England and Wales High Court (Administrative Court) Decisions

IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
ADMINISTRATIVE COURT

		Leeds Combined Court Centre 1 Oxford Row, Leeds, LS1 3BG Judgment delivered at the Central Criminal Court
		23/7/2015

Mr Justice Simon:

Introduction

1. The issue before the Court is whether it is lawful for the Defendant to have a policy of retaining child protection records for a period of 35 years after a case has been closed. The Defendant contends that it is; the Claimant and the Information Commissioner, as an Interested Party, contend that it is not.
2. The policy in question is set out in tabular form and covers:

Child protection: investigations under section 47 of the Children Act 1989, child protection conference minutes, child protection plans and associated case recording.

3. The policy is specifically to retain the records for 35 years after the case is closed, unless the child is or becomes looked after (in which case the retention policy is 75 years from the date of birth) or adopted (in which case the retention period is 100 years from the date of the Adoption Order).
4. In child protection cases, the period starts from 'case closure', and the stated justification is expressed to be 'Executive Director's Delegated Decision, December 2014'.

The history of the action

5. The Judicial Review claim was issued in June 2014 on behalf of members of the Claimant's family and, by way of relief, sought the destruction of child protection information held by the Defendant, an order quashing the Defendant's policy and/or a declaration that it is unlawful. The Claimant felt that he and his family had been the subject of a significant injustice in their dealings with the Defendant, and that the Defendant held material about them as a result. He wanted all the data relating to the case to be destroyed. The Defendant agreed to this, and the particular facts of the Claimant's case (however painful they are to him and his family) are no longer relevant to the broader issue of principle which now arises.
6. In late November 2014, and in specific response to the claim, Stephen Corlett (a senior manager in a joint organisational unit of the Defendant and Northumbria Healthcare Foundation NHS Trust) was asked to advise the Defendant on the period for which records in child protection cases should be kept. Having consulted a number of identified individuals, he produced a report headed 'Retention Period for Child Protection Records' ('the Policy Report'). It is common ground that this is to be read as part of the Retention Policy. I have set out most of the material parts of the Policy Report as an annex to this judgment, and will refer to some of its provisions. It is clear from Mr Corlett's evidence that in preparing the Policy Report he had in mind the potential conflict between two important principles: minimizing intrusion into people's private lives and ensuring the safety of children (see §4.2 of the Policy Report). At §9 of his witness statement he observed:

A balance needs to be struck between these two considerations. It would be highly desirable for that balance to be arrived at nationally, with an opportunity for full public debate about the conflicting expectations placed on local authorities and other agencies, but no authoritative national guidance for social services authorities addressing this issue has to date been produced.

7. In a recorded decision, dated 22 December 2014, the Defendant's Executive Director accepted the recommendations in the Policy Report for the reasons given within it.
8. The Defendant also produced witness statements from Stephen Day, its Safeguarding Standards Manager and Principal Social Worker, and from Fiona Brown, one of its Senior Children's Services Managers. These provide further background information and the context in which the issues raised in the case can be considered.
9. It is convenient to record two points at this stage. First, although the Claimant relies on principles which are founded on the Common law, Article 8 of the Human Rights Act 1998 ('the HRA') and the Data Protection Act 1998 ('the DPA'), it is common ground that the requirements of each source of law is materially the same^[1]. Secondly, the parties are also agreed that it is for the Defendant to show that its Retention Policy is justified. Thus, although the Claimant and the Information Commissioner have argued that a retention period of longer than 6 years cannot be justified, they remind the Court that it is not for them to justify their argument as to what retention period would be lawful. Subject to this qualification, the Claimant's proposed policy was that information should be retained for 6 years after the child in question turned 18, unless the child (ex-child) requested earlier destruction or retention, or the Defendant was able to demonstrate a specific and material reason for its retention. I shall come to the Information Commissioner's proposed policy later in this judgment.
10. The present lack of authoritative national guidance or uniformity among local authorities is clear from the appendix to the Policy Report, which shows a wide variation of retention periods. Among those consulted, Devon has a policy which limited retention to the period up to the subject's 21st year, while Cambridgeshire, Newcastle, Norfolk, Staffordshire and Swindon have a retention period which lasted to a period of 75 years from date of birth. Somerset has a retention period of 6 years from 18th birthday (a policy similar to that advocated by the Claimant and Information Commissioner); and Liverpool, the London Borough of Richmond, Sunderland and Thurrock have a period of retention of 35 years from case closure (the same, or similar, to the Defendant's Retention Policy).

