar to journal readers is that no matter how good a prevention service, some people still succumb to the condition. Even in health, diseases that are now rare have not been eliminated. Thus, while the prevalence of the problem may be reduced by early intervention, there is still a need to help those who succumb and to protect others who might be affected by them. Hence, the balance between early intervention, treatment and social prevention is as important as the interventions themselves for good results.
Fashioning a preventative service
So, what are the lessons of all this for service design? Four seem salient:
Since nearly all children in need live with their families and many looked after children return home, effective prevention and early intervention must take account of the ordinary features of family life and incorporate their strengths. Much expertise concerning the solution of children’s problems rests with the children and families themselves, and the process of establishing an effective service often begins by asking how children and families cope with the problems in question. While the fashioning of the service might rely on sophisticated causal theories, the service may take the form of help with the practical problems that users face. What seems a simple service may rest on sophisticated knowledge. There are considerable strengths in the current arrangements in the UK for children in need as well as obvious weaknesses. Effective prevention and early intervention tend to build on agencies’ known strengths and employ clear strategies to overcome identified weaknesses. It is not a matter of starting from scratch. Effective inter-agency work does not necessarily require or imply that work is done from the same geographical or administrative location. The ability to work across conventional boundaries is a particularly important component of preventative innovation. The effects of preventative action may not be immediate and may only appear a generation later. While it may be true that early intervention reduces later difficulties, professionals have to convince auditors and politicians that the wait is worthwhile.
Conclusions
There have been many attempts to boost preventative services over the years but the current situation is unusual in that there will hopefully be a potential coincidence between new resources and a commitment to building an evidence base for children’s services.
But there are still many difficulties to overcome. The knowledge base of children’s services remains limited. Most agencies struggle to know the patterns of the need they seek to address, and many find it hard to chart the volume and nature of their work. Similarly, knowledge of ‘what works’ is improving but validation by ‘gold standard’ methodologies is still uncommon. When sound evidence is lacking or perceived as equivocal, there is a danger that too much hope is accorded to administrative reforms that are unlikely to significantly improve clinical effectiveness.
Alongside this, we must hold realistic expectations about what better prevention and early intervention can actually achieve. As said previously, early intervention, treatment and social prevention are not alternatives, and it is the combination that will make a difference to children’s lives. But, again, it must be emphasised that we are unlikely ever to eradicate a need; the best we can hope for is to change the pattern of a problem by blocking or diverting known causal chains. Those expecting prevention to end or significantly reduce perennial problems, such as offending or child maltreatment, may be deeply disappointed.
I suggest that we can usefully begin the process of service development by seeking agreement about what we mean by prevention and how the agreed definition fits into the four categories of prevention described earlier. We can then identify the problems likely to benefit from these approaches and explore what is known, what services are available and what innovations are needed. From that sound start, we can venture into the less charted waters that still confound us.