Tag Archives: ICO

Virgin Media, and a stray email

Anyone who’s worked for a large organisation is likely to be familiar with the situation when someone mistakenly sends an email to everyone who works there. Replies – to all – start straight away: “Hi, I don’t know what this means?” “Hi, nor me” “Hi, I don’t think you meant to send this to me” “Nor me” “Hi everyone, please don’t ‘reply to all'” “Hi, you just did the same thing!!!” “Stop replying to all!” “You too!!!” “AAAAGGGHHHH!!!” etc etc, until eventually it settles down.

And then two weeks later someone comes back from leave and replies to all “Hi, I don’t know what this means”…

I imagine the frustration felt by fellow employees in those circumstances doesn’t begin to equate to that felt by some Virgin Media customers, if stories about an incident yesterday are correct. As The Register reports

The broadband biz emailed Brits using its virgin.net email service, which is provided by Google, to warn them of some forthcoming changes…But any email replies to that message were sent to everyone on the mailing list: the email address the update was sent from acted as a conduit to the full list of virgin.net customers. This not only spewed hundreds of extra missives into inboxes, it also shared the senders’ email addresses with everyone on the list

And the BBC says

Some people reported receiving hundreds of emails, including spam messages and light-hearted exchanges between other customers.

I’ve added the emphasis there, to highlight how excruciatingly annoying it must have been to be on the receiving end of hundreds of light-hearted messages like “I don’t know why you’re emailing me” “Stop replying to all!!!” “You’re doing it too LOL!!” ad nauseum.

Virgin Media have apologised, and tell customers that the issue is now resolved

A small proportion of our customers have received an email from one of our suppliers which, if they reply-all, it is sent to a wider group…We are confident that this issue has now been resolved, the problem stopped and further messages prevented.

I’ve just got a couple of observations to make. One is that “a small proportion of our customers” does not necessarily mean a small number, and while this is not quite a simple “reply to all” issue (it seems that the mailing list was wrongly configured) it clearly caused considerable disruption for those affected. And if Wikipedia is correct Virgin Media has several million customers – a “small proportion” of those could well number the 130,000-odd that some news outlets are claiming were affected. And the other observation is that as far as I can see Virgin Media don’t say whether they have informed the Information Commissioner, who will, no doubt, be wanting to ask some questions to establish whether this incident was as a result of a serious contravention of the data controller’s obligations under the Data Protection Act 1998. After all it only takes one careless individual to send a wrongly-addressed email, but it might point to information security failings if a mailing list is wrongly configured.

 

1 Comment

Filed under Breach Notification, Data Protection, Information Commissioner

Opting patients out of care.data – in breach of data protection law?

The ICO appear to think that GPs who opt patients out of care.data without informing them would be breaching the Data Protection Act.  They say it would be unfair processing

In February of this year GP Dr Gordon Gancz was threatened with termination of his contract, because he had indicated he would not allow his patients’ records to be uploaded to the national health database which as planned to be created under the care.data initiative. He was informed that if he didn’t remove information on his website, and if he went on to add “opt-out codes” to patients’ electronic records, he would be in breach of the NHS (GMS contract) Regulations 2004. Although this threatened action was later withdrawn, and care.data put on hold for six months, Dr Gancz might have been further concerned to hear that in the opinion of the Information Commissioner’s Office (ICO) he would also have been in breach of the Data Protection Act 1998 (DPA).

A few weeks ago fellow information rights blogger Tim Turner (who has given me permission to use the material) asked NHS England about the basis for Health Services Minister Dan Poulter’s statement in Parliament that

NHS England and the Health and Social Care Information Centre will work with the British Medical Association, the Royal College of General Practitioners, the Information Commissioner’s Office and with the Care Quality Commission to review and work with GP practices that have a high proportion of objections [to care.data] on a case-by-case basis

Tim wanted to know what role the ICO would play. NHS England replied saying, effectively, that they didn’t know, but they did disclose some minutes of a meeting held with the ICO in December 2013. Those minutes indicate that

The ICO had received a number of enquiries regarding bulk objections from practices. Their view was that adding objection codes would constitute processing of data in terms of the Data Protection Act.  If objection codes had been added without writing to inform their patients then the ICO’s view was that this would be unfair processing and technically a breach of the Act so action could be taken by the ICO

One must stress that this is not necessarily a complete or accurate respresentation of the ICO’s views. However, what appears to be being said here is that, if GPs took the decision to “opt out” their patients from care.data, without writing to inform them, this would be an act of “processing” according to the definition at section 1(1) of the DPA, and would not be compliant with the GPs’ obligations under the first DPA principle to process personal data fairly.