The statutory regime for child protection

11. This too was largely common ground.
12. Section 47 of the Children Act 1989 requires and authorises the gathering of child protection information. Thus:

(1) Where a local authority …

…

(b) have reasonable cause to suspect that a child who lives, or is found, in their area is suffering, or is likely to suffer, significant harm …

the authority shall make, or cause to be made, such enquiries as they consider necessary to enable them to decide whether they should take any action to safeguard or promote the child's welfare.

…

(7) If, on the conclusion of any enquiries or review made under this section, the authority decide not to apply for an emergency protection order, a child assessment order, a care order or a supervision order they shall -

(a) consider whether it would be appropriate to review the case at a later date; and

(b) if they decide that it would be, determine the date on which that review is to begin.

(8) Where, as a result of complying with this section, a local authority conclude that they should take action to safeguard or promote the child's welfare they shall take that action (so far as it is both within their power and reasonably practicable for them to do so).

13. Although section 47 authorises the making of enquiries and (at least implicitly) the retention of material collected in the course of those enquiries there is no specific statutory provision authorising a local authority to retain information.

The arguments of the Parties

The Claimant

14. Mr Greatorex submitted that the protection of human autonomy and dignity was an important aspect of protection of rights under Article 8(1); and that justifications by the state for the retention of information must be scrutinized with care; see the decision of the ECtHR in S and Marper v UK (2009) EHRR 50 at [102]-[104] and of the CJEU in Digital Rights Ireland Ltd v. Communications Minister [2015] QB 127 at [48]. The scrutiny should be particularly careful where it involves the whole-scale retention of highly sensitive information for a long time and where it concerns child protection data relating to physical or mental health or condition, see s.2(e) of the DPA and S and Marper (above) at [124].
15. He criticized the reasoning in the Policy Report in a number of respects: for example, the failure to set out how it was going to filter information passed on to other organizations, as envisaged in §4.16. However, the claim for which permission to bring Judicial Review proceedings was given was not directed to the detail of the Policy Report but to the Retention Policy of 35 years. In any event, it seems to me that Ms Steyn QC is correct in her submission that a close textual analysis of the Policy Report, as if it had the status of a statute, was unlikely to yield significant insights.
16. Nevertheless, it is clear from §§4.18-4.20 of the Policy Report that Mr Corlett considered the option of setting separate retention dates for each case or reviewing each file at regular intervals to decide whether there was any continuing need for retention, and rejected those options: on the basis of the administrative burden that this would create, and due to the difficulty in forming reliable predictive judgments about the future.

The Interested Party

17. The Information Commissioner was given leave to intervene on the basis that he was likely (through his expertise) to be able to assist the Court in understanding the legal issues and the consequences which may flow from the Court's decision, see for example, De Smith, Judicial Review, 7th Ed. §2-068. He has, of course, particular expertise in, and understanding of, the operation of the DPA.
18. The Information Commissioner's view about the Retention Policy has changed. In a letter dated 2 July 2012, he wrote to the Claimant expressing the view that it was likely that the Defendant's 35 years policy was compliant with the DPA and was lawful. In his submissions to the Court (dated 5 January 2015) he indicated a revised position, which is that the 35 year Retention Policy was unlawful and that it would only be lawful under the DPA for the information to be retained by the Defendant's legal department, and then for only 6 years after the child in question turned 18 and for the purposes of defending itself against litigation, unless on the facts of any particular case, a shorter or longer period should apply.
19. Although I accept that it is open to anyone to change their mind, or even change sides, and that sometimes the change may show that a party has thought with particular care about the issue and has concluded that it was wrong in its initial view, the change in position is striking.
20. Mr Hopkins submitted that the Defendant had failed adequately to justify why it was reasonably necessary for it routinely to retain those records for 35 years after the closure of a case by reference to the purposes for which the child protection records were processed. He argued that the primary purpose for which the personal data in the records was processed was the protection of the children to whom the records relate. In order to achieve that purpose, it was not reasonably necessary routinely to retain those records for 35 after years after case closure. He submitted that his client's proposed retention period was appropriate to deal with a secondary purpose: namely, to permit the Defendant to defend claims which might be brought against it. This proposal was a retention period for the period for the bringing of a child protection-related claim in tort, i.e. 6 years after the individual turns 18. On this basis, it would be sufficient for such purpose for the Defendant's legal department routinely to retain the records for such a purpose up to the expiry of this period, at which point it should assess the records on a case-by-case basis in order to determine whether it is reasonably necessary to retain them. If it were not, they should be destroyed.
21. The 6-year period was not proposed as an inflexible rule, since inflexible blanket policies on retention periods are difficult to justify under the DPA. Some records might warrant earlier destruction, for example, if a child protection case was closed following a categorical 'no case to answer' determination; while other records might warrant longer retention periods, in light of the particular features of the case. Such decisions would be based on the need for objective, case-specific justifications.

