UK Family Law Reform

Free information index

The 228 child deaths they didn't want to tell you about

The death of Baby P appalled the nation. But after a six-month investigation, Live reveals an even greater tragedy: the devastating hidden toll of young children who died in the care of their families despite social services being alerted.

To complete this, the first-ever study of all 150 child-protection authorities in England, we used Freedom of Information rights to demand files in the face of cover-up tactics by agencies all the way up to Children's Secretary Ed Balls.

On the eve of yet another review into child protection, Live publishes the full damning portrait of a nation incapable of protecting its most vulnerable members

By David Rose Research by GEORGE ARBUTHNOTT 14th February 2009

It doesn't take long to tell the story of Hylene Essilfie's life, because it lasted only six months.

The immediate cause of its ending, on April 25, 2007, was that she brought up some of her food. In their flat in the east London suburb of Ilford, Hylene's mother, Faustina Osei-Agyapong, 21, had asked the baby's father, Francis Essilfie, 29, to give the baby rice pudding.

Instead, he fed her yoghurt, and when he tried to wind her, she was sick. Francis, a supermarket shelf-stacker, held Hylene upside down by her ankles. He screamed at Faustina: 'You two girls are something else! You are stressing me out!'

He strode across the room and punched his partner's face. Hylene was yelling, so Francis slapped her mouth. Then, boasting that he 'didn't give a damn', he hurled the baby, head-first, to the floor. She landed with a thud. 'Look at what I've done', Francis shouted accusingly. 'You let me kill my own baby.'

In fact, although her skull was fractured, Hylene was still alive. Faustina held her stricken daughter while she gasped for breath and dribbled blood, her head beginning to swell. She begged Francis to call an ambulance, but he refused, saying that if he did, the police would arrest him for murder. He got dressed, grabbed Faustina's mobile phone, and left.

Outside the flat, he finally phoned for an ambulance before boarding a bus for central London. Hylene was rushed to hospital, where, as prosecutor Crispin Aylett told the Old Bailey last April, an 'army of experts' battled to save her life. It was too late. According to the post mortem, Hylene's injuries were so severe that they were on a level normally found in car-crash victims. Convicted of murder, Francis was jailed for life.

After his trial, Redbridge Council, the London borough that covers Ilford, published the 'executive summary' of a Serious Case Review (SCR) into Hylene's death, a study by an independent expert into the way that the agencies responsible for child protection had dealt with her family.

92 - percentage of cases where children's services failed the victim

It reveals that on December 18, 2006, when Hylene was not quite two months old, Faustina went to the police in Tower Hamlets, where the family was then living. She told them Francis had been beating her since the start of her pregnancy. Once he had pointed a knife at her stomach and threatened to kill her unborn baby. She said that when Hylene cried, Francis would pinch the tiny baby to try to make her stop.

After Faustina made her statement, Francis was arrested, but merely cautioned and released. Moreover, the SCR summary says, 'the pinching of Hylene was not recorded as a crime'. The police did discuss the case with staff from Tower Hamlets children's services (as the social services that deal with children are now called).

But although research has shown that men who abuse their partners are likely to be violent towards their children, 'no child protection action was taken by either agency'. A social worker was asked to conduct an assessment, but according to the SCR, he did nothing at all.

On January 16 there was another confrontation, and Faustina called the police again. Afterwards she fled, managing to get herself and her baby re-housed in Ilford. The Tower Hamlets social worker duly passed their file to Redbridge. Having considered it, children's services there deemed the case needed 'no further action'. For a few weeks, Francis left Hylene and Faustina alone, but by February 26 he was accompany-ing them to the doctor's, where a nurse was so concerned about his 'controlling behaviour' that she insisted on speaking to Faustina alone, advising her how to get help for dealing with domestic violence.

Later that day, Faustina called the police yet again, saying Francis was refusing to leave the flat and threatening to abduct Hylene. The police alerted Redbridge children's services. As before, their staff failed to make a single phone call, much less pay a visit. Once again, they decided to take 'no further action'.

The SCR discloses that if the police had consulted their own records, alarm bells should have rung more loudly: they contained details of 'six previous incidents or episodes involving alleged domestic violence by Mr Essilfie with a discernible pattern including serious assaults on young, pregnant women'. There should, it adds, 'have been a clearer focus by police and children's services on the child'.

