Based on a Freedom of Information request with data from 75 per cent of all English children’s services departments covering over half a million children, this paper shows that 22.5 per cent of children born in the 2009–10 financial year were referred to children’s social care before their fifth birthday. Three-quarters of them were at some point assessed, almost two-thirds found to be in need and a quarter formally investigated. These findings show the full extent of children’s involvement in children’s social care before the age of five. One in every nine children born in 2009–10 was suspected by social workers of being abused and this high level of involvement is only justifiable if it is demonstrably reducing harm and promoting well-being of children—an outcome which is contested. Early Help’s introduction was associated with high proportions of children being referred and assessed and rapidly increasing numbers of investigations, thus questioning its ability to prevent entry to the child protection system. The paper calls for a change from the current emphasis on individualised and investigative approaches to child protection in order to provide an effective and humane response to children, the majority of whom live in families affected by high levels of deprivation and poverty.

Introduction

Despite statistics on safeguarding being collected in England for many years, numerous inquiries and decades of research, there has never been information published at a national level to show how many children are involved in the system over their lifetime. This paper reports on a Freedom of Information (FoI) request that gathered data from 75 per cent of all local authorities (LAs) in England covering over half a million children: 80 per cent of all those born in the 2009–10 financial year. It provides, for the first time, information on how many children were referred to children’s services before reaching the age of five. It also shows how many were formally assessed; suspected of abuse; subject to a child protection investigation; on a child protection plan; and entered care. More than one in five of all children were referred and many progressed deeper into the formal child protection system. Given the shame and fear that accusations of abuse can cause (Gibson, 2013), this high level of involvement is only justifiable if it is demonstrably reducing harm and promoting well-being of children—a goal which is certainly contested (Edwards et al., 2015; Featherstone et al., 2014b; Frost and Parton, 2009; Gilbert et al., 2012).

The FoI request shows that this information on numbers of children involved in the safeguarding system over five years is easily produced. The fact that such important information has, until now, been unavailable is an indication of how research agendas and policy making in many cases have been forged either in the glare of publicity or from a focus on the worst cases, rather than based on a firm understanding of how the system operates as a whole.

Background

The long-standing concern about over-intervention in the child protection system was strongly voiced in the 1990s in a summary of government-funded research which identified as central:

… the criticism that many investigations are undertaken, many families are visited and case conferences called but that in the end, little support is offered to the family. In such situations, it is unsurprising that participants become angry, alienated and bewildered. Furthermore, the children are not helped and a chunk of valuable child care resource has been consumed with little apparent benefit (Department of Health, 1995, pp. 54–5).

Because of the harm done to families by suspicion of abuse, the report went on to call for a change in the balance between section 47 enquiries into a child suspected of or suffering significant harm (hereafter investigations) and family support by reducing the number of children investigated and increasing the numbers receiving support. Concerns continue to be voiced about the level of investigations (Featherstone et al., 2014a, 2014b; Peckover, 2014) and a too strong focus on risk management and incidents of harm rather than help (Hayes and Spratt, 2014; Parton, 2014). The Munro Review (Munro, 2011) also called for a greater focus on Early Help to reduce abuse and neglect.

Child protection process

The statutory guidance on safeguarding children (DfE, 2015b, p. 15) requires the publication of local thresholds for assessment and intervention for Early Help and referral for assessment for statutory services from children’s social care (CSC). The London Children’s Safeguarding Board’s (LCSB, undated, p. 6) threshold document suggests four levels of need:

  1. No additional needs: where all the child’s health and developmental needs are met by universal services such as health and education.

  2. Early Help: where children have additional needs and/or ‘are showing early signs of abuse or neglect’ (LCSB, undated, p. 6); Children receive a multi-agency assessment and their needs are met by universal or targeted services not including CSC.

  3. Children with complex multiple needs: this is the threshold for children to be referred to and assessed by CSC under section 17 because ‘children require specialist services in order to achieve or maintain a satisfactory level of health or development or to prevent significant impairment of their health and development and/or who are disabled’ (LCSB, undated, p. 6).

  4. Children in acute need: these children have more complex needs and meet the criteria for an investigation that there is ‘reasonable cause to suspect that a child … is suffering, or is likely to suffer, significant harm’.

The statutory child protection process starts with a referral to CSC. Anyone with a concern about a child’s well-being can make a referral. Some referrals do not meet the statutory threshold for assessment by CSC and are directed into universal services or Early Help. Referrals that appear to be at level 3 or 4 move to assessment. Where there is a suspicion that the child is suffering or likely to suffer significant harm (i.e. level 4), a strategy discussion (an inter-agency consultation involving the police) is held to decide whether an investigation under section 47 is needed. Where the investigation finds that the child is suffering or likely to suffer significant harm, a child protection conference is held to bring together ‘family members (and the child where appropriate), with the supporters, advocates and professionals most involved with the child and family, to make decisions about the child’s future safety, health and development’ (DfE, 2015b, p. 41).

Where further action is needed, the child protection conference makes a child protection plan to ensure the child is safe, promote the child’s health and development, and support the family in doing this.