The Defendant

22. Ms Steyn QC submitted that the safeguarding of children from abuse or neglect was one of the most important functions of a local authority. The information which is collected and maintained is vital for the purposes of child protection, both in relation to individual cases and so as to enable it to form a broader picture. The records are also part of the individual's personal life, to which they may wish to have access as adults.
23. Permission to bring Judicial Review proceedings was granted before the Defendant's Retention Policy was put in place, and before the Defendant had served its evidence: in particular the witness statements of Stephen Corlett, Fiona Brown and Stephen Day. In the light of that evidence, Ms Steyn argued that the claim was without merit. The policy was designed to address its purposes and is flexible enough to allow information to be deleted sooner, where the particular circumstances indicate that this is appropriate. As such, it strikes the right balance between the broad purposes for retention and the rights of the individual.

The Law

24. The most recent case on the retention of personal information by public bodies is the decision of the Supreme Court in R (Catt) v. Association of Chief Police Officers and another [2015] 2 WLR 664. The case concerned the retention by the police of electronic data about individuals. The Court held, unexceptionally, that the storing of personal information about individuals relating to their private life engaged Article 8 of the ECHR and required to be justified. At [14] Lord Sumption JSC, with whom the other members of the Court agreed, made clear that the requirement for the interference with Article 8 rights to be in accordance with the law had to be viewed realistically.

The infinite variety of situations in which issues of compliance may arise and the inevitable element of judgment involved in assessing them make complete codification impossible.

25. At [31] he added some observations, relevant to the present case, about the processing of information whose significance may only be realized in hindsight.

Most intelligence is necessarily acquired in the first instance indiscriminately. Its value can only be judged in hindsight, as subsequent analysis for particular purposes discloses a relevant pattern. The picture which is thus formed is in the nature of things a developing one, and there is not always a particular point of time at which one can say that any one piece of the jigsaw is irrelevant.

26. This was also a point which was addressed by Baroness Hale DPJSC at [54], where she noted that it may be relevant to a police enquiry into a particular complaint (in her example, domestic violence) for the police to know whether there have been similar incidents in the past.
27. In JR60's Application [2013] NIQB 93, Horner J was concerned with the retention of similar material to that which is in issue in the present case, although the applicant in that case had been a looked after child, a category in respect of which the state has particular responsibilities. Having reviewed decisions of the ECtHR, including S and Marper, the Judge concluded on the evidence before him that a retention period of 75 years was compliant with the law, and was proportionate and necessary. The case is therefore authority for the proposition that, at least in the case of a looked-after child, a retention policy of 75 years was appropriate. However the policy itself was not limited to looked-after children. The court was satisfied in that case that the applicant would be given notice of any request to see her records and would have an opportunity to challenge and considered that was when the issues involved should be resolved: see [24].
28. I am not sure that most of the other cases about retention of criminal records, referred to by the parties, added significantly to the argument.

Discussion

29. Section 4(4) of the DPA requires a data controller to comply with the five 'data protection principles' set out in Schedule 1.
30. The first question to be considered is the question which is implicit from each of the five data protection principles: what is the purpose or purposes of the personal data which has been processed and retained?
31. On this point there was a divergence of view. In the Claimant's skeleton argument Mr Greatorex submitted that the purpose of retention was to enable the Defendant to defend claims brought against it: hence the 6 year period. It was only in the course of questioning by the Court in the course of his argument that he accepted that the purpose of retention was wider in its ambit. The Information Commissioner regarded the defence against claims as a secondary purpose. In any event, although they differed slightly on how a 6 year period should be calculated, in each case the argument was based on the 6-year period for bringing tort claims against the Defendant under s.2 of the Limitation Act 1990.
32. If that were the purpose then their suggested policies would not allow for the possibility that the period for bringing a claim may be extended under s.14(1)(c), s.28 and s.33 of the 1990 Act.^[2]
33. However, in my judgment the purpose or purposes for processing and retaining the personal data extends beyond the ability of the Defendant to defend claims brought against it, for which (in any event) it has a separate retention policy. It is clear that the Defendant's Retention Policy is designed for the purpose of (a) protecting other children, (b) allowing those whose data is retained access to the information in later life, and (c) making it available (if required) in the case of later investigation, enquiries or litigation.