But the SCR's conclusion seems strangely at odds with its contents. There were, it says, lessons to be learned, especially about the need for agencies to share information. But overall, Hylene's murder 'could not have been anticipated'. Asked by Live to discuss the case, Redbridge declined.

One death every week

Fourteen weeks after Hylene's barely noticed death, another child was murdered a few miles away, in the north London borough of Haringey. Last November, when the cruelty inflicted on Baby P (as he must be called for legal reasons) became public, together with the total incompetence of the system that should have protected him, it triggered national outrage.

Baby P's mother, her boyfriend and another man had tortured him for eight months, a period in which he was seen by health and social workers 60 times.

Vainly protesting that an official inspection had awarded her department the maximum 'three star' rating, Sharon Shoesmith, Haringey's Director of Children's Services, was ignominiously sacked by Ed Balls, the Children's Secretary. Balls claimed that the case was 'exceptional'. But as he admitted, its horror was intensified by the fact that it happened in Haringey - the borough where, in April 2000, Victoria Climbie died aged eight from a combination of hypothermia, starvation and 128 separate injuries, inflicted by her guardians over many months.

Like Baby P, the Climbie case was a cause celebre. After a public inquiry chaired by Lord Laming, former Chief Inspector of Social Services, it led to a series of radical new policies, enshrined in the 2004 Children Act, which were supposed to ensure that in future, children would be safer than ever before.

Far from protecting children, however, a Live investigation shows that these policies have fundamentally failed. The annual numbers of 'other Baby Ps' - the heartbreaking cases like Hylene Essilfie's, of children under six who die through abuse and neglect but who do not make national headlines - have not declined at all, and may be increasing. In England alone, they are running at almost one dead child each week.

Our investigation has identified a shocking 228 children who have died at the hands of their parents or people known to the child in the five years since the Government introduced its Children Act. We know that there have been many more, although the authorities refuse to disclose details.

A system in crisis

Only a small proportion of these cases receive publicity, usually when the injuries have led to murder trials. Baby P aroused so much media attention because his abuse and sustained injuries were so extreme. The abuse of the other Baby Ps, such as Hylene, is less headline-grabbing but equally heart-wrenchingly cruel.

This cruelty is supposed to be spotted by the authorities. The 228 other Baby Ps were supposed to have been protected by the Labour Government, which put the policy 'Every Child Matters' at the heart of its government.

As part of our investigation, Live obtained details of 108 Serious Case Reviews into such child deaths, more than 40 per cent of the total undertaken in the past five years. This is by far the biggest sample ever seen by journalists. The reviews reveal that the same systemic failings on which Lord Laming blamed Climbie's death are still happening, and if anything, have got worse.

'Children's services are like a computer infected with several viruses,' says Professor Michael Preston-Shoot, who spent years as a social worker after qualifying in 1976 and now, as Dean of Health and Social Sciences at Bedfordshire University, is one of the field's leading experts.

'The viruses are poor management and leadership, a shortage of resources, political interference and the sheer irrationality of much of what the Government has put in place. Social workers find themselves in an environment that does not facilitate good practice.'

The only effective 'anti-viruses' are whistleblowers within the system, Preston-Shoot says. But as he points out, when Haringey social worker Nevres Kemal tried to blow the whistle over Baby P, she became the target of a witchhunt - investigated over bogus allegations that she had threatened another child.

'There are people who will kill their children: that's a hard fact we have to deal with,' says Nushra Mansuri, a child protection specialist who is in constant contact with colleagues across the country as a professional officer for the British Association of Social Workers.

'But the way the system works now may be making such deaths more likely. When you've been in social work a long time, you learn there are certain words you shouldn't use. "Crisis" used to be one of them. That's no longer true. Child protection is in crisis, and morale is lower than I've known it since I started working almost 20 years ago.'

'This is cover-up territory'

The shocking state of affairs is partly a result of the total lack of openness about what is happening. As we found in our investigation, one of the problems bedevilling any attempt to investigate child protection is an extraordinary level of secrecy. This makes even the most basic question - how many children are murdered by people they know or die from neglect each year? - difficult to answer.