Involvement in child protection investigations places considerable stress on parents (Ghaffar et al., 2012) even where there is a good relationship with the social worker and the investigation finds that there has been no abuse (Davies, 2011). During much of the five-year period covered by the FoI request, the safeguarding policy attempted to reduce statutory assessments and investigations through provision of Early Help and the Common Assessment Framework. The national statistics show that, instead of a reduction, there was a rapid increase in investigations. Figure 1 shows that child protection investigations increased by 79.4 per cent between 2009–10 and 2014–15 (DfE, 2015b) whilst the numbers of referrals to LA children’s services (+5.3 per cent) and children in need (+6.8 per cent) were little changed. Thus, despite the increasing focus on Early Help, there was a rise from 89,300 to 160,200 investigations a year. A higher proportion of referred children (14.8 per cent in 2009/10 increasing to 25.2 per cent in 2014/15) and children in need (from 23.6 per cent to 39.7 per cent) were investigated. Whilst there was an increase in children placed on a child protection plan (40.4 per cent) over this six-year period, the much larger increase in investigations meant that the number of children who came under suspicion and were investigated but were not found to be significantly harmed more than doubled from 45,000 to 98,000. In 2014–15, it is likely that fewer than two in every five investigations led to a child protection plan. Whilst there is a duty to make arrangements for ‘step down’ for children in this situation to receive Early Help, the effectiveness of these arrangements to improve the well-being and reduce the need for statutory interventions is unclear. There are no statistics on the numbers of children moving from an investigation to support services; few of the arrangements made by agencies for Early Help have been evaluated on outcomes for children (Luckock et al., 2015); and an Ofsted (2015) audit found that opportunities to help children were missed. However, it has been found that parents frequently rejected voluntary support following an investigative response to referral (Thorpe et al., 2012). This increasing trend, in which over 60 per cent of investigations do not find that children are suffering or likely to suffer significant harm, is increasing the stress on and causing harm to a growing number of families and simultaneously reducing the likelihood that these children receive help.

Figure 1

Increasing investigative response to referrals 2009–10 to 2014–15 This figure draws on annual statistical returns to show a graph of the number of children investigated under section 47 and the number of child starting a child protection plan. It also shows a table giving the proportion of referrals and children in need who were subject to a child protection investigation.

Figure 1

Increasing investigative response to referrals 2009–10 to 2014–15 This figure draws on annual statistical returns to show a graph of the number of children investigated under section 47 and the number of child starting a child protection plan. It also shows a table giving the proportion of referrals and children in need who were subject to a child protection investigation.

This trend for an increasing involvement of children in the child protection system is seen in a number of countries (Bilson et al., 2015; Munro and Manful, 2012) where the broadening of criteria from maltreatment to responding to children’s well-being leads to more investigations (e.g. Trocmé et al., 2014; Bilson et al., 2016) and children entering state care (Gilbert et al., 2011; Bilson et al., 2016). Trocmé et al. (2014) found that, over a ten-year period, the rates of urgent child protection referrals (any abuse of children aged under five, physical abuse requiring medical attention or sexual abuse) were unchanged, whilst investigations because of concerns about the well-being of children, often because of concerns about neglect and emotional abuse, more than doubled. A similar pattern is seen in England where, between 2009–10 and 2014–15 (DfE, 2015a), the numbers on child protection plans because of neglect (+28 per cent) and emotional abuse (+54 per cent) grew sharply whilst physical abuse fell and sexual abuse changed little (Figure 2). The number of children in care also increased by 7.9 per cent between 2009–10 and 2014–15 alongside the 93.4 per cent increase in children adopted or placed in special guardianship.

Figure 2

Increasing use of child protection plans following neglect and emotional abuse.This figure draws on national statistics to show a graph of the number of children on a child protection plan at 31 March showing the type of neglect or abuse that led to the plan.

Figure 2

Increasing use of child protection plans following neglect and emotional abuse.This figure draws on national statistics to show a graph of the number of children on a child protection plan at 31 March showing the type of neglect or abuse that led to the plan.

Bywaters et al. (2014b, p. 7), in a large study in the West Midlands, found that 40 per cent of children in care or on a child protection plan lived in communities in the tenth decile of deprivation (i.e. the 10 per cent of the population who are most deprived) and 60 per cent in the ninth and tenth deciles. There was a clear dose–response relationship between measures of deprivation and being in care or on a child protection plan and those in the tenth decile of deprivation were around eleven times more likely to be on a child protection plan or in care than those in the least deprived decile.

There is much written about the harm caused by neglect and emotional abuse (e.g. Gould et al., 2012; Norman et al., 2012) and there are increasingly ‘now or never’ calls for urgent, early intervention such as Brown and Ward’s (2012) paper for the judiciary. Wastell and White (2012, p. 400) show how these calls misinterpret the neuroscientific evidence on which they are frequently based. Further, Edwards et al.’s research demonstrates the rhetorical use of this misinterpreted neuroscientific ‘evidence’ made by policy makers and practitioners (2015, p. 167) that leads to ‘justifying gendered, raced and social inequalities’ policy responses that are ‘positioning poor mothers as architects of their children’s deprivation’ (see also Featherstone et al., 2014a). Most officially recognised neglect and emotional abuse occurs in the context of poverty and investigative responses place blame on parents without reference to the role poverty plays in creating the conditions that reduce the ability to parent well (Bywaters et al., 2014b). Additionally, the evidence on individualised interventions to prevent further neglect and emotional abuse is sparse (Glaser et al., 2012). MacMillan et al. (2009, p. 250) state that ‘no intervention has yet been shown to be effective in preventing recurrence of neglect’ and ‘there are currently no known approaches to prevent emotional abuse or exposure to intimate-partner violence’. Like Mikton and Butchart (2009, p. 1), they note a few interventions which show promise for helping children overcome the effects of neglect or emotional abuse but these findings are tentative due to methodological shortcomings of the reviews and outcome evaluation studies. This raises the question of whether increasing investigations followed by individualised interventions which necessarily involve an imbalance of power and parental blame is the best response to the majority of neglect and emotional maltreatment (Barnard and Bain, 2015; Department of Health, 1995; Edwards et al., 2015; Featherstone et al., 2014a, 2014b; McDonell et al., 2015).

