The NHS has opened a public consultation on "the wider use of patient information", which essentially boils down to the sharing of anonymised and partially anonymised patient data for research purposes - potentially with commercial and other bodies beyond the confines of the NHS. This is a subject that is likely to arouse strong reactions from all concerned, and is not likely to go away. Patient data has, as the consultation paper rightly states, been used for planning and research purposes for a long time, but the modern techno-health machine relies more and more on statistical research, particularly to underpin decisions on allocating increasingly tight budgets as treatment costs escalate, so non-clinical uses of patient data are likely to be extended significantly. This is all the more likely as patient records get centralised, making such uses easier to accomplish.

 

So what's in this consultation paper? Once I had recovered from the adverse initial impression resulting from an animated NHS logo on every page of the document (yes, really - it jiggles about like a hula dancer on speed), I was quite impressed by the consultation document. Unlike many such consultations emanating from government departments, this one seems to be driven by a genuine desire to obtain the views of the public at large - the questions do not seem to be loaded. Whether the answers actually influence the ultimate decisions is of course another question we will have to wait until February next year to find out the answer to.

 

Various appoaches to data handling are described in the consultation paper. "Sealed envelopes" within the medical record of an identifiable person can hide certain information but will show that something is concealed. Alternatively the record can contain "Sealed and locked" information, the very existence of which is hidden. Anonymisation, which does not conceal the facts but attempts to remove personal identifying information from the factual record, supposedly breaks links between independent records for the same individual. A half-way house to anonymisation merely replaces the personal identity with an arbitrary token but otherwise keeps the whole of all records intact. The public is being asked to give their views of the pros and cons of these approaches. From my perspective, this is asking quite a lot. The issues are complex, even supposing the proposed systems themselves are robust.

 

One of the biggest concerns will be the robustness of the processes - something our government is not particularly good at getting right to date. Informed consent and opt-out are also likely to be bones of contention. Will the "seals" and "locks" remain intact under all circumstances, and if not, who has control? Ultimately, the question "who owns your medical records?" has not yet been satisfactorily answered, and until it has there will always be legitimate concerns about their use in other than a purely therapeutic context.

 

Until there are clear undertakings that no more than the necessary information will be abstracted from the record for any given purpose, that its confidentiality will be protected, and there is clear evidence that such undertakings are being fulfilled consistently, we should worry. The very same considerations went by the board in the infamous HMRC incident - a data set containing massive amounts of redundant sensitive information was mishandled and leaked - for no other reason than departmental cost-cutting. So in this case, I firmly believe the time is not right for increased information sharing, however it is implemented - simply because the government and its contractors have proved themselves both incompetent at data security and uncaring as to the repercussions of their incompetence. They may get their act together - we must hope so - but until they do, can we really afford to acquiesce to increased circulation of our sensitive medical records? I say "No" emphatically - "not on your life!" - it just might be.

Permalink:
http://www.infosecurityadviser.com/view_message?id=82