Before Baby P, ministers often claimed that numbers of child abuse and homicide cases were steadily falling, citing World Health Organisation tables as evidence that Britain's record is among the best in the world. Indeed, the latest WHO figures say that in 2006 only 11 children aged 0-14 died through homicide in the entire UK - just over one child per million.

These statistics are supplied to the WHO by the Department of Health and, it can be safely said, are worthless. A more reliable source is the Home Office, which publishes figures for homicides for each age group recorded by police. To take only children aged 0-4, there were 32 such killings in England and Wales in 2005-6, and 41 in 2006-7. For 2007-8 the figure, published this month, was 45 - a level 50 per cent higher than those recorded annually through most of the Eighties.

However, in terms of assessing how many children die at the hands of their parents or carers, even these figures give only a partial picture. They do not include deaths by neglect, or cases which, while not obviously murder, were not exactly accidents - such as a parent who goes to bed drunk or high on drugs and rolls on to a child, smothering it.

7 - Authorities with most SCRs: Northamptonshire, Kent, Doncaster, Hampshire, Peterborough, Surrey, Southwark

The best statistics ought to be the number of SCRs carried out by each local authority, supposedly commissioned each time 'a child dies and abuse or neglect is known or suspected to be a factor in the death'. Full SCR reports are confidential, but summaries are meant to be published, though usually the subjects are anonymous.

On July 1 last year, John Hemming, the Liberal Democrat MP for Birmingham Yardley, wrote under the Freedom of Information Act to all local authorities in England that deal with child protection. He wanted to know the number of child death SCRs commissioned by the authorities, along with the ages of the children and the dates of their deaths. Hemming's findings - which he has shared with us exclusively and form part of our investigation- are important.

But they are also incomplete, because 21 councils, including huge authorities such as Birmingham, Manchester and Sheffield, refused to cooperate. Hemming soon discovered that the reason was a directive from Children's Secretary Ed Balls.

The non-cooperating authorities claimed that child deaths were not covered by the Freedom of Information Act, adding that they had been told by Balls's department that it would be replying to Hemming on their behalf. When it came, the department's reply was far from satisfactory. It said that, under its own rules, 'we cannot disclose information on authorities where there have been fewer than six cases'.

Earlier this month, Balls partially relented, sending Hemming SCR figures for the financial year 2007-8, including those from 'refusenik' councils. The total involving death was 89. A further 60 SCRs involving serious abuse but where death didn't occur were also carried out.

But the list did not include the victims' ages, gender or dates of death. Some will have been teenagers, who might have died through suicide or gang violence. According to Balls, 'providing such details... could prejudice the interests of children and their families'. He did not specify how.

Why the secrecy over something so fundamental? Hemming says: 'This is cover-up territory. They are avoiding telling the truth because they know the truth is embarrassing. We have one of the world's most intrusive systems for protecting children, and it fails.'

Beverley Hughes, the Children's Minister, in an apparent disagreement with her boss Balls, told Live that Hemming should have been given the full figures, including ages, as 'this is not sensitive information'.

Discovering the shocking truth

It may have gaps, but Hemming's research is compelling. Leaving out the 21 refuseniks, it reveals that there were 182 SCRs into the deaths of children aged 0-5 in England caused by confirmed or suspected negligence and homicide from the end of 2003 until the middle of 2008, plus others where the age was not specified.

The annual totals make uncomfortable reading: 37 in 2004, 36 in 2005, 47 in 2006 and 49 in 2007. Including the refuseniks would boost these numbers considerably: we have established that Birmingham, for example, has commissioned at least eight SCRs since 2004.

Attempting to dig deeper, we also approached all the authorities requesting copies of SCRs. We also trawled local newspaper reports for child deaths to get a more detailed picture of the scale of the problem. Combining our research with Hemming's, we have arrived at a figure that is the closest disclosure of child deaths under the age of six where parents or people known to them, such as step-parents or a mother's new partner, have been involved - and where social services have been alerted to the child. We have determined that in the five years since the Government introduced its Children Act there have been a shocking 228 such deaths. The majority - 60 per cent - were under the age of one.