Longitudinal studies

Internationally, there are few longitudinal studies of involvement in child protection and only one in England (Sidebotham et al., 2002). This study of children born in 1991–92 had a number of limitations including voluntary enrolment in the sample which led to an under-representation of poorer families and the sample being from a single LA. It found that 2.0 per cent of the children were recorded as being statutorily investigated before their sixth birthday and 0.8 per cent placed on the child protection register. Across Australia, statistics from client information systems and longitudinal studies have shown referral rates to the child protection system before the age of eighteen to be around one in four children in New South Wales, Queensland and Victoria (Scott, 2014, p. 16), with the lowest published ratio being one in eight in Western Australia (Bilson et al., 2015). A study undertaken in California (Magruder and Shaw, 2008, p. 178) found that 19.8 per cent of children were reported before the age of seven. These high figures show how examining annual rates gives little indication of the huge extent of involvement in the child protection system.

The FoI request

The FoI request was sent to all 152 English LAs in mid-2015. It requested data that are collected for the national statistics on the cohort of children born between 1 April 2009 and 31 March 2010 (hereafter 2009–10). Questions were based on categories from the national children in need survey which commenced in 2009, as this is a format used in client information systems across the country. A copy of the FoI request is available from the internet (http://bilson.org.uk/foi).

The FoI responses

Out of 152 English LAs, seven did not reply and thirty-one, for various reasons, supplied no data. Thus, 114 (75.0 per cent) provided some or all of the data. Throughout this paper, Office for National Statistics (2015) Mid-year Population Estimates for children aged zero or one on 30 June 2010 (75 per cent of those aged zero and 25 per cent aged one) are used as a proxy to calculate the number of children in each LA in the 2009–10 cohort. The total population of the 2009–10 cohort in the 114 LAs was 527,550 (79 per cent of children in England). Whilst most LAs could answer all the questions, there was an error in the initial round of requests for data and the categories of need concerning low income and absent parenting were missed from the question list. In later requests, these were included and answered by twenty-five LAs for low income (16.4 per cent) and twenty-one for absent parenting (13.8 per cent). Only seventy LAs could provide information on the number of children who had been placed on a supervision order and other questions were answered by at least ninety-nine LAs.

It is the duty of an LA responding under FoI to ensure no data are provided that can identify an individual. In some cases, the response to a particular question concerned a small number (usually under five) of children and here the LAs mainly applied the good practice of indicating this and not producing the actual figure.

Issues and limitations

In reporting on the children in need census, the Department for Education (DfE) gives an indication of the quality of responses from LAs based on: the quality of the data received; whether the LA reports issues of data quality in their return; and year-on-year comparisons. A three-point scale is used, with 3 indicating the highest quality. In the most recent report (DfE, 2015a), out of 152 LAs, one did not provide data on referrals; three were given a rating of 1; thirty a rating of 2; and the majority of 118 (78.1 per cent) received the highest rating of 3. Similar scores of quality are given for other elements of the data. Thus, the DfE recognises that there is some variation in the quality of these data but that the majority are of the highest quality.

In the responses to the FoI request, there were some data returns that seem particularly high and others that were very low when compared to other results and taking into account the national statistics. It was not possible to identify whether these ‘outliers’ were incorrect data. For example, four LAs (Islington, Liverpool, Luton and North Yorkshire) provided data showing that above 50 per cent of children had been referred. Islington and Liverpool had reported high rates of referrals in their annual data and a high proportion of their population live in the most deprived areas in the country. In contrast, Luton and North Yorkshire are considerably lower on both these measures and seem on this basis to be more likely to be erroneous. There are similar issues at the other end of the scale, where nine LAs had a much lower-than-average proportion of children referred and the FoI data differed markedly from data reported for annual referrals. Interestingly, seven of these nine LAs also had a quality rating of 2 for their data in 2014. Excluding these ‘outliers’ lowered the average rate of children referred before the age of five reported below by less than 2 per cent to 21.82 per cent.

Because FoI requests can be refused if the time to create the response is excessive, the authors chose not to ask for information to be disaggregated by gender or ethnicity even though other research (Bywaters et al., 2014a, 2014b) has shown that there are certainly differences in engagement with the child protection system when ethnicity is taken into account.

The data focus on official reactions to child protection and do not indicate levels of actual abuse of children. Additionally, whilst there have been attempts at a national level to rationalise practice, there are large differences in decisions about what is recorded as a referral or an investigation (Platt and Turney, 2014).

