House of Commons - Education and Skills

Select Committee on Education and Skills Ninth Report

6 Integrated processes

93. As part of measures to improve communication between professionals, and ultimately outcomes for children and young people, the DfES intends to implement IT-based, multi-agency index(es) containing basic details on all children in the country. This has been one of the most controversial aspects of the Every Child Matters reforms, and one that has consistently attracted strong criticism in both written and oral evidence.

94. The rationale for the index(es) is briefly summarised as follows—it will serve as a tool to:

help practitioners identify quickly a child with whom they have contact, and whether that child is getting the universal services (education, primary health care) to which he or she is entitled;

enable earlier identification of needs and earlier and more effective action to address them by providing a tool for practitioners to identify who else is involved with or has a concern about a child; and

encourage better communication and closer working between different professionals and practitioners.[71]

95. The failure of other recent Government-funded IT-based initiatives, some commissioned by the DfES, makes us more cautious than might otherwise be the case about these proposals. This Committee has recently inquired into two other projects—the UK e-University venture and Individual Learning Accounts. The first of these failed to achieve its ends and has consumed £53 million of public funds, while the second collapsed under suspicion of fraud and an overspend of £60 million. Consequently, the development of child indexes needs to be approached with the utmost caution, not least because it would be unreasonable to expect the taxpayer to bear the cost of another IT failure.

INITIAL DEVELOPMENT WORK ON THE INDEXES—CONTEXT

96. In 2002, six Local authority areas or groupings of areas (since expanded to ten) were selected to become Identification, Referral and Tracking Trailblazers, now re-branded as Information Sharing and Assessment Trailblazers (ISAs). Each was provided with £1 million to test and develop ways of improving co-ordination and information-sharing between agencies involved with children and young people. The work being carried out includes: clarification of how and when practitioners should share information; development of child indexes or databases; development of a common assessment framework (CAF) and a range of activities to support better information sharing within and across children's services. A team from the Royal Holloway College, University of London, led by Prof. Hedy Cleaver, were commissioned to analyse the initial development of Information Sharing and Assessment Trailblazers and analysis of the development of child indexes took place as part of this project. The researchers completed collecting evidence in August 2004, and their report was published in November 2004[72].

WILL CHILD INDEXES IMPROVE OUTCOMES FOR CHILDREN?

97. The development of child indexes is potentially extremely costly. We have received a number of different estimates of the resources—up to £1 billion pounds in one case—that are likely to be needed for initial development and maintenance. However, when we asked the Minister about development costs, we were told that the estimate of £1 billion was "absurd", and that the real figure would be in "the low hundreds" of millions of pounds[73].

98. Unarguably, several hundred million pounds remains a very significant amount of money. A key consideration must be whether it is justifiable to spend such a large sum on the development of child indexes when there are other very substantial calls on funding and, more importantly, other available ways of fostering improved communication between professionals—for example, through multi-agency training which can develop a culture of openness and trust between traditionally separate professionals. The evidence we have received has left us with doubts about whether investment in child indexes can currently be justified in terms of the contribution it is likely to make to improving outcomes for children.

99. Professor Cleaver told us that almost all the Trailblazers, at the end of the research period, had functioning index systems or had such systems in development and were preparing for imminent implementation. On the likely value of these systems, she has written that "outcomes for children will be improved if practitioners communicate and services are delivered in a co-ordinated way. A child index with details of how to contact other practitioners could aid this process but must not be seen as the whole solution".[74] Dr. Eileen Munro, London School of Economics told us categorically, however, that she did not think investment in improving information-sharing in general and indexes in particular was justifiable, especially in the light of other pressing demands on resources:

She went on to explain further:

100. It is clear that there are several specific and interrelated operational challenges which will need to be successfully overcome if there is to be any prospect of child indexes serving as the useful professional tool that the Government envisages. We address some of these specific challenges below, dealing in turn with: registering 'flags of concern' and contact with sensitive services; security; ensuring the accuracy of information held; and a source for the 'unique identifier'.