Even this figure is not complete, as there are still a handful of councils that have refused to provide any information to either John Hemming or Live, but it provides the closest estimate ever produced. It is reasonable to suspect our figure is incomplete by about ten per cent, suggesting the number of children under six killed at the hands of their parents in just England could be at least 250 in the past five years.

The only year where Ed Balls has provided the figures broken down by age (but not by authority) is 2007. This shows that there were 51 SCRs carried out involving death of a child under six, a figure that is totally in tune with our estimates.

Failing the child

In our approach to the local authorities we also asked them to supply the SCR summaries to which these figures relate. Many councils refused to provide them. But the 108 summaries that we obtained reveal patterns that a Government that is committed to improving child protection should find deeply worrying.

The Government's response to Lord Laming's inquiry into the Climbie murder began with 'Every Child Matters', a green paper issued in 2003.

It said Victoria's death had highlighted 'problems of long standing'. The 'common thread' linking the failure to prevent hers and other child murders were 'poor coordination (between different services); a failure to share information; the absence of anyone with a strong sense of accountability; and frontline workers trying to cope with staff vacancies, poor management and a lack of effective training'. These failings, the paper said, were going to be put right.

Our investigation shows those words could easily be written today. Just eight of the SCRs we obtained made no criticism of children's services - just eight per cent of the total. That means in a shocking 92 per cent of the cases we have looked at, children's services have failed the child to some degree.

Lessons are clearly not being learned. Of the 100 that were critical, a quarter cited poor training as contributing to deaths, and a further quarter a failure to keep proper records - problems that had passed unnoticed by children's services management.

According to Laming's report, 'improvements to the way information is exchanged within and between agencies are imperative if children are to be adequately safeguarded'.

Often, there was a 'huge chronology' that only emerged when a child was already dead: if only different agencies such as the police, GPs and hospitals had shared their concerns with social workers, their life might have been saved.

60 - percentage of children who died before their first birthday

The Government claims that the 2004 Act solved this problem by imposing a 'duty to cooperate to improve (children's) wellbeing' and creating a vast, £250 million database, ContactPoint, which will contain children's details from the different agencies. This started to come online only last month.

Yet of our 100 critical SCR summaries, a staggering 75, three-quarters of the total, said poor coordination and information sharing between different services were significant factors in children's deaths, both within the same and between different localities.

Some of the most shocking SCR summaries echoed the Hylene Essilfie case, where police and children's services remained oblivious to the killer's violence towards previous victims.

Consider, for example, the murder of Deraye Lewis, three, who was beaten to death in Milton Keynes by his mother's boyfriend, Nicholas Halling, in January 2005. Like Baby P, he had been tortured for months: in the words of the judge who sentenced Halling to life, the boy had been treated with 'contempt, hostility and violence' for at least a year - burnt with cigarettes, beaten about the head and kicked in the stomach.

Halling, who had been to prison numerous times, moved in with Deraye's mother, Donna, when the boy was 16 months old and they were living in Dunstable, Bedfordshire.

According to the SCR obtained by Live: 'He was known to social services because he had been involved, two years previously, with a family where the children were observed to be terrified... social services and the police in Bedfordshire had taken the decision in 2002 that dog handlers should be used if the children needed to be removed and that the police should always be present during visits if Halling was in the household.'

That family's mother had always covered up for him, but only, records said, because she was also terrified. The police national computer had Halling flagged as 'dangerous'.

Yet when social workers first visited Deraye in July 2004, alerted by neighbours that Halling was physically and racially abusing him, they accepted Donna's assurances that everything was fine and closed the file.

The social workers were told of escalating abuse on three further occasions, and when they visited saw that Deraye was bruised, and once - like Baby P - had an untreated dog bite. The SCR says they were told by the neighbours that Halling 'had trained his dog to attack Deraye'. But 'the community paediatrician decided, with little evidence, his injuries were not deliberate,' and again the case was closed.

In September 2004, Donna and Deraye turned up at the social services office, both covered with cuts and bruises. She made a statement to police, saying Halling had attacked her when she tried to stop him beating her son. But it wasn't passed on to other agencies, and nothing was done to protect them.