The FoI request results

The results of the FoI request are summarised in Table 1. Items in the table numbered 1 to 10 provide summaries of answers to questions of the form ‘Had the child at any time before his or her 5th birthday been …?’. The primary need codes (N1 to N8) record the main reason why a child started to receive service by the LA. The category recorded is that which fitted best at the point at which the first assessment was carried out and, if there was difficulty choosing a single category, then the category highest in the list was used (this is the order shown in Table 1). Children can have more than one episode of being a child in need and the request asked for the category given on the first episode. The category of abuse is allocated when a child is placed on a child protection plan. Children can be on a child protection plan more than once and the FoI asked for the category given at the start of the first plan for each child. Since not all agencies were able to answer all the different questions, the table shows the number of LAs which replied to each question. The percentage of children born in 2009–10 was then calculated using the population of children in those LAs that replied.

Table 1

Results of the Freedom of Information request showing number and percentage of children born between 1 April 2009 and 31 March 2010 who had reached various stages of the children’s services before their fifth birthday

 Percentage of all children born in 2009–10 Number of children Number of LAs providing data# 
1. Children referred 22.5 115,735 111 
2. Children assessed by children’s services 17.0 83,713 106 
3. Children in need 14.3 72,181 108 
Primary need code on the first episode as a child in need 
N1 Abuse or neglect 7.8 38,365 102 
N2 Child’s disability 0.6 2,737 101 
N3 Parental disability or illness 0.5 2,404 99 
N4 Family in acute stress 1.8 8,775 102 
N5 Family dysfunction 3.3 16,084 102 
N6 Socially unacceptable behaviour 0.2 1,001 100 
N7 Low income 0.1 80* 24 
N8 Absent parenting 0.1 52* 21 
4. Children subject of a strategy discussion 6.2 29,895 104 
5. Children subject of a section 47 enquiry 5.4 27,794 111 
6. Children subject of a child protection conference 3.7 19,014 110 
7. Children subject of a child protection plan 3.5 17,823 111 
Category of abuse as recorded when the first child protection plan commenced 
a) Neglect 1.6 7,721 105 
b) Physical abuse 0.5 2,662 103 
c) Sexual abuse 0.1 633 103 
d) Emotional abuse 1.0 4,701 104 
e) Multiple/not recommended 0.3 1,272 104 
8. Children subject of a supervision order 0.3 1,010 78 
9. Children subject of a care order 1.1 5,312 107 
10. Children ever having been looked after [in care] 1.7 8,929 111 
 Percentage of all children born in 2009–10 Number of children Number of LAs providing data# 
1. Children referred 22.5 115,735 111 
2. Children assessed by children’s services 17.0 83,713 106 
3. Children in need 14.3 72,181 108 
Primary need code on the first episode as a child in need 
N1 Abuse or neglect 7.8 38,365 102 
N2 Child’s disability 0.6 2,737 101 
N3 Parental disability or illness 0.5 2,404 99 
N4 Family in acute stress 1.8 8,775 102 
N5 Family dysfunction 3.3 16,084 102 
N6 Socially unacceptable behaviour 0.2 1,001 100 
N7 Low income 0.1 80* 24 
N8 Absent parenting 0.1 52* 21 
4. Children subject of a strategy discussion 6.2 29,895 104 
5. Children subject of a section 47 enquiry 5.4 27,794 111 
6. Children subject of a child protection conference 3.7 19,014 110 
7. Children subject of a child protection plan 3.5 17,823 111 
Category of abuse as recorded when the first child protection plan commenced 
a) Neglect 1.6 7,721 105 
b) Physical abuse 0.5 2,662 103 
c) Sexual abuse 0.1 633 103 
d) Emotional abuse 1.0 4,701 104 
e) Multiple/not recommended 0.3 1,272 104 
8. Children subject of a supervision order 0.3 1,010 78 
9. Children subject of a care order 1.1 5,312 107 
10. Children ever having been looked after [in care] 1.7 8,929 111 

This is the percentage of children in the local authorities that provided data.

#

This is the number of local authorities that provided data for this question.

*

There are many returns in this category with fewer than five children and the total is estimated.

Discussion of results

There are many issues raised by these FoI results. The 22.5 per cent of all children referred to social care is likely to be concentrated in the areas of greatest deprivation as shown by the research of Bywaters et al. (2014a, 2014b) discussed earlier. These children live in the most deprived areas where poverty, instability of housing, crime, poor health and many other stressors are commonplace. In these communities, parents will experience higher-than-average levels of suspicion and investigation with the stigma and anxiety described above (Hooper et al., 2007). These issues will be considered further in a future paper. This paper focuses on the high proportion of children who are involved in children’s services and the ineffectiveness of the system in reducing the need for this statutory intervention.

High proportions of children involved in the child welfare system

Table 1 shows that 111 of the 114 LAs provided data and 117,875 children were referred to children’s services departments before their fifth birthday—22.5 per cent of the estimated 521,688 children in the 2009–10 cohort in these LAs. Thus, more than one in every five children born in 2009–10 had concerns referred to children’s services before the age of five.