Protecting other children

34. When it is intervening in the best interests of a child, or deciding whether to do so, it is imperative that Local Authorities have as much relevant information as is possible. The assessment of a child's needs requires close consideration of the parents' or carers' capacity to respond to those needs. That assessment will need to be wide ranging and will aim to use all available information, see for example, HM Government's Publication: 'Working together to safeguard children (a guide to inter-agency working to safeguard and promote the welfare of children)', March 2015 at §§29, 35-36, 43-46.
35. As the Court of Appeal noted in the case of In re U (A Child) [2005] Fam 134 at [26], in the context of a judicial determination of whether intervention is necessary:

… the judge invariably surveys a wide canvas, including a detailed history of the parents' lives, their relationship and their interaction with professionals. There will be many contributions to this context, family members, neighbours, health records, as well as the observation of professionals such as social workers, health visitors and children's guardian.

36. The evidence of Stephen Day makes clear that information relating to a particular child can be relevant to future care proceedings for other children many years later, and thus form part of the broader picture to which the Defendant must have regard. If authority were necessary for such a proposition, it is available on the facts outlined by Lord Wilson JSC in In re B (A child) (Care Proceedings: Threshold Criteria [2015] 1 WLR 1911 at [7]-[13].
37. The Defendant's evidence describes the inherent dangers of incidents being viewed in isolation, rather than in their historical context, with the possible consequence that escalating or recurring patterns go unnoticed, to the detriment of children who could otherwise have been protected had the full facts been known. An incomplete picture can result in 'disguised compliance', in which repeatedly dishonest or unreliable accounts are relied on because they are taken in isolation. This is a matter which was highlighted in a publication of The National Society for the Prevention of Cruelty to Children, 'Disguised compliance: learning from case reviews' (March 2014).
38. Fiona Brown's evidence describes several categories of historic data, relating to children whose interaction with the Defendant (or other local authorities), is of benefit in making assessments, some of which extended back for 35 years.
39. The first category of such data is of parents of children who are being assessed. The abuse or neglect in childhood of the parents can have serious, long-term repercussions. This view is supported by a letter of 10 April 2015 from the Association of Directors of Children's Services Ltd:

… there is strong evidence that maltreatment in childhood can have a profound effect on an adult's capacity to parent, not least in the creation of inappropriate sexual behaviours and the perpetuation of harsh or physically abusive parenting styles.

40. The second category is the data of siblings of children who are being assessed. Parents who present risks may have children many years apart. In such circumstances, the records of a parent's earlier behaviour may provide crucial information and context. This was a matter considered at §§4.7 and 4.10 of the Policy Report.
41. The third category is children being cared for by older relatives, where background information may be important.
42. Ms Brown's witness statement gives a number of specific examples, from her recollection, of cases where these types of record have proved relevant long after the subject has reached the age of 18. Her evidence was that in many of these types of case the records would have been destroyed before access to them was sought for the purpose of protecting children if the Claimant's or the Commissioner's proposed retention period had applied. Mr Greatorex was critical of reliance on these examples and suggested that the 6 year period for which he advocated might have afforded sufficient protection; but that is to overlook the fact that Ms Brown's evidence was directed to the purpose of retention. She described specific examples of the types of case where it was necessary to review records long after the records had ceased to become directly relevant as a result of the subject reaching the age of 18 or where a longer term view of the matter was necessary. It is likely that Mr Greatorex would have been very much more critical if Ms Brown's arguments had not given concrete examples.

Access by the data subjects themselves

43. Those whose data has been retained may themselves wish to access the information in later life. Article 8 includes the right to access records relating to one's childhood held by the state, as was recognized in the short and dispositive judgment of the ECtHR in MG v United Kingdom (Access to Social Service Records) (2003) 36 EHRR 3; and s.7 of the DPA creates a specific right of access to personal data.
44. The Defendant's Retention Policy recognises these rights (see §4.13 of the Retention Policy), and gives effect to them. In most cases where adults take advantage of this right, they are seeking information dating back approximately 25 years. Since November 2013, the Council has received 60 subject requests. In several cases, the records would not have been available had they been destroyed according to the Claimant's or the Information Commissioner's preferred retention policy.
45. The records may relate to complex and traumatic childhoods, which the subjects may have failed to understand or remember, or will have suppressed. Subjects may not become aware of their background, or may not be motivated to investigate, until well into adulthood. The requests may be of crucial importance to a sense of identity and emotional or mental well-being. It may also have practical implications for finding relatives.
46. In the JR60 case (see above) at [19]-[20] Horner J recognised that requests for access may occur, particularly at the time of marriage or the birth of a child, or at times of crisis, or after the death of a foster or adoptive carers. The right of an individual to be able to establish details of his or her identity was found to be relevant, even though the applicant (who by then had a successful career and son attending university) did not herself want to know, both because it was possible that she might change her mind and because the right belonged to others, too, such as the applicant's own offspring.
47. The arguments of the Claimant and the Information Commissioner have nothing to say about those who might wish to have access to their own data after they have reached the age of 24.