Afterwards, she moved to Milton Keynes and applied for council housing. While she was waiting, she was still in harm's way: she and Deraye were staying with Halling's brother. However, the Bedfordshire agencies did not inform their Milton Keynes counterparts of any of the previous background.

The last chance to save Deraye's life came on December 31, when Donna dialled 999, telling the operator she was being attacked by her partner. In the middle of the call, the line went dead, but when the police arrived, she said the problem had been an exboyfriend who had left. She did not mention Halling's name, and unaware of any reason to be concerned, the officers took no action.

Next morning, Donna dialled 999 again. This time, it was to call for an ambulance, though by now, Deraye was dead.

Something to hide?

The Deraye SCR summary is unusual in that it is incredibly detailed. Other reports we received have been censored, a small number sanitised to the point where it is difficult to comprehend why things went wrong.

Some councils take this tendency to hide embarrassing details still further: after weeks of trying, when we finally obtained four summaries from Birmingham, they had been 'redacted' - huge chunks of text replaced with black ink. The council claimed this censorship had been done for legal reasons, citing its requirement to protect the identity of the child. In fact this is nonsense and something that Beverley Hughes has criticised.

It appears that Birmingham has something to hide. In the last three years it has carried out eight SCRs, the highest number of any authority. And Live has established that between 2000 and 2005 it carried out a further 12 SCRs, a total of 20 in eight years. Manchester, another council that refused to cooperate, also carried out 12 SCRs between 2000 and 2005.

Lord Laming saw SCRs as an essential link in a chain of accountability, and a way to avoid repeating mistakes. Indeed, our investigation suggests it is crucial that SCRs are published and distributed in full across the country.

Otherwise, how are social workers ever going to learn from past mistakes?

3/4 - Amount of SCRs citing poor communication as a factor in the death

Last year, a damning report by Ofsted found that 40 per cent of SCRs were 'inadequate' - despite their average cost of about £20,000. But something Ofsted did not point out is that the full SCR reports remain confidential to the local 'Safeguarding Children Boards' that commission them.

To services elsewhere that might face similar problems, they might as well not exist. The result is that the same errors are replicated, time and time again.

Ultimately, Deraye Lewis died because none of the agencies in Milton Keynes knew about the threat Halling posed. Almost identical conclusions are drawn by the SCR into the murder of Amaraye Bryan in Sheffield, who was killed at the age of 11 weeks by his father Courtney, on May 14, 2007, more than two years after Deraye's death.

It was only then that it emerged that he had assaulted two children almost the same age during previous relationships. One suffered severe brain damage and still needed full-time care as Bryan was convicted of murdering Amaraye in 2008. Bryan had also been suspected of seriously injuring two older children, one aged two, the other three.

However, these earlier cases took place in Nottingham. Even within that city, agencies had failed to identify that he was the common link between four families where children had been attacked. As the SCR puts it, 'as with most SCRs, inter-agency communication failings have been identified; but what is more unusual is a sequence of intra-agency communication failings'.

Needless to say, the authorities in Sheffield had no idea of the potential threat that Bryan posed to Amaraye: 'The significant information that (he) posed a threat to children existed but was not recognised or utilised.'

Hence, then, the critical lost opportunity when Amaraye might have been saved. Eleven days before he died, he was seen to have suffered bruising and bleeding in his eyes after being left alone with Bryan. But according to the SCR, 'medical opinion at the time was that there was plausible explanation for this condition'.

According to the Government, the ContactPoint database is its key weapon to improve communication.

But the information it contains will be very limited, and will concern each child, not the people who may be looking after them. It is hard to see how it would have made any difference to Amaraye, because it does not contain any details of potential perpetrators. Even with the new database Sheffield would still have been unaware of Amaraye's vulnerability.

Too many forms, too few meetings

Aside from better communication, both Laming's report and 'Every Child Matters' stressed the need for social workers to spend more time getting to know families properly: the Government, as the green paper said, was determined to 'address bureaucracy and identify ways of freeing up time for face to face work'.

By such means, families who deceived children's services - such as Baby P's - would be more easily detected, and killers like Halling and Bryan less likely to get away with intimidating their partners into silence.