What this study cannot answer is how many of these children and their parents actually received Early Help or responses other than assessment, investigation or care. Research by Thorpe et al. (2011) preceding much of the current focus on Early Help suggested that almost three-quarters of referred children did not receive services. They studied around 4,200 referrals in fourteen LAs, with each child followed for a year after referral. Only 19.5 per cent of these referrals were initially treated as potentially being a child in need where CSC staff negotiated with parents about what help or advice was required. Other referrals were treated with an investigative orientation in which, for example, checks with other agencies were made without parental consent and in some cases joint visits were undertaken with the police. Not all of these cases were classified as formal child protection investigations, though all were treated with a forensic orientation seeking to identify evidence. If this rate applied to the children in this study, we would see around 95,000 children, 18 per cent of all children, having experienced suspicion and an investigatively oriented response to referral. In their study, 44.0 per cent of all the referrals were offered services, though mostly these ‘were referred on to or “signposted” to other agencies’ (Thorpe et al., 2011, p. 32), and the report states that the figures for take-up of these services was probably over-estimated but well below this at ‘just over one quarter’ (p. 33) of referrals. Few of the referrals treated with an investigative orientation took up services. An Ofsted (2015) study of Early Help also found there was little follow-up on referred on/signposted services. Morrison (2010, pp. 314–15) sums up how many parents feel about their engagement with CSC which may explain the low level of service acceptance:

Much is at stake for these families: fear of exposure, stigma, removal of their children or even prosecution. Fearful and anxious at the intrusion of external agencies, they may be defensive, angry and unwilling to acknowledge the reality of any difficulties. Negative previous experiences of statutory agencies may exacerbate this.

Figure 3 shows the proportions reaching each stage of the children’s service process before their fifth birthday. One in every six children had been assessed by children’s services and more than one in every seven were considered to be in need and thus required services to achieve or maintain a reasonable standard of health or development. However, it cannot be assumed that these children, classified by the statistics as being in need, actually received services or help for their families. The most common reasons for children to be in need that social workers recorded were ‘Abuse and neglect’ (7.78 per cent of children born in 2009–10), which includes but is not limited to children formally investigated; and ‘Family dysfunction’ (3.43 per cent), which includes ‘families where the low parenting capacity is at risk of, or actually is, impairing the child’s health and development’ and should ‘include children whose safety is in concern as a consequence of family dysfunction, but for whom there is not yet hard enough evidence to invoke child protection measures’ (DfE, 2015a, p. 45). Thus, social workers recorded their suspicion of abuse or neglect through these two categories for more than one in every nine children born in 2009–10. However, in some cases in which maltreatment is very minor and/or the response by family and others meet his/her needs, the child will not be recorded in these categories. Thus, the 11.2 per cent in these categories is an underestimate of the number of children found to be or suspected by social workers of being abused or neglected, though more than half of this was below the threshold of concern about suspicion of significant harm that would trigger an investigation. Figure 3 also shows that 5.34 per cent of children, more than one in every nineteen, were formally investigated under child protection legislation and 1.74 per cent had spent some time in care.
Figure 3

Proportions involved in children’s social care This figure uses findings of the Freedom of Information request to show the proportion of children born in 2009–10 who reached each stage of the children child protection system before reaching the age of five.

Figure 3

Proportions involved in children’s social care This figure uses findings of the Freedom of Information request to show the proportion of children born in 2009–10 who reached each stage of the children child protection system before reaching the age of five.

To illustrate what this means, Figure 4 shows what involvement in the social care system would be found amongst children in an ‘average’ class of thirty with a fifth birthday in 2014–15. Seven of the thirty children would have been referred to CSC and four of these would have complex/multiple needs. Between one and two children would have been the subject of a child protection investigation and one child found to be suffering or likely to suffer significant harm.

Figure 4

Referrals and investigations in an ‘average’ class of thirty five-year-olds This figure illustrates the proportions of children reaching different stages of the child protection system if the results of the Freedom of Information request were seen in an ‘average’ class of thirty children.

Figure 4

Referrals and investigations in an ‘average’ class of thirty five-year-olds This figure illustrates the proportions of children reaching different stages of the child protection system if the results of the Freedom of Information request were seen in an ‘average’ class of thirty children.

As discussed above, over the period of this study, there was a growing focus on Early Help intended to reduce the need for statutory involvement or entry to the child protection system. The FoI shows that three-quarters (76 per cent) of children who were referred reached the level of being assessed for statutory services and just under two-thirds (63 per cent) became a child in need. Thus, a high proportion of referred children progressed, at some time, to the formal statutory system. This contrasts with the findings of the Committee on Education (House of Commons Education Committee, 2012, p. 5), which recommended research into whether ‘thresholds for section 17 and section 47 interventions are too high and/or rising in some area’. Ironically, during the period, whilst this ‘evidence’ that thresholds for involvement of children’s services were too high was being given to the committee, section 47 investigations rose by 25 per cent in a single year whilst referrals fell by 1.6 per cent. However, those giving this evidence may have been more concerned about the lack of help being offered to children as found by Thorpe et al. (2011).

There is currently no way to ascertain the number of children at a national level receiving ‘Early Help’ (Holmes et al., 2012, p. 16). It is thus not possible to ascertain how Early Help is being used and whether it is reaching out to a wider group still than the 22.5 per cent referred. What is clear is that a high proportion of children are involved in statutory interventions despite the existence of these supposedly preventive measures and it is not known whether Early Help has contributed to rather than reduced this trend. This high level of statutory involvement and growing use of section 47 investigations indicates that the system has an increasingly forensic orientation (Hayes and Spratt, 2014). A worrying possibility is that Early Help has the unintended consequence of net-widening—drawing more children into this investigative maelstrom—and the concerns expressed by some parents that involvement in the Early Help system ‘might lead to professionals across agencies making judgements about their parenting skills and subsequently result in a referral to children’s social care’ (Holmes et al., 2012, p. 8) may be well founded.