'FLAGS OF CONCERN' AND THE RECORDING OF SENSITIVE INFORMATION

101. The DfES has proposed that electronic 'flags of concern' should be attached where appropriate to children's files. These flags would serve as a means of alerting other practitioners, thus building up a more complete picture of the circumstances of the child. They would then serve as a trigger for contacting other professionals. It is also proposed that where children have contact with additional—often specialist—services the contact details of the practitioner should be placed on the file—again, enabling practitioners involved with the same child to contact each other.

102. These two aspects of the proposed indexes have been seen as inherently problematic by a very large number who have given evidence on the topic, including the Information Commissioner, Richard Thomas[77]. We were especially interested to hear that Professor Cleaver shared these concerns, based on her own analysis of trailblazer authorities. She told us:

103. The DfES responded to these concerns—which have been expressed over a long period—by issuing a consultation on this aspect of the database. The Government response to this consultation has not yet been released, so we are unable to be sure at this stage what plans are being made in this area, although the Minister did tell us that "the key to this [the child indexes in general] is simplicity, and I am determined to have that."[79] We take this—and the consultation—as positive signs that concerns over the adding of flags of concern and practitioner details are being taken extremely seriously, but nevertheless, we think that the Government should aim to clarify their intentions in this area as soon as is feasible.

104. A further concern here is that the use of a flag should indicate that a professional has a 'cause for concern', but what constitutes such a concern is left undefined and seen as a matter for professional judgement. Since professional groups work to diverse definitions and interpretations, there is unlikely to be a consensus on meaning. This could lead to an unnecessary volume of information on the database and a consequent diversion of professional energies from real concerns. The information Commissioner shares this concern, noting in his written evidence that practitioners may simply add a flag of concern as a defensive measure to ensure they are legally covered.[80] This in turn could trigger superfluous work and may lead to undue intervention in a child or young person's life. We feel there needs to be a much clearer and shared understanding of what constitutes a cause for concern and urge the DfES to clarify this matter before the indexes become operational.

SECURITY

105. The security of information on IT-based indexes has also been a major concern for those submitting evidence—and for us as a Committee. Inappropriate access and retrieval of information by those looking to harm children clearly risks undermining the ultimate aim of the indexes—to protect and safeguard children's and young people's welfare. Women's Aid, for example, told us that they were: "concerned that electronic databases specified in the current Children Bill could be used by abusers to track down women and children fleeing from domestic violence, and recommend that appropriate safeguards are introduced into these systems as a matter of urgency."[81]

106. In oral evidence, we were told by Professor Hedy Cleaver that research carried out by Trailblazer authorities with children themselves had revealed that many had serious concerns about the potential for data to serve, paradoxically, as a 'resource' for those looking to harm them. Professor Cleaver explained:

107. Evidence from the Information Sharing and Assessment Trailblazers further showed, however, that the issue of security is one that has been taken extremely seriously by those trialling systems, and a number of technical and other solutions have been piloted in different circumstances—including, for example, the necessity to enter a series of passwords and other personal information before being able to access records. While we would not contend that security presents an insurmountable problem, we would argue that existing publicly available analysis of child indexes does not demonstrate beyond doubt that workable solutions have been found.

MONITORING AND ENSURING ACCURACY OF INFORMATION ON THE DATABASES

108. Much of the evidence we have heard has stressed the importance of keeping information on the database accurate and up to date—otherwise its value as a source of information will be undermined. It is clear that potentially disastrous consequences could ensue should outdated information be retained on the system when circumstances change. The Information Commissioner outlined to us a case whereby details about a child had changed but these had not been updated on the records:

"a father was alleged to have been abusing his child. Another person was then prosecuted for that matter and the father was entirely innocent, and yet the records were not updated, and that was still there against that particular parent's name. It is fundamental not just to keep track of the names and addresses, but to keep track of the changes of circumstances as the various processes move forward.".[

109. The Information Commissioner went on to point out that the administrative burden of keeping indexes fully up to date should not be underestimated. The risk to children and their families posed by the potential for storage of false or outdated information should therefore be assessed as the Government moves to produce its business case on the indexes—and there should be clear, transparent explanations of how the administrative burdens of updating the system would be met.