On a very strict reading of the DPA this may be technically correct – for processing of personal data to be fair data subjects must be informed of the purposes for which the data are being processed, and, strictly, adding a code which would prevent an upload (which would otherwise happen automatically) would be processing of personal data. And, of course, the “fairness” requirement is absent from the proposed care.data upload, because Parliament, in its wisdom, decided to give the NHS the legal power to override it. But “fairness” requires a broad brush, and the ICO’s interpretation here would have the distinctly odd effect of rendering unlawful a decision to maintain the status quo whereby patients’ GP data does not leave the confidential confines of their surgery. It also would have the effect of supporting NHS England’s apparent view that GPs who took such action would be liable to sanctions.

In fairness (geddit???!!) to the ICO, if a patient was opted out who wanted to be included in the care.data upload, then I agree that this would be in breach of the first principle, but it would be very easily rectified, because, as we know, it will be simple to opt-in to care.data from a previous position of “opt-out”, but the converse doesn’t apply – once your data is uploaded it is uploaded in perpetuity (see my last bullet point here).

A number of GPs (and of course, others) have expressed great concern at what care.data means for the confidential relationship between doctor and patient, which is fundamental for the delivery of health care. In light of those concerns, and in the absence of clarity about the secondary uses of patient data under care.data, would it really be “unfair” to patients if GPs didn’t allow the data to be collected? Is that (outwith DPA) fair to GPs?

Leave a comment

Filed under care.data, Confidentiality, Data Protection, data sharing, Information Commissioner, NHS

Kent Police get £100,000 penalty for poor data security

I blogged last week about “data breaches”, and the need to define and sometimes to differentiate between a breach of the Data Protection Act 1998 (DPA) and a general data security breach. Well, I’m (not at all) pleased to say that today’s news of the latest monetary penalty notice (MPN) served by the Information Commissioner’s Office (ICO) on Kent Police doesn’t need any such nuanced analysis. Here was a data security breach which was also a manifest breach of the DPA.

A police officer, by chance, discovered in some premises video tapes clearly marked as police material. He subsequently ascertained that the owner had found them, and much more besides, in the basement of a former police station which he had purchased. It is difficut to think of more sensitive information than the kind which was involved here. In part it consisted of

documents and video/audio tapes containing confidential and highly sensitive personal data about a significant number of individuals. These included files relating to threats to kill, rape, grievous bodily harm and child abuse cases; interviews with victims, witnesses/informants and suspects

Although the force had initially

taken some steps to safeguard the information by carrying out inspections of the former police station which identified that items were still in situ

the failure to have any policies in place, or to assign responsibility to anyone, meant that this was a clear and serious contravention of the seventh data protection principle (relating to data security measures) of a kind likely to cause, at least, substantial distress. I would add, although the ICO does not, that it might well have been also a serious contravention of the fifth principle (“Personal data processed for any purpose or purposes shall not be kept for longer than is necessary for that purpose or those purposes”). Given this, it is somewhat surprising that this case falls (admittedly at the top end) into the lowest category of cases qualifying for an MPN (the ICO’s internal guidance says that these cases will attract an amount of £40,000 to £100,000). Bearing in mind that Brighton and Sussex University Hospitals NHS Foundation Trust got an MPN of £325,000 for failing to dispose of computer hard drives properly, this current MPN seems low.

It also, once again, draws attention to the importance of good records management within police forces. I wrote only recently, in the context of the Ellison Review of policing relating to the Stephen Lawrence inquiry, about how records management is essential for the operation of the rule of law and the current case just gives even greater strength to this.

1 Comment

Filed under Data Protection, enforcement, Information Commissioner, monetary penalty notice, police, records management

Sale of patient data – time for an independent review?

The Sunday Times reports that a billion patient records have been sold to a marketing consultancy. Is it time for an independent review of these highly questionable data sharing practices?

In 2012, at the behest of the then Secretary of State for Health, Andrew Lansley (driver of the Health and Social Care Act 2012), Dame Fiona Caldicott chaired a review of information governance in the NHS. Her report, which focused on the issue of sharing of information, was published in April 2013. At the time a statement in it, referring to the Information Commissioner’s Office (ICO) stood out to me, and it stands out even more now, but for different reasons. It says

The ICO told the Review Panel that no civil monetary penalties have been served for a breach of the Data Protection Act due to formal data sharing between data controllers in any organisation for any purpose

At the time, I thought “Well duh” – of course the ICO is not going to take enforcement action where there has been a formal data sharing agreement, because, clearly, the parties entering into such an agreement are going to make sure they do so lawfully, and with regard to the ICO guidance on data sharing – lawful and proportionate data sharing is, er, lawful, so the ICO wouldn’t be able to take action.