One in every sixteen children (6.2 per cent) was the subject of a strategy discussion before reaching the age of five and thus thought to have been suffering or likely to suffer significant harm. The number formally investigated under section 47 was slightly lower at 5.4 per cent (one in every nineteen children) and this is because some strategy discussions will provide information to assuage the suspicion that there is significant harm. One in twenty-seven children (3.7 per cent) reached a child protection conference and one in twenty-nine (3.5 per cent) were placed on a child protection plan. Thus, 2.7 per cent of all children, 43 per cent of children reaching the stage of a strategy discussion, were not placed on a child protection plan before their fifth birthday having been formally suspected of abuse.

Approaching half (45 per cent) of the occasions that a child was first placed on a child protection plan, and for 1.6 per cent of all children, the category of abuse was neglect and for another quarter (27 per cent, 1.0 per cent of children) it was emotional abuse. Physical abuse (16 per cent and 0.5 per cent of children) and sexual abuse (4 per cent and 0.1 per cent) combined accounted for fewer than one in five of the child protection plans for this young group of children. These figures reflect the trends in the national statistics for children on child protection plans shown in Figure 2.

Alternative non-blaming approaches

Despite the lack of evidence that the child protection system reduces harm to young children (e.g. Gilbert et al., 2012), this investigative and blaming approach continues to increase and there are even calls by campaigning organisations such as the NSPCC for more children to be statutorily involved. The high incidence of suspicion shown by this study is likely to have ‘traumatic effects on families’ (Department of Health, 1995, p. 44) and create a climate of fear for those living in the poor communities (Hooper et al., 2007) where, as shown by Bywaters et al. (2014a, 2014b), higher rates of suspicion will be concentrated. Additionally, unless there has been a considerable improvement in recent years, a high proportion of those entering the system are not offered support; are signposted without help to access support; or do not accept support if it is offered (Thorpe et al., 2011).

The findings of this FoI add more weight to the arguments of those calling for a new paradigm for social work (e.g. Featherstone et al., 2014b). Developmental social work provides such an alternative paradigm with its focus on advocacy, poverty alleviation, and developing family and community strengths (Midgley and Conley, 2010; Bilson and Larkins, 2013). It specifically aims to create ‘tangible improvements in standards of living, health and education, and a concomitant reduction in poverty, malnutrition, and illiteracy’ (Midgley and Conley, 2010, p. 13). This alternative orientation for CSC would aim to improve the conditions of families and communities enabling them to enhance the well-being of children rather than relying on agents of child protection to seek out harm. Its community orientation would allow it to embrace approaches such as ‘Strong Communities’ that focuses on increasing neighbourhood cohesion to promote the well-being of children and families (McDonell et al., 2015); evidence-based approaches to early years education that increase social capital such as FAST (McDonald et al., 2015); and even advocating to increase incomes of poor families that on its own has been shown to increase the well-being of children (Cooper and Stewart, 2013).

Conclusion

This is the first study to show the extent that children are involved in CSC over their first five years. This lack of longitudinal data in England is extremely worrying. The FoI request shows longitudinal data can easily be produced. Since it is vital to have an accurate picture of the child protection system in order to make policy, plan and assess services, government should make longitudinal data available annually. The findings also highlight the need for further research into the impact of Early Help on the wider child protection system, and the provision and outcomes of services for children referred to CSC.

This paper shows the high levels of suspicion of abuse and investigations and the increasing investigative orientation of the child protection system. Since 57 per cent of referrals and 59 per cent of child protection plans are on children aged five or older (DfE, 2015a), the proportion of children born in 2009–10 who are drawn into the statutory system will, before they reach the age of eighteen, increase substantially above the 22.5 per cent found here. Whilst some children need to be protected, there is little evidence to support this scale of statutory involvement or the growing focus on early, and increasingly investigative, interventions alongside increases in removal of children from families into long-term care, special guardianship and adoption. These policies bring high levels of suspicion, fear and shame on a considerable proportion of families in the most deprived areas where this activity is concentrated. This is done without evidence that the individualised, investigative approach is effective in preventing further harm. Alternatives include a more humane developmental social work orientation and approaches that promote cohesion in neighbourhoods and reduce deprivation and poverty. But, to achieve this, we need to step away from our current preoccupation with the search for ever more parents to investigate and blame.

Acknowledgements

Many thanks to all the information officers who responded to our FoI request; to Brid Featherstone and Paul Bywaters for comments on early versions; and to the School of Social Work, Care and Community at the University of Central Lancashire for support for the research.