UNIQUE IDENTIFYING NUMBER

110. A network of child indexes depends for operational integrity on the association of a unique identifier with each child's record, in order to ensure that data retrieved or entered relates to the appropriate child and that there is no confusion between, for example, children with the same name. Almost all of the evidence we have received on this subject—including that from the Department itself—has indicated that finding such a unique identifier is in practice proving to be extremely difficult. In oral evidence, Officials told us that that they were currently involved in discussions with Ministers and colleagues about the issue and were due to present a recommendation to the cross-departmental Committee of Ministers in April 2005. We appreciate that this is not something which can be subject to a 'quick fix'. It has to be a cause for concern that at this late stage in proceedings, a source for the unique identifier has not been decided upon.

111. Although the Minister sought to reassure us that she would not proceed to the commissioning stage until she was certain that indexes were "not going to be an IT disaster, but […] a good additional tool,"[84] we remain concerned about the development of child indexes and would urge the Government to proceed with the very utmost caution in this area. The Minister has confirmed that further research will be commissioned to examine the impact of indexes in Trailblazer areas, the findings of which would be used to inform the business case for implementation,[85] and we welcome this.

112. In the past, this committee has been concerned that crucial policy decisions are sometimes taken without sufficient research or evaluation of existing practice. In this case, the fundamental decision to go ahead with child indexes appears to have been taken before the activities of the Information Sharing and Assessment Trailblazers could be fully analysed.

113. We are not convinced that sufficient evidence currently exists to justify the commissioning of the proposed IT-based child indexes. We have significant reservations about whether this will represent the best use of resources and very significant concerns about critical issues such as security, confidentiality and access arrangements. We are concerned in particular that the current research evidence does not conclusively demonstrate that expenditure in this area is the best way of improving outcomes for children.

114. We welcome the news that further evaluative work on the impact of indexes in Trailblazer areas is now being planned, and that the results of this will be used to inform the business case for implementation. This research should analyse the comparative benefit of the indexes as a means of improving outcomes and other ways of improving information-sharing within and between professionals.

Common Assessment Framework

115. Along with child indexes, the development and implementation of the Common Assessment Framework (CAF) is a key means of achieving closer integration of services at the process level. It aims is to "provide a national, common process for early assessment to identify more accurately and speedily the additional needs of children and young people".[86] Specifically, the CAF aims to:

"provide an easy-to-use assessment of all the child's individual, family and community needs, which can be built up over time and, with consent, shared between Practitioners […] ;
improve the quality of referrals between agencies by making them more evidence-based;
help embed a common language about the needs of children and young people;
promote the appropriate sharing of information; and reduce the number and duration of different assessment processes which children and young people need to undergo."[87]

116. The Children Act 1989 defines two sets of circumstances where assessments and interventions should be carried out. On the one hand, Section 17 lays out the need for assessment where children are in need. Section 47 defines the duty to carry out assessment/ intervention where a child is thought to be in need of protection. The Common Assessment Framework has been widely perceived as a means to align different agency assessment procedures, but it can also be understood as a method of bringing together sections 17 and 47 of the Children Act 1989.

117. Information Sharing and Assessment Trailblazer authorities were charged with developing common assessment processes as part of their activities. Trailblazers' experience has fed into the design of the national Common Assessment Framework, which was issued in first draft form for consultation in August 2004 and was subsequently released as draft guidance subject to refinement in January 2005.[88] It is expected that the guidance will be issued in its final format by April 2006 so all Local Authorities can begin implementation.

118. While we have encountered very widespread support for the aims and objectives of the CAF, and the potential benefits it could provide, we are also aware of concerns about the lack of clarity surrounding its final form and the ways in which it will be used in practice. This is no doubt inevitable given that CAF is still in the final stages of development, and there is clearly a trade-off to be made between valuing consultation as a means of developing policy and the benefits of providing a clear sense of direction and certainty from the very outset. We have also been told of concerns about particular operational issues, many of which resonate with discussions about the implementation of other aspects of the Every Child Matters programme, including: pressures on the workforce; the cost of training staff in its use and the cost of implementation; ensuring take-up of the CAF; and the link between assessment and entitlement to assistance. In the main, these issues were also, unsurprisingly, raised by agencies who responded to the consultation carried out in late 2004.