But now, with the frequent and worrying stories emerging of apparent data sharing arrangements between the NHS Information Centre (NHSIC), and its successor, the Health and Social Care Information Centre (HSCIC), I start to think the ICO’s comments are remarkable for what they might reveal about them looking in the wrong direction, when they should have been paying more attention to the lawfulness of huge scale data sharing arrangements between the NHS and private bodies. And now, The Sunday Times reports that

A BILLION NHS records containing details of patients’ hospital admissions and operations have been sold to a marketing consultancy working for some of the world’s biggest drug companies

I think it is time for a wholesale review, properly funded, by the ICO as independent regulator, of these “formal data sharing” arrangements. They appear to have a questionable legal basis, based to a large extent on questionable assumptions and assurances that pseudonymisation equates to anonymisation (which anyone who looks into will realise is nonsense).

And I think the review should also consider how and why these arrangements appear to have deliberately been taking place behind the backs of the patients whose data has been “shared”.

Leave a comment

Filed under care.data, Data Protection, data sharing, Information Commissioner, monetary penalty notice, NHS, Privacy

Analysis prompted by Morrisons “data breach”

Yesterday’s data breach involving Morrisons supermarket and its staff payroll illustrates how difficult it is properly to handle such incidents, and perhaps provides some learning points for the future. But also raises issues about what is a “data breach

What do we mean by “data breach”, “personal data breach”, “data security breach” etc?

The draft European General Data Protection Regulation (GDPR), which continues to slouch its way towards implementation, says in its current form that

In the case of a personal data breach, the controller shall without undue delay notify the personal data breach to the supervisory authority [and]

When the personal data breach is likely to adversely affect the protection of the personal data, the privacy, the rights or the legitimate interests of the data subject, the controller shall, after the notification referred to in Article 31, communicate the personal data breach to the data subject without undue delay

“without undue delay” is, by virtue of (current) recital 67, said to be “not later than 72 hours” (in the original draft it was “where feasible, within 24 hours”). However “personal data breach” is not defined – it is suggested rather that the proposed European Data Protection Board will set guidelines etc for determining what a “breach” is.What is not clear to me is whether a “breach” is to be construed as “a breach of the data controller’s legal obligations under this Regulation”, or, more generally, “a breach of data security”. Certainly under the current domestic scheme there is, I would argue, confusion about this. A “breach of data security” is not necessarily equivalent to a breach of the Data Protection Act 1998 (DPA). To give a ludicrous example: if a gunman holds a person hostage, and demands that they unencrypt swathes of personal data from a computer system and give it to them, then it is hard to see that the data controller has breached the DPA, which requires only that “appropriate technical and organisational measures shall be taken against unauthorised or unlawful processing of personal data and against accidental loss or destruction of, or damage to, personal data” (which clearly cannot be construed as an unlimited obligation) but there has most certainly been a breach of data security.

It is unclear whether Morrisons chose to inform the Information Commissioner (ICO) about their incident, but the wording they’ve used to describe it suggests they are seeing this not as a breach of their obligations under the DPA, but as a potentially criminal act of which they were the victim: on their Facebook page they describe it as an “illegal theft of data” and that they are liaising with “the police and highest level of cyber crime authorities” (a doughnut to anyone who can explain to me what the latter is, by the way). If an offence has been committed under section 55 of the DPA (or possibly under the Computer Misuse Act 1990) there is a possible argument that the data controller is not at fault (although sometimes the two can go together – as I discuss in a recent post). Morrisons make no mention of the ICO, although I have no doubt that they (ICO) will now be aware and making enquiries. And, if Morrisons’ initial assessment was that they hadn’t breached the DPA (i.e. that they had taken the appropriate technical and organisational measures to mean they were not in breach of the seventh DPA principle), they might quite understandably argue that there was no need to inform the ICO, who, after all, regulates only compliance with the DPA and not broader issues around security breaches. There was certainly no legal obligation under current law for Morrisons to self-notify. Plenty of data controllers do, often ones in the public sector (the NHS Information Governance toolkit even automatically delivers a message to the ICO if an NHS data controller records a qualifying incident) but even the ICO’s guidance is unclear as to the circumstances which would trigger the need to self-notify. Their guidance is called “Notification of data security breaches to the ICO” but in the overview at the very start of that guidance it says

Report serious breaches of the seventh principle
Ultimately I see it boiling down to two interpretations: report a data security breach so that the ICO can assess whether it is a serious breach of the seventh principle, or, assess the data security breach yourself, and if you assess it as a serious breach of the seventh principle, report that to the ICO. This is not obligatory under the current domestic data protection law, so to an extent it is an arid discussion, but if the obligation to notify does become obligatory under the GDPR it will become much more important.
There is one domestic law under which it is obligatory to report a “personal data breach”. The Privacy and Electronic Communications (EC Directive) Regulations 2003 amended by 2011 Regulations, require a provider of a public electronic communications service to notify the ICO of
a breach of security leading to the accidental or unlawful destruction, loss, alteration, unauthorised disclosure of, or access to, personal data transmitted, stored or otherwise processed in connection with the provision of a public electronic communications service
This notably does not specify that the breach has to constitute a breach of the service provider’s DPA obligations, and one wonders if this is the sort of thing that will be specified as a breach once the GDPR is implemented.
Morrisons’ notification to data subjects

