I was pleased to give evidence to the Committee on 5 February. At that session I said I would write to you to provide so far as possible additional information in response points raised by you and other members of the Committee.

  As I said to the Committee, the Government attaches great importance to tackling the problem of knife crime and was keen to use the data gathered to assess the progress of the Tackling Knives Action Programme (TKAP) to inform the public about the true prevalence of knife crime as soon as it was reasonable and appropriate to do so. As we discussed when I gave evidence, it would be odd if Members of Parliament and the public could secure the publication of information, however preliminary, using for example the Freedom of Information Act, whereas the Government was unable to provide relevant information pro-actively. The Government believes that the public perception of the prevalence of knife crime is self-reinforcing, especially among young people. Unfortunately, on this occasion a mistake was made in prematurely including NHS data in the fact sheet on the TKAP. We have acknowledged that and are determined to learn the lessons. The Government also attaches great importance to the reforms we introduced to the governance of official statistics, with the firm aim of enhancing public confidence in them. It welcomes the new Code of Practice for Official Statistics published by the UK Statistics Authority in January.

  Sir Gus O'Donnell's letter of 16th January to you sets out what happened in this case.

  In addition, you asked about who authorised the release of statistics in the fact sheet. I can confirm the fact sheet, which was a Home Office publication, was published under HO authority.

  As Sir Gus's letter explains, the decision was taken to include numbers supplied by the NHS Information Centre alongside those supplied by police forces and others. It was the use of NHS data that the National Statistician raised with the Permanent Secretary in No10 on 12 December last.

  While I appreciate the Committee's reasons for probing, it would not be right to identify the individuals involved for reasons set out in the section of the Cabinet Office's guidance on Departmental Evidence and Response to Select Committees (usually known as the "Osmotherley rules") dealing with the conduct of individual officials.

  At the hearing I undertook to let you know about numbers of special advisers. The Government publishes names and numbers of special advisers to Parliament on an annual basis. This was last published on 22 July 2008 when there were 73 special advisers across government of whom 24 worked in No 10. Information for the current year will be published when it is ready at the end of the financial year. The Committee will know that, under the Code of Conduct for Special Advisors, special advisers cannot give instructions to permanent civil servants. You asked about the role of special advisers in the release of the fact sheet. I can confirm that special advisers in DH, and No10 were among those involved in the discussions about whether to release the fact sheet but that the Home Office, and not special advisers, took the final decision to do so. The involvement of the special adviser in No 10 and the circumstances leading up to the decision to publish are set out in Sir Gus O'Donnell's letter.

  You asked about the role of Ministers. As stated in Sir Gus O'Donnell's letter, Ministers in the Home Office (the lead department for the TKAP) were keen to use the data gathered to assess the progress of the initiative to inform the public about the prevalence of knife crime as soon as it was reasonable and appropriate to do so. Home Office Ministers were not aware that there were outstanding concerns around the use of the NHS data until after the fact sheet had been published. As you know, the Home Secretary has apologised to the House for what happened.

  You asked about targeted action, disciplinary or educational, being taken in respect of any of the individuals involved. For the reasons I set out above, I do not think it appropriate to discuss individuals, although I have no doubt the individuals involved, and others not directly involved, very much regret the damage that has been done and are now much more fully aware of the need for care in publishing data that relates to official statistics.

  As I advised the Committee, a one-page summary of key do's and don'ts has been drafted. Sir Gus O'Donnell has sent the summary to permanent secretaries to circulate to private offices, special advisers, press offices, policy officials and others in the department likely to be involved in considering the publication of data. It draws their attention to the new Code of Practice for Official Statistics and to the key things they should consider as users of data. I am pleased to attach a copy of this one-page guidance. Sir Michael Scholar and Karen Dunnell spoke to a meeting of permanent secretaries on Wednesday 11 February where it was agreed that the way civil servants treated statistics was a key part of civil service integrity and that Sir Gus O'Donnell's letter set out the main lessons which departments would follow.

  You asked whether statisticians in the Home Office were involved in the publication of the fact sheet. As Sir Gus O'Donnell's letter says the crime figures used for monitoring the TKAP were not under the control of Home Office statisticians. I understand they were collected quite separately from the Police Reported Crime Statistics and only from the police forces taking part in the programme in order to monitor its effects. The Home Office statisticians did not regard those figures as official statistics or falling within the remit of the Code of Practice for Official Statistics, and as I said in evidence, were therefore content for them to be published without the need to review those figures before publication.

  You asked what reason HO had to believe the concerns of NHS statisticians had been addressed. As Sir Gus O'Donnell set out in his letter this situation arose due to a misunderstanding about the status of the NHS data compared to the Home Office data and the rules that applied.

  You asked to be copied into the response to Mark Easton's questions of 9 February. As this letter addresses the points he raises, the Cabinet Office press office will respond by referring Mr Easton to it. I have also placed a copy of this letter in the House of Commons Library.

  I am pleased to offer this further clarification of what happened, as reported to the Cabinet Office. The NHS data being used within Government for monitoring the TKAP was published inappropriately. This should not have happened.

BACKGROUND

  The Statistics and Registration Service Act 2007 established the UK Statistics Authority and required it to draw up a Code of Practice for official statistics.

  The Code of Practice was published in January 2009 and provides a statement of good practice to be applied to official statistics in the United Kingdom.

  The key points to note are:

— When preparing any publication containing statistics, including those drawn from administrative or management information, you must involve statistical professionals at the earliest opportunity.

— You must not use unpublished statistics without the advice of a statistical professional.

— You must not selectively quote favourable data from any unpublished dataset.

— Decisions taken by statistical professionals are final.

— Any publication containing official statistics must provide information relating to their quality, reliability and usability.

— Pre-release access to official statistics is a privilege. You must not disclose any information, nor seek to alter it in any way.

— Ignoring any of the above may constitute a breach of the Code and result in an investigation by the UK Statistics Authority and a published report to Parliament.

FURTHER HELP

  If you are unclear about what course of action to take, or with any of the advice provided above:

— Contact your departmental Head of Profession for Statistics who is [Add Details for your department].

— If you cannot reach them or a deputy you can contact the National Statistician through the Helpline on 020 7014 2350.

— The Code of Practice for Official Statistics is available at http://www.statisticsauthority.gov.uk/assessment/code-of-practice/index.html