References

Barnard
M.
Bain
C.
(
2015
)
‘Resisting your good intentions: Substance‐misusing parents and early intervention to support and monitor children in need’
,
Child and Family Social Work
 ,
20
(
2
), pp.
171
80
.
Bilson
A.
Cant
R.
Harries
M.
Thorpe
D. H.
(
2015
)
‘A longitudinal study of children reported to the child protection department in Western Australia’
,
British Journal of Social Work
 ,
45
(
3
), pp.
771
91
.
Bilson
A.
Cant
R.
Harries
M.
Thorpe
D. H.
(
2016
) ‘Accounting for the increase of children in care in Western Australia: What can client information systems tell us?’, manuscript in preparation
Bilson
A.
Larkins
C.
(
2013
)
‘Providing alternatives to infant institutionalisation in Bulgaria: How gatekeeping can benefit from a social development orientation’
,
Children and Youth Services Review
 ,
35
(
9
), pp.
1566
75
.
Brown
R.
Ward
H.
(
2012
)
Decision-Making within a Child’s Timeframe
 ,
2nd edn
,
London
,
Childhood Wellbeing Research Centre, Institute of Education
.
Bywaters
P.
Brady
G.
Sparks
T.
Bos
E.
(
2014a
)
‘Child welfare inequalities: New evidence, further questions’
,
Child and Family Social Work
  Advanced Access published March 2014, 10.1111/cfs.12154.
Bywaters
P.
Brady
G.
Sparks
T.
Bos
E.
(
2014b
)
‘Inequalities in child welfare intervention rates: The intersection of deprivation and identity’
,
Child and Family Social Work
  Advanced Access published June 2014, 10.1111/cfs.12161.
Cooper
K.
Stewart
K.
(
2013
)
Does Money Affect Children’s Outcomes? A Systematic Review
 ,
York
,
Joseph Rowntree Foundation
.
Davies
P.
(
2011
)
‘The impact of a child protection investigation: A personal reflective account’
,
Child and Family Social Work
 ,
16
(
2
), pp.
201
9
.
Department for Education (DfE)
(
2015a
) Characteristics of children in need: 2014 to 2015, London, DfE, available online at https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/469737/SFR41-2015_Text.pdf.
Department for Education (DfE)
(
2015b
) Working Together to Safeguard Children: A Guide to Inter-Agency Working to Safeguard and Promote the Welfare of Children, London, DfE, available online at https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/419595/Working_Together_to_Safeguard_Children.pdf.
Department of Health
(
1995
)
Child Protection: Messages from Research
 ,
London
,
HMSO
.
Edwards
R.
Gillies
V.
Horsley
N.
(
2015
)
‘Brain science and early years policy: Hopeful ethos or “cruel optimism”?’
,
Critical Social Policy
 ,
35
(
2
), pp.
167
87
.
Featherstone
B.
White
S.
Morris
K.
(
2014a
)
‘A marriage made in hell: Early intervention meets child protection’
,
British Journal of Social Work
 ,
44
, pp.
1735
49
.
Featherstone
B.
White
S.
Morris
K.
(
2014b
)
Re-Imagining Child Protection: Towards Humane Social Work with Families
 ,
Bristol
,
Policy Press
.
Frost
N.
Parton
N.
(
2009
)
Understanding Children’s Social Care: Politics, Policy and Practice
 ,
London
,
SAGE
.
Ghaffar
W.
Manby
M.
Race
T.
(
2012
)
‘Exploring the experiences of parents and carers whose children have been subject to child protection plans’
,
British Journal of Social Work
 ,
42
(
5
), pp.
887
905
.
Gibson
M.
(
2013
)
‘Shame and guilt in child protection social work: New interpretations and opportunities for practice’
,
Child and Family Social Work
 ,
20
(
3
), pp.
333
43
.
Gilbert
N.
Parton
N.
Skivenes
M.
(
2011
)
Child Protection Systems: International Trends and Orientations
 ,
Oxford
,
Oxford University Press
.
Gilbert
R.
Fluke
J.
O’Donnell
M.
Gonzalez-Izquierdo
A.
Brownell
M.
Gulliver
P.
Janson
S.
Sidebotham
P.
(
2012
)
‘Child maltreatment: Variation in trends and policies in six developed countries’
,
The Lancet
 ,
379
(
9817
), pp.
758
72
.
Glaser
D.
Prior
V.
Auty
K.
Tilki
S.
(
2012
) ‘Does training in a systematic approach to emotional abuse improve the quality of children’s services?’, available online at https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/181602/DFE-RB196.pdf.
Gould
F.
Clarke
J.
Heim
C.
Harvey
P. D.
Majer
M.
Nemeroff
C. B.
(
2012
)
‘The effects of child abuse and neglect on cognitive functioning in adulthood’
,
Journal of Psychiatric Research
 ,
46
(
4
), pp.
500
6
.
Hayes
D.
Spratt
T.
(
2014
)
‘Child welfare as child protection then and now: What social workers did and continue to do’
,
British Journal of Social Work
 ,
44
(
3
), pp.
615
35
.
Holmes
L.
McDermid
S.
Padley
M.
Soper
J.
(
2012
)
Exploration of the Costs and Impact of the Common Assessment Framework
 ,
London
,
DfE
.
Hooper
C. A.
Gorin
S.
Cabral
C.
Dyson
C.
(
2007
)
Living with Hardship 24/7: The Diverse Experiences of Families in Poverty in England
 ,
London
,
Frank Buttle Trust
.
House of Commons Education Committee
(
2012