The people whose data was apparently compromised in the Morrisons “breach” were its staff – it was payroll information which was allegedly stolen and misused. It appears that Morrisons emailed those staff with internal email addresses (how many checkout staff and shelf-stackers have one of those?) and then, as any modern, forward-thinking organisation might, it posted a message on its Facebook page.However, I really wonder about that as a strategy. The comments on that Facebook page seem to be threatening to turn the incident into a personnel, and public communications disaster, with many people saying they had heard nothing until they read the message. Moreover, one wonders to what extent some staff might have been misled, or have misled themselves, into assuming that the comments they were posting were on some closed forum or network. As was suggested to me on twitter yesterday, some of the comments look to be career-limiting ones, but by engaging on its social media platform, might Morrisons be seen to have encouraged that sort of robust response from employees?

Much of this still has to play out – notably whether there was any contravention of the DPA by Morrisons – but, in a week when their financial performance came under close scrutiny, their PR handling of this “data breach” will also be looked at very closely by other data controllers for lessons in case they are ever faced with a similar situation.

4 Comments

Filed under Breach Notification, Data Protection, employment, Information Commissioner, PECR, social media

We thought you cared(ata)

David Evans is Senior Policy Officer at the Information Commissioner’s Office (ICO). In an interview with “The Information Daily.com” uploaded on 12 March, he spoke about data sharing in general, and specifically about care.data (elsewhere on this blog passim). There’s a video of his interview, which has a backdrop with adverts for “Boilerhouse Health” and “HCI Daily“, both of which appear to be communications companies offering services to the health sector. David says

care.data…the overall project is very good because it’s all about making better use of information in the health service…what care.data appear to have done is failed to get that message across

Oddly, this view, that if only the people behind care.data had communicated its benefits better it would have sailed through, is very similar to that expressed by Tim Kelsey, NHS National Director for Patients and Information and head cheerleader for care.data. Tim said, for instance, after the announcement of a (further) six-month delay in implementation

We have been told very clearly that patients need more time to learn about the benefits of sharing information and their right to object to their information being shared

Both David and Tim are right that there has been a failure of communication, but I think it is completely wrong to see it merely as a failure to communicate the benefits. Any project involving the wholesale upload of confidential medical records, to be processed and disclosed, at various levels of deidentification, to third parties, is going to involve risk, and will necessitate explanation of and mitigation of that risk. What the public have so far had communicated to them is plenty about the benefits, but very little about the risks, and the organisational and technical measures being taken by the various bodies involved to mitigate or contain that risk. Tim Gough has argued eloquently for a comprehensive and independent Privacy Impact Assessment to be undertaken (while criticising the one that was published in January

To be fair, NHS England did publish a PIA in January 2014, which does appear a little late in the day for a project of this kind.  It also glosses over information which is extremely important to address in full detail. Leaving it out makes it look like something is being hidden

As far as I am aware there has been no official response to this (other than a tweet from Geraint Lewis referring us to our well-thumbed copies of the ICO’s nearly-superseded PIA Handbook).

To an extent I can understand Tim Kelsey feeling he and his colleagues need to do more to communicate the benefits of care.data – after all, it’s their job to deliver it. But I do have real concerns that a senior officer at the ICO thinks that public concerns can be allayed through yet more plugging of the benefits, with none of the detailed reassurances and legal and technical justifications whose absence has been so strongly noted.

In passing, I note that, other than a message from their very pleasant Senior Press Officer for my blog, I have had no acknowledgement from the ICO of my request for them to assess the lawfulness of previous health data upload and linking.

UPDATE: 14.03.14

The ICO has kindly acknowledged receipt of my request for assessment, saying it has been passed to their health sector team for “further detailed consideration”.

1 Comment

Filed under care.data, Data Protection, data sharing, Information Commissioner, NHS

The FOI ministerial veto – why not?

The Court of Appeal has ordered disclosure of private correspondence between Prince Charles and the government. The judgment is potentially a triumph for transparency, but I have my doubts whether it reflects Parliament’s intentions when passing the FOI Act. And there will be a further appeal…

In September 2012 the Administrative Appeals Chamber of the Upper Tribunal (UT) handed down a judgment which struck me then, as it does now, as a remarkable work of research and scholarship. It was ruling on requests by the Guardian journalist Rob Evans – made as far back as April 2005 – under the Freedom of Information Act 2000 (FOIA) and the Environmental Information Regulations 2004 (EIR) for disclosure of information in private letters sent by the Prince of Wales to government ministers on matters of official policy. The UT’s judgment ran to 65 pages with three annexes, went into detailed analysis of constitutional conventions regarding the heir to the throne, and its decision was that the correspondence should be disclosed (overturning the prior decisions of the Information Commissioner (IC)). Subsequently, the Attorney General issued a certificate under section 53 FOIA – a “ministerial veto” – whose effect was to disapply the UT’s decision. The Attorney General’s certificate, in rather wider-spaced text, ran to ten pages.

