Learning from death

Child
care professionals must avoid generalising about the circumstances leading to
the death or serious injury of a child, caution Ruth Sinclair and Roger
Bullock.

Whenever
a child dies through abuse or neglect it is important that lessons are learned.
For all the professionals and agencies involved, there is a great desire to
consider, as Lord Laming has in the case of Victoria Climbie, "how such an
event may, as far as possible, be avoided in the future"¹.

The treatment of Victoria was both tragic and
scandalous, and it seems right that there has been a public inquiry into the
circumstances surrounding her death. But it is also important that the learning
from such deaths is not treated in isolation, but is placed within the wider
body of professional or child care knowledge.

Public inquiries into child deaths are rare.
A more usual response is for the local area child protection committee (ACPC)
to instigate a serious case review (formerly known as a part 8 review) as
required by Working Together to Safeguard Children². Serious case reviews (SCRs) should be
undertaken whenever a child dies or is seriously injured and abuse or neglect
is known or suspected to be a factor. Their purpose is to establish whether
there are any lessons to be learned about the way in which local professionals
and agencies work together; to identify what those lessons are and how they
will be acted on; and to improve inter-agency collaboration to safeguard
children better.

The Department of Health estimates that in
England around 90 child deaths are the subject of SCRs each year. Each review
will contain important messages and an action plan for improving policy and
practice locally. But reviews are also an important source of information for
more general learning. Working Together requires the DoH to commission
an overview report of SCRs at least every two years, and we were commissioned
to undertake an overview, which was published recently³.

Forty SCRs were randomly selected, 31 of
which followed a child’s death and nine a serious injury: 20 were undertaken
under the old guidance⁴ and 20 under the Working Together guidance.
We used a comprehensive framework to analyse the reports, which included all
factors identified in previous child protection inquiries and research.

In addition, interviews were sought with the
report author and the chairperson of the ACPC in half the cases. Our study
findings covered the background characteristics of children who were the
subject of a SCR and the services they received, indications of the main
lessons to be learned from the cases, the operation of the SCR process, and the
impact of the new guidance.

There were some common features in the
children’s circumstances, including poor standards of care, emotional neglect,
domestic violence and mental health problems. However, there was a variety of
situations and this highlighted the need to locate individual cases in a wider
evidence base if we are to learn lessons. Also, because SCRs are a relatively
uncommon experience for many of those concerned, there was a tendency to
over-generalise from individual cases. This is illustrated by a comment made by
a senior manager: "It’s young mothers, who are depressed, and simply
cannot cope with their babies, in poor living circumstances, especially when
their situation is compounded by a violent partner."

In fact, our findings do not bear out this
generalisation. Only nine of the 40 main carers were under 21 when the child
was born; fewer than half (18 out of 40) had mental health problems; fewer than
half the children (19) were babies under 12 months; for 16 children no concerns
about their welfare had ever been expressed; and in 23 cases there was no
significant poverty or accommodation problems – although 22 out of 31 current
partners were violent.

As well as this variety, there was also great
diversity in the incidents that resulted in death or serious injury. In some
cases it seemed that the incident appeared "out of the blue", while
in others it occurred in a context of chronic low level need and occasionally
it arose in situations where it was "waiting to happen". We
classified these incidents into groups (see below).

The very different nature of the cases in
these groups emphasises the need for a wide range of responses related to the
particular attributes of the case. Here too, however, there is a tendency for
stereotyping to apply, as this comment from another respondent shows, with
actual findings interspersed in brackets:

"It’s usually social workers who are in
the front line (12 of the 40 children were not known to social services), and
despite their continual contact with families (only 12 were open social
services cases) so often miss the warning signs (this applied to 23 of the 40
cases) that should have alerted them to the possibility of a tragedy (four
cases were high risk or high priority)."

The involvement of social services was low.
Twelve cases were unknown to them, but in 24 cases concerns had been expressed
– more than once in 20 cases – but only 16 of these referrals led to a strategy
discussion, and only nine to a child protection inquiry under section 117 of
the Children Act 1989. At the time of the incident only six children were on
the child protection register. In 17 cases social services were criticised for
not undertaking assessments, and in six cases assessments had been completed
but not acted on.

Interestingly, health rather than social
services had most involvement with the 40 children studied. At the time of the
incident, 18 were open cases to health visitors, and five to other health
professionals. GPs were involved with more children and families than other
professionals.

Despite a considerable range of involvement,
in only 12 cases did agencies consider the families to have a high level of
need for services. Only four cases were seen as high priority, or as situations
where the child was at high risk of significant harm.

The reports identified deficiencies in all
agencies. The most common causes of concern were inadequate sharing of
information, poor assessment processes, ineffective decision making, lack of
inter-agency working, poor recording of information and lack of information on
significant males. Generally, however, practice was seen as poor rather than
grossly neglectful or incompetent, and in most cases the death or injury was
seen as unpredictable and largely unpreventable, even if the vulnerability of
some children was beyond doubt.

While any overview of SCRs provides evidence
about factors associated with child abuse, they have limited predictive value.
This is because it rarely has a single cause, but is the result of a chain of
effects that in particular circumstances leads to harm. Hence the application
of factors known to be associated with child abuse identifies high numbers of
likely abusers, most of whom do not abuse their children, known in research
parlance as "false positives". The value of SCRs therefore lies in
what they can tell us about managing the risks that are identified.

We developed some broader conclusions for the
enhancement of child protection. In brief, these are: the importance of
establishing through research good epidemiological and clinical evidence on
factors associated with children suffering significant harm, better
understanding of the process of change to implement effective public services,
and the development of practice tools to improve decision-making and practice
consistency.

Incidents resulting in death or serious injury

Number
of cases

Accidental/natural
cause of death but possible neglect                10

Known
significant protection risks or long-term neglect              6

Baby
"battered" by father/stepfather                                          5

Teenagers
living in chaotic circumstances                                   4

Murder
by mentally ill father/stepfather(one-off incident)            4

No
known protection risks but suspicious death/injury               4

Murder
by mentally ill mother (one-off incident)                         3

Dramatic
change in parenting following arrival of new male         2

Concealed
pregnancy/abandonment                                          1

Fabricated
or induced illness                                                     1

Ruth
Sinclair is director of research at the National Children’s Bureau, and Roger
Bullock is professor at the Dartington Social Research Unit

References

¹
Lord Laming; Opening Statement to the Inquiry into the Death of Victoria
Climbie, May 2001

²
Department of Health, Working Together to Safeguard Children: A Guide to
Inter-agency Working to Safeguard and Promote the Welfare of Children, DoH,
Home Office and DfES, 1999

³
R Sinclair and R Bullock, Learning from Past Experience: A review of Serious
Case Reviews, 2002, www.doh.gov.uk/qualityprotects
 

⁴
Department of Health, Department of Education and Science and Welsh Office, Working
Together under the Children Act 1989, HMSO, 1991