Yet bureaucracy, in the shape of new, intensely demanding computerised forms, has got much worse, not better. A research team from Lancaster University has just finished compiling a devastating study based on two years' observation in five very different children's services departments, which was quietly published earlier this month in the British Journal of Social Work.

Among the report's authors was Professor Susan White, newly appointed to a Government task force charged with looking into child protection. The report concludes that the Government's reforms, far from improving protection for children, have actually created 'acute challenges to safe practice'.

The Government, it says, has become obsessed with computerised forms and targets. Computerised methods - the 'common assessment framework' and the 'integrated children's system' - are at the heart of the Government's post-Laming programme. It was because they had been successfully implemented that Haringey's Shoesmith made her disastrous claims about three-star ratings after Baby P.

But among their many damaging aspects are requirements to make an initial decision whether to pursue a case within 24 hours, and a more considered assessment of those not then rejected within seven days - creating, as the Lancaster study says, 'a rapid, but not necessarily reliable, response'. High absentee and sickness rates - other studies estimate that children's services departments are, on average, at least 20 per cent understaffed, with some at just 40 per cent strength - exacerbate this pressure.

Meanwhile, the computers punish those who file their forms late with flashing 'traffic lights,' and in some departments management 'print out weekly graphs of levels of attainment in meeting targets, alongside tables exposing individual failures'.

The inevitable consequence was that 'well-intentioned but very busy workers' found themselves with little time to meet families at all, much less weigh their decisions: 'There was a tendency to abort an assessment whenever the opportunity arose.'

Twice, for instance, Hylene Essilfie's case was marked 'no further action'.

Laming had wanted to 'reduce the distance' between social workers and those they served, but instead this had increased.

'The real tragedy', concludes the report, 'is that, in busy departments, demands to support families will be routinely subordinated to pressures to maintain "workflow"'.

Dr Karen Broadhurst, the study's lead author, told us that the Government has created 'a difficult, if not a lethal cocktail,' with social workers not only placed under impossible pressures but finding it impossible to voice their concerns for fear of being victimised.

Like Nushra Mansuri, she says children's social work 'is in a state of real crisis'. With the recession likely to increase the risks to children as families slide into poverty, the time has come 'for some very frank and honest discussions between policymakers and practitioners'.

According to Mansuri, some social workers spend no time with children at all because of the system's bureaucratic demands. An average estimate from the academics and professionals we spoke to is that most spend 80 per cent of their time at their computers. The social workers' union, Unison, backed up their claims, while everybody involved in the profession that we spoke to complained about the appalling bureaucracy and chronic levels of electronic form-filling that preoccupies their time.

A new Children Act?

Belatedly, in the wake of Baby P, the Government has started to recognise that things are going wrong. However Hughes rejects the claim that child protection is in crisis.

Later this month, Lord Laming will issue a 'review' - much less than a fullscale inquiry - of the way the system is working, while Ed Balls's department has set up its child protection 'task force'.

This, says Hughes will 'get under the skin of the practitioners', and if it finds problems actions will be taken. However, she added that she rejects the idea that social workers do not need to make careful records in order to spend almost all their time 'having tea in people's front rooms.'

Some of the improvements that could be made would be small-scale. One example comes from Staffordshire, where Rachel Bramble, a social worker since 1982, works not from a children's services office but at Wolgarston School in Penkridge. There, she gets to know children and their families by a natural process, making herself available to give help and advice and thus to notice problems early on.

80 - percentage of time social workers spend at their PCs

'You can see how it works,' she says. 'It could start with a kid with a bruise.' If social workers were based in schools and GPs' surgeries widely, Bramble says, many of the problems of inter-agency cooperation would not exist. Yet her initiative has no official backing, and depends on the school funding it from its own budget.

Other changes are more strategic. John Fox, the Laming inquiry's police advisor, says that one of the system's central problems is that Laming's main recommendations were never properly implemented, especially its call for stronger accountability. 'Laming wanted accountability to run from top to bottom,' Fox says. 'That meant that the new local

Safeguarding Children's Boards should have been chaired by councils' chief executives, the most powerful local official and the person who controls the cheque book. Instead - as in Haringey - they're often chaired by the director of children's services, creating an immediate conflict of interest.'