Section 53 requires only that the accountable person (a minister)

gives the [Information] Commissioner a certificate signed by him stating that he has on reasonable grounds formed the opinion [that there had not been a failure to comply with the FOIA]

It is, as I’ve argued before , a bludgeon of an executive weapon, but it is, as are all acts of public authorities, potentially amenable to judicial review. So it was that, despite any statutory right of appeal, the Guardian made such an application. However, in July 2013, the High Court effectively decided that, although the ministerial power to override a superior court of record (let alone the statutory decision-maker, in the form of the IC) appeared to be a “constitutional aberration”, the proposition that “the accountable person is not entitled simply to prefer his own view to that of the tribunal” must be rejected. As Davis LJ said (para 111)

why not? It is inherent in the whole operation of s.53 that the accountable person will have formed his own opinion which departs from the previous decision (be it of Information Commissioner, tribunal or court) and may certify without recourse to an appeal. As it seems to me, therefore, disagreement with the prior decision (be it of Information Commissioner, tribunal or court) is precisely what s.53 contemplates, without any explicit or implicit requirement for the existence of fresh evidence or of irrationality etc. in the original decision which the certificate is designed to override

However, Davis LJ refused to accept that the wording of section 53 (“…stating that he has on reasonable grounds formed the opinion…”) permitted of an interpretation that:

the accountable person can, as it were, self-certify as to the availability of reasonable grounds

rather,

In my view, the language chosen clearly is sufficient to connote that an objective test is to be applied

But how to conduct that objective test? For Davis LJ, it must be that the reasonable grounds are “cogent”:

if an accountable person is to interfere, by way of exercise of the power of executive override, with the decision of an independent judicial body then that accountable person must be prepared and able to justify doing so. I am reluctant to talk in terms of burden of proof. But in terms of burden of argument the burden is in practice on the accountable person to show that the grounds for certifying are reasonable

Lord Dyson in the Court of Appeal has taken issue with this, saying (para 38) that

I do not consider that it is reasonable for an accountable person to issue a section 53(2) certificate merely because he disagrees with the decision of the tribunal. Something more is required […]
Examples of “something more” are given as
a material change of circumstances since the tribunal decision or that the decision of the tribunal was demonstrably flawed in fact or in law
Accordingly, as the Attorney General failed to give this “something more” but “simply disagreed with the evaluation made by the UT”, he failed to give reasons amounting to “reasonable grounds”. Thus (putting to one side a crucial other ground on which the appeal succeeded, relating to the EIR and European law, which I will deal with in a later blog post) the certificate had to be quashed.

As Dr Mark Elliot argues Lord Dyson here “adopted a significantly more exacting conception of reasonableness” than had the High Court and I would commend Dr Elliot’s piece to you as an expert analysis I am not competent to give.

However – and it pains me to say it, because I really don’t like section 53 – wasn’t it precisely Parliament’s intention that the accountable person did “merely” have to state that he had formed – on reasonable grounds – a different opinion to the preceding tribunal? If he cannot arrive at a different opinion, in the absence of “something else”, isn’t section 53 fundamentally weakened, even sidestepped? Indeed, Lord Dyson in my view arrives at this point, when he says

On the approach of the Divisional Court to section 53(2), the accountable person can override the decision of an independent and impartial tribunal which (i) is reasonable, (ii) is the product of a detailed examination (fairly conducted) of the issues after an adversarial hearing at which all parties have been represented and (iii) is not challenged on appeal. All that is required is that the accountable person gives sensible and rational reasons for disagreeing with the tribunal’s conclusion. If section 53(2) has that effect, it is a remarkable provision not only because of its constitutional significance (the point emphasised by the Divisional Court), but also because it seriously undermines the efficacy of the rights of appeal accorded by sections 57 and 58 of the FOIA
to which I am tempted to respond, adopting Davis LJ’s rhetorical device, “why not?” – that seems to have been what Parliament intended.

No doubt we shall see this explored more – the Attorney General is reported to have sought, and been given, leave to appeal to the Supreme Court.

1 Comment

Filed under Environmental Information Regulations, Freedom of Information, Information Commissioner, Uncategorized, Upper Tribunal

Reflections on the monetary penalty notice served on British Pregnancy Advisory Service

On 28 February the Information Commissioner’s Office (ICO) served a Monetary Penalty Notice (MPN), pursuant to powers under section 55A of the Data Protection Act 1998 (DPA), on the British Pregnancy Advisory Service, in the sum of £200,000 (which would be reduced to £160,000 if promptly paid). The ICO’s new release explains

An ICO investigation found the charity didn’t realise its own website was storing the names, address, date of birth and telephone number of people who asked for a call back for advice on pregnancy issues. The personal data wasn’t stored securely and a vulnerability in the website’s code allowed [a] hacker to access the system and locate the information.