Implications for workload and costs of staff training.

119. The introduction of the CAF has inevitably raised concern among some about the potential impact on the workload of those who will be expected to incorporate it into their professional practice. Additionally, training in the use of CAF will need to be provided for those staff that are expected to use it in day-to-day practice. The LGA comments on both of these issues:

120. In the draft guidance, the DfES proposes a number of ways of addressing the costs of training and potential impact on workload. With regard to the former, it proposes: a phased approach, whereby 'key' staff members in certain agencies are trained in its use and expected to deploy it; other staff will contact these designated members if they think an assessment needs to be carried out; previously, it has proposed integration of CAF into initial training programmes for professionals, and into Continuing Professional Development; and the production of centrally produced training materials which will be 'cascaded' downwards to the front line. Decisions about who shall receive training initially will be made at local level. These steps seem in general to be a rational response to the challenges and concerns identified by practitioners, but also give rise to questions of their own: Will, for example, the selection of only some staff for training mean that CAF will only be used with a small proportion of children who could have additional needs? Will local authorities find sufficient funds to train staff? How exactly will CAF become part of initial training programmes?

121. With regard to reducing pressures on workload, it is explained that CAF will ultimately lead to a rationalisation of existing assessments and thus reduce the amount of time spent on this activity. It is also proposed that research (discussed in more detail below) will be undertaken to examine the implications of CAF for staff in a variety of settings and that the findings will be used to inform implementation. The rationalisation of assessment through CAF again strikes us as an aspiration rather than a likelihood at this stage, and we will therefore look forward to the results of the research.

LINK BETWEEN ASSESSMENT AND ASSISTANCE

122. Responses to the DfES consultation raised concerns about the issue of how and whether being assessed through CAF would trigger the provision of services. Some respondents argued that:

"practitioners would have to be very careful about raising expectations of families by engaging them in the process of a common assessment only to find the needs identified cannot be met, or met within a relatively quick timescale. It was felt this experience could deter families, YP [young people] and children from engaging in the process and would undermine the trust in both the process and public or other services."[

123. This seems to us a crucial point, which has also been raised with us in relation to our inquiry. The Refugee Children's Consortium, for example, state that "Assessments should result in action, and the action assessed as needed should be recorded."[91] We would concur that achievement of the aims of CAF are only likely to be attained if there are sufficient resources in the system to actually provide the services for the need(s) identified.

124. Groups representing parents have pointed out that it is sometimes very difficult currently for parents to obtain access to the assessment that they need to 'trigger' assistance for their children. Currently, they argue, the withholding of assessment can be a gate-keeping measure employed by hard-stretched service providers, who have no other choice but to limit access to services in some way[92]. In its response to the consultation, the DfES indicated that:

not

This stance has been maintained in the draft guidance. While this is in keeping with the strong presumption toward devolved decision-making running throughout Every Child Matters, it remains to be seen whether devolving responsibility again to the front line on the issue of when and where the Common Assessment Framework should be used will be a productive way forward.

125. The DfES makes the following comments on how it is intending to progress:

126. It is essential that the design and implementation of the Common Assessment Framework takes place at a pace that allows informed development. The commitment to further testing and assessment before national rollout is therefore extremely welcome. While it is sensible that the assessments will examine the impacts of Common Assessment Framework on services, we would also hope that they take a broader view and examine the extent to which the Common Assessment Framework is leading to improved outcomes for children, young people and families.

71 Summary adapted from DfES Information Sharing and Assessment website, http://www.dfes.gov.uk/ISA/IndexProp/indexProp.cfm Back

72 Ibid Back

73 Qq 565, 566 Back

74 Department for Education and Skills, Cleaver et. al. 'Developing Information Sharing and Assessment Systems', pp. 68, 2004. Back

75 Q 322 Back

76 Q 326 Back

77 See EVCM 66. Here, the Information Commissioner lays out a detailed response to the ongoing consultation. Back

78 Q 322 Back

79 Q 543 Back

80 EVCM 66 Back

81 EVCM 34, para 2.4. Back

82 Q 364 Back