Investigations, inquiries and litigation

48. The third purpose of maintaining records relates to matters which may become the subject of investigations or inquiries in which retained information may become important. This was a matter addressed in the Policy Report at §§4.3 and 4.11. There are several types of proceedings which might call for this type of evidence.
49. First, there are public inquiries. The most important is the current Independent Inquiry into Child Sexual Abuse, headed by Justice Lowell Goddard ('the Goddard Inquiry'), whose wide-ranging task is to investigate the failures of various institutions to protect children from abuse over the past several decades.
50. The Goddard Inquiry follows several other high-profile inquiries in recent years into the abuse or exploitation of children. These include one in Rotherham, whose remit was for the years 1997-2013, and Oxford, which investigated abuse as far back as 1999. A review by Peter Wanless and Richard Wittham QC into the extent of the Home Office's knowledge of organised child abuse covered the years 1979-1999. One of the issues which has given rise to particular public concern is the existence and sufficiency of records; and, where records are no longer available, why this may be so.
51. The Defendant argued that these inquiries are of fundamental public importance, and would be severely hampered if records were deleted in accordance with the Claimant's or Information Commissioner's proposed policy. This was a matter taken into account by Horner J in the JR60 case at [20] where, accepting this element of public interest, he referred to an Inquiry into Historical Institutionalised Child Abuse in Northern Ireland for the period from 1922 to the present day.
52. The second type of enquiries are police investigations. Revelations about historic child abuse and exploitation give rise to criminal investigations, most infamously in the case of Operation Yewtree, led by the Metropolitan Police. This operation investigated allegations that well-known media personalities and others had engaged in criminal sexual activity, in some cases with underage and vulnerable people. Some of the crimes took place as long ago as 1967. Records of the neglect and abuse of children may be of significant interest to criminal investigators and prosecutors many years after the events themselves; and it is plainly in the public interest that critical evidence be preserved to enable justice to be done, whether this is by corroborating a victim's allegation or exonerating someone who is wrongly accused.
53. The third area of potential relevance is internal reviews. Local Authorities may have concerns about the provision of care and wish to review its procedures in order to make improvements. This is in keeping with their general duties to safeguard children's welfare, as supplemented by the mandatory recommendations in Chapter 2 of the March 2015 Inter-agency Guide, referred to above. Old records may also be relevant to performance reviews or disciplinary proceedings relating to one or more employees, advancing not only the protection of children, but also public confidence in social services.

Conclusion

54. Mr Greatorex pointed out that the reference in §4.10 of the Retention Policy Report to the 35 year period having 'some basis in national advice' was to the 'Retention Guidelines for Local Authorities' published by the Records Management Society in 2003 which had been withdrawn. However, as Ms Steyn pointed out, they were withdrawn because the Records Management Society ceased to exist. That did not mean that the advice and the reasoning ceased to have effect or that the support for the Defendant's Retention Policy was undermined by reliance on the Guidelines. As noted in the introduction to the Guidelines, they were based on best practice and prepared with the help of experienced practitioners in the profession. At §2.15 of the Policy Report, it was acknowledged that the Guidelines had been withdrawn, but noted that there had been no significant changes in legislation, practice or guidance that would have the effect of making the Guidelines redundant.
55. It is clear that Mr Corlett consulted Mr Day who expressed the clear view that understanding and evaluating the present circumstance of children depended very heavily on a view of 'generational behaviours (sic), attitudes and history'; and, as noted at §4.10 of the Policy Report the period of 35 years was not based on any mathematical calculation, but rather on a reasonable view of the various considerations which had been identified.
56. The Claimant and the Information Commissioner accepted that there might be exceptional cases where the case files would need to be retained; but submitted that this could be done by forming a prospective view of their future utility at the end of the 6 year period which they advocated. In my view this approach would involve a cumbersome and time-consuming predictive exercise, which would necessarily err on the side of retention; and the argument overlooks the importance of one of the purposes for which the information is retained: its later use in order to analyse what may be a pattern or risk which can only be identified with hindsight, see Lord Sumption at [31] and Baroness Hale at [54] in Catt (above). The possibility of considering different retention periods for different files was considered and rejected for good reason in §§4.4 and 4.18 of the Policy Report.
57. One of the unusual aspects of this case is that the Claimant is not advancing a case that his own circumstances highlight a particular deficiency in the Defendant's Retention Policy. The challenge is made at a relatively high level of abstraction when compared, for example, with the challenge in JR60 case (above). Furthermore it relates to the storage of specific personal material in relation to a confined number of people, and for a particular and identified purpose. It is a very different to the Digital Rights Ireland case (above), where the Grand Chamber of the ECJ considered the impact of Council Directive 2006/24/EC which covered the retention of all traffic data relating to all electronic communication. Its conclusion that the Directive was inconsistent with Articles 7, 8 and 52(1) of the European Charter, was hardly surprising in the light of the Court's view at [56] that it entailed 'an interference with the fundamental right view of practically the entire European Population.'
58. The suggestion that the files should be reviewed every 7 years seems to me to involve a disproportionate use of labour and unproductive use of resources which are better devoted to protecting children. As Ms Brown noted^[3], such reviews could not be carried out by anyone other than an experienced social worker, looking at the entire file and having to assess potential risks.
59. As Ms Brown put it in §58 and 59 of her witness statement:

58. [The Defendant] deal[s] with a large number of cases and at any point will have approximately 3,000 children as open cases to Children's Services …

59. The current retention policy provides for retention for a period long enough to ensure that it is improbable that information from a closed record will be of central importance in any future work to protect children. In some cases, it might in principle be possible to dispose of records after a shorter period – particularly in situations where an investigation found where an investigation has found no evidence that abuse or neglect [has] taken place – but it would be necessary to assess the risk in every case.

60. I accept that evidence. As noted above, the Defendant's investigation into the Claimant and his family found no evidence of abuse or neglect, and the records have been dealt with consistently with the Defendant's Retention Policy by deletion.
61. In the light of the above I accept the Defendant's submissions that there is a need to keep the records for a substantial period, and certainly substantially longer than the period argued for by the Claimant and the Information Commissioner. It might be argued that the period of 35 years is not the only possible period of retention, but in my judgment it falls within the bracket of legitimate periods of retention.
62. I have therefore concluded that the Defendant's Retention Policy of 35 years is in accordance with the law, has been carefully considered, adapted to the purposes for which it is required, and is applied proportionately and flexibly.
63. It follows that the claim fails.


 


Annex The Retention Policy Report

Discussion

4.1 Since there is no specific legislative basis for retention periods for child protection records, and no national guidance from the Government or other statutory sources, this is an area in which each local authority has to form a reasonable view after considering the issues involved. Practice in other local authorities, and advice given by local bodies, provide useful context, but cannot determine what decision the Council should take.

4.2 A balance needs to be struck between two opposing considerations:

a) as a general principle, retaining unnecessary information about the private family lives of people who have not consented to this is unacceptable, and could be a breach of their human rights.

b) but the Council is obliged to give the very highest priority to the safety of children. Any child protection investigation, with the possible exception of one which is founded on a simple and clear misunderstanding (such as a confusion of similar names), engages this obligation. Even if the investigation concludes that there is no evidence of neglect or abuse, it is possible that putting the information which it uncovered together with subsequently available information may lead to a revised conclusion-it is accepted good practice that local authorities should monitor patterns of referrals about child protection concerns raised about specific children, even where they do not individually merit investigation, and will respond if a series of referrals about the same child or family suggests that there may be an issue which needs to be further explored.

4.3 Recent national events make particularly clear the potential importance of historical information about child protection investigations. Perpetrators of serious abuse may succeed in deflecting investigations into their behaviour for many years, and may continue to abuse other children for decades; young people who have been abused and disbelieved may lack the confidence to come forward again until it becomes clear to them that they may now receive a different response.

4.4 Clearly there are different considerations in different circumstances. The potential risks from deleting information about possible child sex abuse or exploitation are different from those from deleting information about alleged neglect or inappropriate punishment of a small child. Special issues arise where the allegation of abuse is made on the basis of disputed medical evidence, either about the causes of a child's health problems or the appropriateness of parental decisions about medical treatment. However it would be difficult to design and administer a policy which provided for different retention limits for each kind of child protection case-particularly so given that one of the reasons for retaining information is to make it possible to reconsider previous conclusions if subsequent evidence makes it appear that the situation may have been interpreted wrongly.

4.5 It is also relevant to consider the nature and extent of the harm to the family life and privacy of the children and adults involved in a child protection investigation which would result from any unnecessary retention of information. Social services child protection records are held under conditions of strict confidentiality. All those with access to these records know that they are required to maintain this confidentiality, and to access them only when they need to. Information from these records is shared only when there is clear legal reason to do so-usually because it appears necessary to do so to protect children from significant harm. While it cannot be wholly ruled out that the unnecessary retention of information could have negative consequences for the subjects of that information, these arrangements significantly reduce the likelihood of that.