The hacker threatened to publish the names of the individuals whose details he had accessed, though that was prevented after the information was recovered by the police following an injunction obtained by the BPAS

The back story to this is that the hacker in question was subsequently jailed for 32 months for offences under the Computer Misuse Act 1990 (no doubt the prosecutors recognised that the criminal sanctions under the DPA were too weedy to bother with).

The section 55A DPA powers are triggered where there has been a qualifying serious contravention by a data controller of its obligations under section 4(4) to comply with the data protection principles in Schedule One. The most pertinent of these in the instant case (and in the large majority of ICO MPNs) was the seventh

Appropriate technical and organisational measures shall be taken against unauthorised or unlawful processing of personal data and against accidental loss or destruction of, or damage to, personal data

which extends to the need to, when contracting with someone to process data on your behalf, require them to take equivalent security measures and evidence this contractual provision in writing. As the ICO’s MPN says

BPAS failed to take appropriate technical and organisational measures against the unauthorised processing of personal data stored on the BPAS website such as having a detailed specification about the parameters of the CMS to ensure that either the website did not store any personal data or alternatively, that effective and appropriate security measures were applied such as storing administrative passwords securely; ensuring stated standards of communication confidentiality were met; carrying out appropriate security testing on the website which would have alerted them to the vulnerabilities that were present or ensuring that the underlying software supporting the website was kept up to date

(Interestingly, the MPN also makes clear that there was a contravention of the fifth principle – which provides that “personal data processed for any purpose or purposes shall not be kept for longer than is necessary for that purpose or those purposes”. This was because “the call back details were kept for five years longer than was necessary for [BPAS’s] purposes”).

 The original crime was a particularly nasty one – the offender appears to have had an ideological, or at least personal, opposition to abortion in general, and the apparently very real threat to publish people’s details, given to BPAS in highly sensitive circumstances, is probably what elevated the BPAS contravention to a level which justifies such a high sum being served on a charity. However, BPAS have announced that they intend to appeal, and their press release about this is interesting. It suggests that the appeal will be not about the issuing of the MPN, but about its amount (section 55B(5) DPA permits appeals on either basis):

We accept that no hacker should have been able to steal our data but we are horrified by the scale of the fine

but it goes on to make the valid point that, by serving an MPN of this large amount, the ICO potentially gives the offender something that he wanted – to harm the charity:

 It is appalling that a hacker who acted on the basis of his opposition to abortion should see his actions rewarded in this way

This, though, seems to be a matter of ethics, rather than law, but it will be interesting to note if the argument makes it in some form into the grounds of appeal. More likely, if the challenge is to be made solely on the amount (under section 55B(5)(b)), focus will fall on to the suggestion that

This fine seems out of proportion when compared with those levelled against other organisations who were not themselves the victims of a crime

Of course, by a circular argument, the “fine” would not have been served, if the data controller had not, by its omissions, permitted itself to be a victim of the crime.

An extra frisson is caused when one considers the compelling argument by the solicitor-advocates for Scottish Borders Council, who successfully helped the latter win an appeal of an MPN last year. Although their argument – that MPNs were more correctly to be considered criminal, as opposed to civil, penalties – did not fall to be decided by the First-tier Tribunal, it did observe that

One general question hovering over this appeal is whether proceedings in respect of monetary penalties are “criminal” in nature. There are certainly enough indications, not least in the title of the amending statute, [the Criminal Justice and Immigration Act 2008] to make an arguable case for them being so…We have concluded that there is no need for us to make any decision or pronouncement in the abstract; but there is a need for us to be vigilant to ensure that the proceedings are fair

If this line of argument continues to be developed – that recipients of MPNs are entitled to be afforded the equivalent rights to fairness, of hearing under Article 6 of the European Convention on Human Rights, afforded to those accused of crimes – then MPNs, and the circumstances and manner in which they are served, may be subject to a much greater level of scrutiny, and the cash-strapped ICO may find itself under even more pressure from legal challenges.

These issues may be aired, and possibly determined, in the forthcoming appeal on the Upper Tribunal of the MPN served on Christopher Niebel, and subsequently overturned by the First-tier Tribunal.

2 Comments

Filed under Data Protection, human rights, Information Commissioner, Information Tribunal, monetary penalty notice

Health data breaches – missing the point?

Breaches of the DPA are not always about data security. I’m not sure NHS England have grasped this. Worse, I’m not sure the ICO understands public concern about what is happening with confidential medical information. They both need to listen.