) Children First: The Child Protection System in England: Fourth Report of Session, 13, 8, available online at www.publications.parliament.uk/pa/cm201213/cmselect/cmeduc/137/137.pdf.
London Children’s Safeguarding Board (LCSB)
(undated) Threshold Document: Continuum of Help and Support, London, London Children’s Safeguarding Board, available online at www.londoncp.co.uk/files/revised_guidance_thresholds.pdf.
Luckock
B.
Barlow
J.
Brown
C.
(
2015
)
‘Developing innovative models of practice at the interface between the NHS and child and family social work where children living at home are at risk of abuse and neglect: A scoping review’
,
Child and Family Social Work
  Advanced Access published March 16, 2015, 10.1111/cfs.12228.
MacMillan
H. L.
Wathen
C. N.
Barlow
J.
Fergusson
D. M.
Leventhal
J. M.
Taussig
H. N.
(
2009
)
‘Interventions to prevent child maltreatment and associated impairment’
,
The Lancet
 ,
373
(
9659
), pp.
250
66
.
Magruder
J.
Shaw
T. V.
(
2008
)
‘Children ever in care: An examination of cumulative disproportionality’
,
Child Welfare
 ,
87
(
2
), pp.
169
88
.
McDonald
L.
Miller
H.
Sandler
J.
(
2015
)
‘A social ecological, relationship-based strategy for parent involvement: Families And Schools Together (FAST)’
,
Journal of Children’s Services
 ,
10
(
3
), pp.
218
30
.
McDonell
J. R.
Ben-Arieh
A.
Melton
G. B.
(
2015
)
‘Strong communities for children: Results of a multi-year community-based initiative to protect children from harm’
,
Child Abuse and Neglect
 ,
41
, pp.
79
96
.
Midgley
J.
Conley
A.
(eds) (
2010
)
Social Work and Social Development: Theories and Skills for Developmental Practice
 ,
New York
,
Oxford University Press
.
Mikton
C.
Butchart
A.
(
2009
)
‘Child maltreatment prevention: A systematic review of reviews’
,
Bulletin of the World Health Organization
 ,
87
(
5
), pp.
353
61
.
Morrison
T.
(
2010
) ‘Assessing parental motivation for change’, in
Horwath
J.
(ed.),
The Child’s World: The Comprehensive Guide to Assessing Children in Need
 ,
London
,
Jessica Kingsley Publishers
.
Munro
E.
(
2011
)
The Munro Review of Child Protection: Final Report: A Child-Centred Approach (Vol. 8062)
 ,
London
,
The Stationery Office
.
Munro
E.
Manful
E.
(
2012
) Safeguarding Children: A Comparison of England’s Data with that of Norway and the United States, Sheffield, Department for Education (DfE), available online at https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/197718/DFE-RB198_1_.pdf.
Norman
R. E.
Byambaa
M.
De
R.
Butchart
A.
Scott
J.
Vos
T.
(
2012
)
‘The long-term health consequences of child physical abuse, emotional abuse, and neglect: A systematic review and meta-analysis’
,
PLOS Medicine
 ,
9
(
11
), pp.
1
32
.
Office for National Statistics
(
2015
) Annual Mid-Year Population Estimates, 2014, available online at https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/datasets/midyearpopulationestimatesexplorabledatasets.
Ofsted
(
2015
) Early Help: Whose Responsibility? Manchester, Ofsted, available online at https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/410378/Early_help_whose_responsibility.pdf.
Parton
N.
(
2014
)
‘Social work, child protection and politics: Some critical and constructive reflections’
,
British Journal of Social Work
 ,
44
(
7
), pp.
2042
56
.
Peckover
S.
(
2014
)
‘Domestic abuse, safeguarding children and public health: Towards an analysis of discursive forms and surveillant techniques in contemporary UK policy and practice’
,
British Journal of Social Work
 ,
44
(
7
), pp.
1770
87
.
Platt
D.
Turney
D.
(
2014
)
‘Making threshold decisions in child protection: A conceptual analysis’
,
British Journal of Social Work
 ,
44
(
6
), pp.
1472
90
.
Scott
D.
(
2014
)
‘Children in Australia: Harms and hopes’
,
Family Matters
 ,
96
, pp.
14
22
.
Sidebotham
P.
Heron
J.
Golding
J.
(
2002
)
‘Child maltreatment in the “children of the nineties”: Deprivation, class, and social networks in a UK sample’
,
Child Abuse and Neglect
 ,
26
(
12
), pp.
1243
59
.
Thorpe
D.
Denman
G.
Regan
S.
(
2011
)
RIEP and ACDS Funded Safeguarding and Promoting Welfare Research Project
 ,
Kendal
,
Practice Research Overbeck
.
Thorpe
D.
Regan
S.
Mason
C.
May-Chahal
C.
(
2012
)
‘Making a case for common assessment framework responses to concerns about children’
,
Social Work and Social Sciences Review
 ,
12
(
3
), pp.
40
56
.
Trocmé
N.
Kyte
A.
Sinha
V.
Fallon
B.
(
2014
)
‘Urgent protection versus chronic need: Clarifying the dual mandate of child welfare services across Canada’
,
Social Sciences
 ,
3
(
3
), pp.
483
98
.
Wastell
D.
White
S.
(
2012
)
‘Blinded by neuroscience: Social policy, the family and the infant brain’
,
Families, Relationships and Societies
 ,
1
(
3
), pp.
397
414
.