4.6 The bare minimum policy which the Council should adopt would be to retain the records for six years, in line with the general guidance about records issued by DH in 2000 (which covered records of all social services involvements, not just child protection work or involvements with children). However it would appear perverse for the local authorities which have statutory responsibility for maintaining child protection registers to retain the relevant records for a shorter period than is required for retention of what will often be the same records by NHS national guidance. It would also be hard to reconcile this with the advice in the 2000 guidance that records should be retained for longer than six years where they "relate to individuals and providers of services who have, or whose staff have been judged unsatisfactory". Such a policy would also in many cases mean that if as a young adult the subject of a child protection investigation wished to take legal action against an adult who had abused or neglected them-or indeed against the local authority for failing to protect them-crucial information would no longer be available. Under the Limitation Act 1980, young adults are able to bring legal actions about deliberately caused personal injuries to them during childhood up to six years after their eighteenth birthday.

4.7 The minimum policy that would avoid the last of these consequences would be retention of records until six years after all children involved had reached the age of majority-or, preferably, to align the social services retention period with the period required for NHS records. However this would not address the wider issues about the potential for new information to cast a different light on an earlier investigation years into the future, and the potential for a pattern of abuse or neglect to be repeated over many years. The NHS policy cannot be taken as a direct precedent for a social services policy, since the NHS does not have the lead statutory responsibility for safeguarding children.

4.8 The maximum potentially justifiable retention policy to address these issues would be to retain child protection records for the same length of time as records of looked after children, as a number of other local authorities currently do. Retention for 75 years would almost certainly ensure that no information relevant to the safety of other children or to adults' ability to understand traumatic events from their own past would be destroyed.

4.9 While there is a case for an extension to 75 years, this does not on balance seem proportionate in the absence of national advice that records should be retained for that length of time. It is possible that new national advice will ultimately result from recent investigations into historic sexual abuse. In the meantime, it appears most reasonable to adopt a policy somewhere between 75 year retention and retention based on the NHS retention period.

4.10 The specific retention period of 35 years is not based on any arithmetical calculation, but on a reasonable view of the balance between the considerations described in this report. The primary merits of this specific period are that it does have some basis in national advice, though not from a statutory agency, and is long enough to provide some reassurance that the issues about historic abuse and adults' ability to recover information about their past would not be of great concern. It is also of the right order of magnitude to ensure that, in most cases, records would still be available at any time when children from a family which had been the subject of child protection concerns themselves became parents-which is a key point at which the consequences of past abuse may again become significant. It is also a long enough period to make it unlikely that parents whose first children had been the subject of child protection concerns would still be parenting other children. Finally, while this is not in itself a justification for the policy, it corresponds to fairly widespread practice elsewhere, at a time when there has still been no guidance from the Government.

4.11 There does not appear to be any obvious basis for any other period between the date based on the NHS guidance and the period based on the regulations for looked-after children. It is possible that the national investigations into historical abuse which are currently taking place-which include an independent inquiry whose scope embraces events said to have taken place in the early 1980's, just within the 35 year period from the present-may lead to specific national recommendations-and this would clearly be desirable, for the reasons given in the Soham serious case review.

What should be included in the Child Protection Category

4.12 The current wording in the retention policy defining the scope of the category of child protection records, which was based on Staffordshire's policy, is not wholly clear, and could be interpreted as including some information which does not need to be retained for a long period, such as missing children from other local authority areas. Proposed revised wording is included in the recommendations of this report.

Use of retained information

4.13 In general, retained information will only be actively used if the Council receives further potentially-related referrals under its statutory social services functions about people involved in the case; or if one of the people involved asks to access the information to help them to understand their past or to support legal action; or to support investigations of alleged historical child abuse. In each of these cases, the risk that the subjects of the information will be adversely affected by the use of the information without good cause is limited, since the trigger for its further use will be directly related to further concerns about matters related to children's social care.

4.14 A greater risk of unjustified adverse impacts could arise when retained information is used for other purposes-for instance if it is consulted when approving a person for employment or as a childminder or foster carer. Currently there are a number of circumstances in which this can happen:

a) the police may request information to support an investigation into current or historic allegations of abuse.

b) Ofsted request information as part of the process of making checks on applicants for registration as childminders-and checks for similar reasons are made by the voluntary organisation SSAFA for people applying to work in child care on military bases in the County.

c) University social work departments request confirmation of details which they have been given by students about their personal involvement with social services.

d) Checks are made on people who apply to be local authority foster carers or adopters in Northumberland, or to provide some other kinds of local authority arranged services for children.

4.15 Where information is being passed on to other organisations, there is a need for some care to ensure that old information which is irrelevant to the reason for the enquiry is not passed on to third parties which might make decisions adversely affecting the subject of the information. Case law on disclosures by the police suggests that the courts would expect there to be a filtering process to ensure that information passed on is relevant and accurate (though the most relevant case law concerns information about minor offences which do not involve any suggestion of harm to children).

