Judgment in Department of Health v IC [2011] EWHC 1430 (Admin) – the ‘abortion statistics’ appeal – was handed down on 20 April this year. Cranston J’s judgment has now been made available. The following salient points from that judgment may be of use to those interested in the concept and extent of ‘personal data’ under s. 40 FOIA and the DPA – especially when looking at the grey area of statistics or other anonymous data which is rooted in or derived from other data which is more overtly personal. The judgment is also essential reading for anyone grappling with the application of the leading House of Lords decision on this subject, Common Services Agency v Scottish Information Commissioner [2008] UKHL47, [2008] 1 WLR 1550 (‘CSA’). (‘Grappling’ is probably apt: even Cranston J conceded that “it would be wrong to pretend that the interpretation of the CSA case is an easy matter”).
Briefly by way of background: the Department refused a request for detailed statistics on the number of late-term abortions carried out on prescribed grounds. It relied on s. 40 FOIA, basing its case on the risk that, given the ‘low cell counts’ in these categories, the relevant patients and/or doctors might be identified by those sufficiently motivated to do so. The Commissioner found that these statistics were not personal data. The Information Tribunal agreed with the Department that they did constitute personal data, but was not satisfied that s. 40 was effective, as there was insufficient risk of identification.
On the Department’s appeal to the High Court, Cranston J agreed with the Commissioner that these statistics are not personal data.
One route to that conclusion was that advocated by the Commissioner, namely to adopt the approach of Baroness Hale in CSA: anonymised statistics remain personal data and therefore subject to the protection of the DPA in the hands of the data controller (who possesses the underlying data from which individuals could be identified) but not in the hands of the general public (who do not). This approach commended itself to the Upper Tribunal in the recent case of All Parliamentary Group on Extraordinary Rendition v Information Commissioner [2011] UKUT 153 AAC (on which, see my post here).
Cranston J, however, rejected that route, as it was the reasoning of Lord Hope rather than Baroness Hale in CSA which had attracted the majority’s support in that case. Lord Hope’s approach can be paraphrased as follows. The definition of personal data under s. 1 DPA provides for two means of identification: either from the data itself (inapplicable in the case of anonymous statistics) or from “from those data and other information which is in the possession of, or is likely to come into the possession of, the data controller”. Lord Hope’s approach to situations such as this is to ask: does the ‘other information’ (if provided to the hypothetical member of the public) add anything to the statistics which would enable them to identify the underlying individuals? If the answer is no, the statistics are not personal data. The underlined words are important: if identification can be achieved from the ‘other information’ in isolation (rather than when added to the statistics) then the statistics themselves are truly anonymous, and are not personal data. The statistics in this case failed Lord Hope’s test, and were thus not personal data.
Cranston J’s conclusion was that the Tribunal had been correct to conclude that the data was ‘truly anonymised’ – but it had erred in treating this as personal data which had been truly anonymised. The Department contended that, because it held the underlying identification data, the abortion statistics remained personal data in all circumstances. Cranston J rejected this submission, stating that:
“If that were the case, any publication would amount to the processing of sensitive personal data… Thus, the statistic that 100,000 women had an abortion in a particular year would constitute personal data about each of those women, provided that the body that publishes this statistic has access to information which would enable it to identify each of them. That is not a sensible result and would seriously inhibit the ability of healthcare organisations and other bodies to publish medical statistics”.
In going on to dismiss the Department’s other grounds of appeal, Cranston J made a number of other points of general application. For example, in rejecting the criticism that the Tribunal had failed adequately to engage with the Department’s expert evidence, Cranston J said this:
“To begin, the issue before the Tribunal was one of assessment: the likelihood that a living individual could be identified from the statistics. That was in my judgment only partly a question of statistical expertise, as regards matters such as the sensitivity of the data. Partly, also, it was a matter of assessing a range of every day factors, such as the likelihood that particular groups, such as campaigners, and the press, will seek out information of identity and the types of other information, already in the public domain, which could inform the search. These are factors which the Tribunal was in as good a position to evaluate as the statistical experts, a point which one of the Department of Health’s experts conceded. The analysis also applies to the evidence of senior civil servants.”
As regards the Department’s contentions that conditions from Schedules 2 and 3 of the DPA were not met, their points were “wounding” to the Tribunal’s judgment, but not “fatal”, in light of the evidence at the Tribunal hearing. Finally, Cranston J described the Department’s argument based on Article 8 ECHR as “very much a jury argument”.
Interestingly, on the same day as judgment was given in this case, the High Court (Kenneth Parker J) gave judgment in R (BT & Anor) v The Secretary of State for Business, Innovation and Skills [2011] EWHC 1021 (Admin) – BT’s unsuccessful application for judicial review of the Digital Economy Act 2010 (on which, see my piece here). One of the grounds of challenge was alleged non-compliance with the Data Protection Directive. In that judgment, IP addresses (anonymous strings of numbers linked to internet subscribers’ accounts) were treated as personal data even in the hands of copyright owners who possessed only those IP addresses. This was by application of the definition of personal data under the Directive: here copyright owners were deemed likely to come into possession of the underlying personal data when taking legal action against the individual internet subscribers who downloaded content in breach of copyright. This conclusion was reached independently of the Lord Hope test. Note, however, that it seems from the judgment that this question – are IP addresses always personal data or not – was not argued in full before Kenneth Parker J. There is talk of a potential appeal, so the application of these principles to IP addresses might be considered in the courts again before too long.
Robin Hopkins