Proponents of the care.data initiative have been keen to reassure us of the safeguards in place for any GP records uploaded to the Health and Social Care Information Centre (HSCIC) by saying that similar data from hospitals (Hospital Episode Statistics, or HES) has been uploaded safely for about two decades. Thus, Tim Kelsey, National Director for Patients and Information in the National Health Service, said on twitter recently that there had been

No data breach in SUS*/HES ever

I’ve been tempted to point out that this is a bit like a thief arguing that he’s been stealing from your pockets for twenty years, so why complain when you catch him stealing from your wallet? However, whether Tim’s claim is true or not partly depends on how you define a “breach”, and I suspect he is thinking of some sort of inadvertent serious loss of data, in breach of the seventh (data security) principle of the Data Protection Act 1998 (DPA). Whether there have been any of those is one issue, and, in the absence of transparency of how HES processing has been audited, I don’t know how he is so sure (an FOI request for audit information is currently stalled, while HSCIC consider whether commercial interests are or are likely to prejudiced by disclosure). But data protection is not all about data security, and the DPA can be “breached” in other ways. As I mentioned last week, I have asked the Information Commissioner’s Office to assess the lawfulness of the processing surrounding the apparent disclosure of a huge HES dataset to the Institute and Faculty of Actuaries, whose Society prepared a report based on it (with HSCIC’s logo on it, which rather tends to undermine their blaming the incident on their NHSIC predecessors). My feeling is that this has nothing, or very little, to do with data security – I am sure the systems used were robust and secure – but a lot to do with some of the other DPA principles, primarily, the first (processing must be fair and lawful and have an appropriate Schedule 2 and Schedule 3 condition), and the second “Personal data shall be obtained only for one or more specified and lawful purposes”).

Since the story about the actuarial report, at least three other possible “breaches” have come to light. They are listed in this Register article, but it is the first that has probably caused the most concern. It appears that the entire HES dataset, pseudonymised (not, note, anonymised) of around one terabyte, was uploaded to Google storage, and processed using Big Query. An apparently rather unconcerned statement from HSCIC (maybe they’ll blame their predecessors again, if necessary) said

The NHS Information Centre (NHS IC) signed an agreement to share pseudonymised Hospital Episodes Statistics data with PA Consulting  in November 2011…PA Consulting used a product called Google BigQuery to manipulate the datasets provided and the NHS IC  was aware of this.  The NHS IC  had written confirmation from PA Consulting prior to the agreement being signed that no Google staff would be able to access the data; access continued to be restricted to the individuals named in the data sharing agreement

So that’s OK then? Well, not necessarily. Google’s servers (and, remember “cloud” really means “someone else’s computer”) are dotted around the world, although mostly in the US, and when you upload data to the cloud, one of the problems (or benefits) is you don’t have, or don’t tend to think you have, a real say in where it is hosted. By a certain argument, this even makes the cloud provider, in DPA terms, a data controller, because it is partly determining “the manner in which any personal data are, or are to be, processed”. If the hosting is outside the European Economic Area the eight DPA principle comes into play:

Personal data shall not be transferred to a country or territory outside the European Economic Area unless that country or territory ensures an adequate level of protection for the rights and freedoms of data subjects in relation to the processing of personal data

The rather excellent Tim Gough who is producing some incredibly helpful stuff on his site, has a specific page on DPA and the cloud and I commend it to you. Now, it may be that, because Google has conferred on itself “Safe Harbor” status, the eight principle is deemed to have been complied with, but I’m not sure it’s as straightforward because, in any case, Safe Harbor itself is of current questionable status and assurance.

I don’t know if PA Consulting’s upload of HES data to the cloud was in compliance with their and NHSIC’s/HSCIC’s DPA obligations, but, then again, I’m not the regulator of the DPA. So, in addition to last week’s request for assessment, I’ve asked the ICO to assess this processing as well

Hi again

I don’t yet have any reference number, but please note my previous email for reference. News has now emerged that the entire HES database may have been uploaded to some form of Google cloud storage. Would you also please assess this for compliance with the DPA? I am particularly concerned to know whether it was in compliance with the first, seventh and eighth data protection principle. This piece refers to the alleged upload to Google servers http://t.co/zWF2QprsTN

best wishes,
Jon

However, I’m now genuinely concerned by a statement from the ICO, in response to the news that they are to be given compulsory powers of audit of NHS bodies. They say (in the context of the GP data proposed to be uploaded under the care.data initiative)

The concerns around care.data come from this idea that the health service isn’t particularly good at looking after personal information

I’m not sure if they’re alluding to their own concerns, or the public’s, but I think the statement really misunderstands the public’s worries about care.data, and the use of medical data in general. From many, many discussions with people, and from reading more about this subject than is healthy, it seems to me that people have a general worry about, and objection to, their confidential medical information possibly being made available to commercial organisations, for the potential profit of the latter, and this concern stems from the possibility that this processing will lead to them being identified, and adversely affected by that processing. If the ICO doesn’t understand this, then I really think they need to start listening. And, that, of course, also goes for NHS England.