4.16 It is suggested that the Council's policy on the filtering of information provided in these circumstances should be clarified, using as a model the statutory guidance cited above on disclosing past safeguarding allegations when providing references for foster carers. The policy would thus be that, before passing on information about closed child protection cases to other organisations in response to any of the forms of request listed above we would review the child protection records, and exclude from disclosure any allegations which we were satisfied were unsubstantiated, unfounded or malicious. This would not affect the existing arrangements for disclosures to the police associated with criminal investigations.

4.17 If this policy is approved, we will contact the organisations concerned to let them know of it, and will review the issue further if they raise any concerns.

Varying retention rates

4.18 There is in principle an option of setting separate retention dates for every case, or reviewing all files at regular intervals to decide whether there continues to be reason for retention. However it would be difficult to recommend any options on these lines, for a number of reasons:

a) One of the primary reasons for retention is that information may take on a new significance in the light of later events. It is not clear how practical it would be to review records against criteria based on hypothetical future possibilities.

b) The practical effect of a policy of regular reviews of closed files during the 35 year retention period would be to create a considerable additional burden on social workers. It would not be safe or reasonable for such reviews to be carried out as an administrative task, so experienced child protection social workers would have to carry them out. Doing so conscientiously would require close reading of often extensive documentation-and potentially a need to make further contact with children, families or the professionals originally involved.

c) Setting individual retention dates at the time of case closure would be practically much easier-but would again raise the question about how to assess the possibility that future information might lead to reinterpretation of the conclusions reached at the time.

4.19 The approach which appears most practical would be to continue to operate a standard retention period, but with explicit provision for marking some records as not needing to be retained for 35 years, and reviews of individual cases if any involved parties request them. For the reasons given in this report, both of these provisions would likely to be invoked only in exceptional circumstances. In many cases, a more appropriate response would be to ensure that any errors are prominently corrected, to remove the risk of future misinterpretation, while leaving the original version in place as a historical record of the information which may have been relied on at a particular time. An example of a situation in which it might instead be reasonable to schedule early destruction of a record, or to agree to early destruction following a review at the request of a person involved, would be an investigation, and possibly a child protection plan, which was based wholly on a medical diagnosis which is subsequently conclusively ruled out as a result of further tests.

4.20 Likewise, it would be reasonable to make provision for marking a record as potentially requiring retention for longer than 35 years, in exceptional cases. The effect would be that at the point when destruction would ordinarily take place, the file would be reviewed before being destroyed. It is not likely that this provision would be often used, since in situations where there is an exceptionally high risk that abuse will have long-term consequences, there would usually also be reasons to maintain ongoing involvement with the child(ren) concerned, which would itself delay the start of the 35 year retention period.

4.21 In any case where a record is destroyed before the standard period has elapsed, experience in the case which led to the current judicial review proceedings has made it clear that before destruction takes place it will be important to consider carefully two issues:

a) whether some elements of the factual record of the actions of the Council need to be retained, possibly outside the main electronic and manual file systems, in case of future legal proceedings by anyone involved;

b) how to communicate with other agencies which may still be retaining records of the case, to inform them of the Council's decision and ask them to consider whether to delete or amend their own records.

Note 1   The Common law requires retention of information by the state to be justified: see for example R (TD) v. Commissioner of Police of the Metropolis [2014] EWCA Civ 585 at [17] (Laws LJ). Retention and storing of personal information by the state constitutes a prima facie interference with article 8(1) rights and so requires justification in accordance with article 8(2): see Leander v Sweden (1987) 9 EHRR 433, at [48], S and Marper v. UK (2008) 48 EHRR 1169, [121]; R (GC) v Commissioner of Police for the Metropolis [2011] 1 WLR 1230 and MM v UK (24029/07, 13 November 2012). Section 4(4) of the DPA requires data controllers ‘to comply with the data protection principles in relation to all personal data with respect to which he is the data controller’. The principles are set out in Part 1 of Schedule 1.     [Back]

Note 2   In NXS v LB Camden [2009] EWHC 1786 (QB), although the claim form was issued 11½ years after the limitation period had expired (in that case, three years after the Claimant’s 18th birthday) had expired, Swift J held that it was equitable to allow the action to proceed. At the trial the claimant was 33 and the claim relied heavily on care records as evidence of Camden’s breaches of duty. See also, Foskett J’s remarks in The Children's Rights Alliance for England v Secretary of State for Justice [2012] EWHC 8 (Admin) at [229] in the context of a policy for retaining records so as to be able to defend a claim which may be brought many years later.     [Back]

Note 3   Paragraph 61 of her witness statement    [Back]