Fox reserves his deepest contempt for the phrase found in so many SCRs, Hylene Essilfie's included - that a child's death 'could not have been anticipated'. 'That's a very questionable claim,' Fox says.

'Maybe the death couldn't have been foreseen given the current level of inadequate practice. But that's a different thing. If there are failings, you should be saying so.'

Ultimately, there is an argument for a new Children Act - one that finally puts the principles identified by Laming into effect.

It is difficult to see how children can be better safeguarded without freeing social workers from bureaucratic tyranny, encouraging them to be more active in the community, establishing a database that includes potential perpetrators as well as vulnerable children and re-establishing accountability by the publishing of full Serious Case Reviews.

Only then could the Government be said to be serious about ensuring that every child really does matter.


Summaries of all Serious Case Reviews are supposed to be made public, or at least available upon request. By sharing SCRs, social workers and other services can learn from past mistakes.

However in Birmingham, where it has been reported that eight children known to social services have died in just three years, local authority officials at first refused to send through any of the SCRs. After weeks of pressing, they finally released just four documents, yet each of them was seriously redacted (blacked out), rendering the exercise almost irrelevant.


How did we arrive at our figures? In July 2008, John Hemming, the Liberal Democrat MP for Birmingham Yardley, wrote to every local authority requesting information about every Serious Case Review (SCR) that involved the death of a child known to social services. He asked for the date of death, the child's initials and the age of the child at death.

All but 21 local authorities responded to his request, but not every one provided full details. So we then contacted each of the authorities to request actual copies of the Serious Case Reviews.

In the meantime, Children's Secretary Ed Balls provided Hemming with a list of SCRs involving death that were carried out between 2007 and 2008. We cross-referenced this list with Hemming's original list to ensure that our numbers were as up to date as possible. Furthermore, we checked our figures against local newspaper reports of child deaths.

From all this data, we compiled a master table - an extract is shown above. Because the process was extremely complicated, we used a colourcoded system to signify where we had got to in our investigation. For example, a red entry means we were unable to obtain an SCR. But even after these checks, our figures are not complete, as we still do not have full data from every authority.

It is important to note that our investigation was only concerned with the deaths of children under six since the end of 2003. Where possible, we struck out every SCR that was irrelevant.


Lord Laming's 2003 inquiry report into the death of Victoria Climbie was supposed to transform child protection for the better. Here are some of its key recommendations - and what has actually happened:


1. Social workers should be freed from time-consuming bureaucracy in order to spend much more time with children and their families.

2. A new child-protection agency be formed to ensure policy is implemented at a local level and that lessons are shared when children are killed.

3. Local agencies must always work together and share information on children and possible perpetrators.

4. Every child at risk should be given their own allocated social worker.

5. Accountability should be ensured by creating powerful local safeguarding children's boards, to be chaired by the council chief executive.

6. No child protection case should ever be closed until social workers have seen the child and their families.


1. Experts say social workers now spend 80 per cent of their time in front of computers, thanks to the demands of the Government's reforms.

2. The agency was never created.

3. About three-quarters of serious case reviews say the failure to share information was a problem.

4.Often four to five different workers will be involved in any given case.

5. Boards are often chaired by children's services directors, as was the case with Haringey's Sharon Shoesmith. It means they are reporting to themselves.

6. Many cases - like Hylene Essilfie's - are marked 'no further action' without further interviews with the parents in order to meet deadlines.


This is an example of the type of form that has to be filled in by social workers. This particular one is a labyrinthine eight-page 'initial assessment' that must be filed within seven days of a case being referred to social services by the police, doctor or hospital, for example. The social worker assigned to the case must then decide within seven days whether to take the case forward, or no further action is required.

Social workers, who maybe working on four or five such forms in a single week, say filling them in takes hours of time and is one of the reasons they are spending 80 per cent of their time in front of computers.

A newly published two-year study by Lancaster University has found that it is simply impossible to collect all the information required, and that as a result social workers not only fail to spend time with children and families but have to 'cut corners'. One social worker told Live: 'Sometimes, you're virtually making it up.' Failing to meet the deadline can lead to individuals being disciplined.

Click here to view the shocking figures of child deaths in the UK

Join us in campaigning for equality and justice.