*“SUS” refers to HSCIC’s, and its predecessor, NHSIC’s Secondary Uses Service

4 Comments

Filed under care.data, Data Protection, data sharing, Information Commissioner, NHS

Why no prison sentences for misuse of medical data?

So, the government, roused from its torpor by the public outrage at the care.data proposals, and the apparent sale of 47 million patient records to actuaries, is said to be proposing, as a form of reassurance, amendments to the Care Bill. The Telegraph reports that

Jeremy Hunt will unveil new laws to ensure that medical records can only be released when there is a “clear health benefit” rather than for “purely commercial” use by insurers and other companies.

Ministers will also bolster criminal sanctions for organisations which breach data protection laws by disclosing people’s personal data. Under a “one strike and you’re out” approach, they will be permanently banned from accessing NHS data

One needs to be aware that this is just a newspaper report, and as far as I know it hasn’t been confirmed by the minister or anyone else in the government, but if it is accurate, I fear it shows further contempt for public concerns about the risks to the confidentiality of their medical records.

The first of the reported amendments sounds like a statutory backing to the current assurances that patient data will only be made available to third parties if it is for the purposes that will benefit the health and social care system (see FAQ 39 on the Guide for GP Practices). It also sounds like a very difficult piece of legislation to draft, and it will be very interesting to see what the proposed amendment actually says – will it allow secondary use for commercial purposes, as long as the primary use is for a “clear health benefit”? and, crucially, how on earth will it be regulated and enforced? (will properly resourced regulators be allowed to audit third parties’ use of data? – I certainly hope so).

The second amendment implies that the Data Protection Act 1998 (DPA) will also be amended. This also sounds like a difficult provision to draft: the Telegraph says

Those that have committed even one prior offence involving patient data will be barred from accessing NHS medical records indefinitely as part of a “one strike and you’re out” approach

But what do we mean by “offence”? The Telegraph falls into the common error of thinking that the Information Commissioner’s Office’s (ICO’s) powers to serve monetary penalty notices (MPNs) to a maximum of £500,000 are criminal justice powers; they are not – MPNs are civil notices, and the money paid is not a “fine” but a penalty. The only relevant current criminal offence in the DPA is that of (in terms) deliberately or recklessly obtaining or disclosing personal data without authority of the data controller. This is an either-way offence, which means it currently carries a maximum sanction of a £5000 fine in a magistrates court, or an unlimited fine in Crown Court (it is very rare for cases to be tried in the latter though). Prosecutions under this section (55) are generally brought against individuals, because the offence involves obtaining or disclosing the data without the authority of the data controller. It is unlikely that a company would commit a section 55 offence. More likely is that a company would seriously contravene the DPA in a manner which would lead to a (civil) MPN, or more informal ICO enforcement action. More likely still is simply that the ICO would have made a finding of “unlikely to have complied” with the DPA, under section 42 – a finding which carries little weight. Are prior civil or informal action, or a section 42 “unlikely to have complied” assessment going to count for the “one strike and you’re out” approach? And even if they are, what is to stop miscreant individuals or companies functioning through proxies, or agents? or even simply lying to get access to the data?

Noteworthy by its absence in the Telegraph reports of the proposed amendments was any reference to the one change to data protection law which actually might have a deterrent effect on those who illegally obtain or disclose personal data – the possibility of being sent to prison. As I and others have written before, all that is needed to achieve this is for the government to commence Section 77 of the Criminal Justice and Immigration Act 2008, which would create the power to alter the penalty (including a custodial sentence) for a section 55 DPA offence. However, the government has long been lobbied by certain sections of the press industry not to do so, because of apparent fears that it would give the state the power to imprison investigative journalists (despite the fact that section 78 of the Criminal Justice Act 2008 – also uncommenced – creating a new defence for journalistic, literary or artistic purposes). The Information Commissioner has repeatedly called for the law to be changed so that there is a real sanction for serious criminal data protection offences, but to no avail.

Chris Pounder has argued that the custodial sentence provisions (discussion of which was kicked into the long grass which grew up in the aftermath of the Leveson inquiry) might never be introduced. Despite the calls for such strong penalties for misuse of medical data, from influential voices such as Ben Goldacre, the proposals for change outlined by the Telegraph seem to support Dr Pounder’s view.

One of the main criticisms of the disastrous public relations and communications regarding the care.data initiative is that people’s acute concerns about the security of their medical records have been dismissed with vague or misleading reassurances. With the announcement of these vague and probably ineffectual proposed legal sanctions, what a damned shame that that looks to be continuing.

3 Comments

Filed under care.data, Data Protection, data sharing, Information Commissioner, Leveson, monetary penalty notice, NHS