When you’re sitting opposite your GP at your local surgery and he or she is accompanying their well-meaning diagnosis and advice with the melodic tap-tap-tapping on their keyboard, do you ever really wonder what happens to the information that they are inputting, where it goes or in what form it is seen?
Needless to say, the details related to your medical history do need to have the freedom to pass beyond the limitations of your local surgery. If you were to have a referral to other specialists, you would want them to know your details rather than have to narrate your entire medical history to every new face you see. Likewise, if you were to have an emergency admission to hospital and be unable to speak for yourself, you would probably rather that they know if you have, for example, an allergy to penicillin, such that their ‘treatment’ doesn’t result in rashes and face-swelling to accompany any misfortune that had accompanied you to A + E originally.
However, not all forms of medical data-usage refer to such direct means of treatment that benefit you individually. Instead, your data can be used for a number of secondary purposes such as, for example, monitoring the performance of medical professionals or using large data-sets to analyse the effects and side-effects of medications. The care.data programme refers to one such usage of medical data that has been subject to much controversy. Care.data refers to a specific data-usage programme using secondary data provided by patients through their engagement with the NHS. It refers to the usage of data for research purposes that is analysed to assess performance of the NHS and to improve understanding of conditions and treatments over time. Names of patients are not given as part of the care.data programme but individuals could easily be identifiable by data that IS contained such as NHS numbers, date-of-births and full postcodes.
The benefits of the care.data programme are clear. The large amounts of data would enable detailed analysis of trends related to disease prognosis and effectiveness of treatments but, if you were to think that all the numerical fun and games would be restricted to within the amorphous walls of the NHS, you would be mistaken because, alas, the proposal would include the handing out of this data to interested third-parties. Whilst rigorous control procedures are designed to ensure that this data will not fall into the hands of anyone who sends in a plea (and it could certainly be argued that it would be beneficial to allow certain third-parties to have access to data, e.g. pharmaceutical companies), it remains to be seen how these controls will be enforced and, given the not-so-easily-forgotten media-revelations of medical data falling in the wrong hands, one could be forgiven for reserving some cynicism in this regard.
I recently had the opportunity to run ten care.data workshops whereby an NHS London representative was put before rooms full of older Londoners in order to detail what the programme involves and then to field any questions that were to follow. To his credit, the NHS London representative was candid and responded to all questions as well as he was able, neither joining in with the more vocal critics in the crowd nor trying to defend the programme against all criticism.
The two key issues highlighted were as follows:
1. At present, and in contrast to most research undertaken within the NHS, the proposal is for this to be an opt-out rather than a consent-based scheme. Essentially, this means that all individuals’ data WILL be used through the care.data programme unless they pro-actively head down to and opt-out at their local surgery. The argument for this is that the scheme requires large data-sets and, due to public tendency to general passivity as well as the aforementioned lack of faith in data-confidentiality, an opt-in scheme is unlikely to get the required numbers. By bucking the normal trend of making research ‘opt-in’, however, this is likely to increase suspicions and adversely impact upon public view of the NHS.
2. Confidentiality and control of access to data are clearly big issues and more details of precisely who would be permitted access along with the controlling mechanisms are currently lacking. Whilst the identifiability of patients provides better data from which to analyse outcomes over time, it does appear that any risk-assessment of this conception of the care.data programme has not focused on the legitimate concerns of the public in relation to how their information is stored and used.
Ultimately, any final decision regarding the care.data programme will need to strike a balance between the public view of what is acceptable use of their data and the research-orientated perspective of the data-set requirements necessary to draw meaningful conclusions from analysis. Where the final line is drawn will give an interesting insight into the relative priorities held by those making the decisions.