Abstract
In this article I review the principal arguments in favour of and against the UK government’s recent proposals to allow access to NHS patient records to life sciences companies as part of the NHS–Life Sciences partnership scheme.
Introduction
Towards the end of 2011 the UK Government announced that under its proposed NHS–Life Sciences partnership NHS patient records held by both hospitals and GP practices will be made available, in an anonymized format, to healthcare companies in the private sector for the purposes of assisting and advancing medical research (BBC, 2011; Sparrow, 2011; Walsh, 2011). Or to quote Sir Paul Nurse, president of the Royal Society, ‘What I think is most interesting about this is the potential for a cultural change at the National Health Service. To consider it not just as an organization delivering care, but as a fantastic research resource as well. That could make a very significant impact’ (Feilden, 2011).
However, this proposal has been criticized on the grounds that it will endanger patient confidentiality and that it represents an inappropriate deepening of the NHS’s relationship with the private sector. 1
I will begin by setting out and discussing the principal arguments in favour of pursuing this scheme. I will then consider the principal argument that has been advanced against these proposals, namely that they will endanger patient confidentiality. I also look in detail at the proposal for the system to operate a model of presumed consent, borrowing arguments from the organ donation, and assess whether this would be acceptable.
Why are the proposals a good thing?
Even after a brief survey of the media discussion of these proposals it is possible to pick out three principal arguments that have been advanced in support of allowing access to NHS records under the NHS–Life Sciences partnership. These can, broadly speaking, be characterized as follows:
The proposals will be economically beneficial.
The proposals will benefit patients.
The proposals allow people to fulfil their duties to participate in research.
The first two arguments could be characterized as being consequentialist in nature in so much as they both rest on the notion that allowing access to patient records, under the NHS–Life Sciences partnership, will bring about considerable benefits to the UK as whole without, so it has been claimed, any deleterious consequences. Or to put it another way, the good consequences of pursuing the proposals for patient data access far outweigh the negative consequences pursuing such a plan.
As an interesting aside, it is possible to view both of these arguments through the prism of ethical egoism in so much as both, as we shall see, can be expressed as appeals to enlightened self-interest. That is to say, both can be construed as claims that it is in our own self-interest to approve of the proposals because each of us could potentially benefit personally from them.
So what lies at the heart of these arguments?
1. The proposals will be economically beneficial
At the centre of the first argument is a claim that allowing access to patient records, as part of the NHS–Life Sciences partnership, will bring about economic benefits to the UK. It has been estimated that the life sciences/medical research sector is worth around 50 billion pounds to the UK economy and employs around 160,000 people 2 and has a strong expertise base in both the private and public sectors via, for example, industry and the university sector (BBC, 2011; Walsh, 2011). As such, the primary goal of the NHS–Life Sciences partnership is not only to maintain this boon for the UK economy but to grow it (Feilden, 2011).
So how will the provisions for allowing access to patient records help support this strategy? The answer to this question lies in the sheer amount of patient data the NHS both generates and holds. 3 Now when seen in the context of a resource the patient records held by the NHS can be utilized to the economic benefit of the UK in two ways: a) by encouraging companies to either stay in, or relocate to, the UK so that they can access this resource to benefit themselves and in turn benefit the UK economy through employment and wealth generation; and b) it has been claimed that allowing access to patient records will ‘speed up’ the research project, thus speeding up the time needed to take a drug from concept to market.
2. The proposals will benefit patients
As was noted above, central to the plan to allow access to patient data is the claim that this will ‘speed up’ the research process, and thus increase the speed at which new treatments are developed and deployed. Alongside these financial benefits to the country the faster research process would benefit patients. This argument can be characterized as follows: allowing access to patient records will ‘speed up’ research, thus ‘speeding up’ the development and delivery of new treatments. This in turn will both increase the number of treatment options for patients and improve treatments themselves. Therefore, if we assume that increased and better treatment options for patients is beneficial for patients, then allowing access to patient records under the NHS–Life Sciences partnership is easier to justify.
3. Allowing fulfilment of duties to participate in research
Harris has argued that we have a moral obligation to support and/or participate in medical research (Harris, 2005). Now if we assume that Harris is correct, then making patient records accessible under the NHS–Life Sciences partnership might provide a means by which this duty could be discharged.
Briefly, the notion that there might be a duty to participate in and/or support medical research can be derived from two, albeit closely linked, arguments. The first appeals to the idea of beneficence and/or the rule of rescue in so much as it holds that if it is within our power to help those in need, or prevent, minimize, or relieve a particular harm, and doing so will not be overly burdensome to us then we ought to do so. For example, my friend is currently waiting to move into his new house when he is told that there will be a delay. However, because my friend had assumed he would be able to move into his new home on time he has already given notice on his rental property. Unfortunately, this means that between leaving his rental property and moving into his house my friend has a week in which he will be effectively homeless. Fortunately, I have a spare room in my own house and, having recognized that he is in need of somewhere to stay, and knowing that inviting him to stay in my spare room will be of no burden to me, I invite him to stay with me. Now with regard to medical need Harris notes that, ‘Most, if not all diseases create needs, in those who are affected, and in their relatives, friends and carers and indeed society. Because medical research is a necessary component of relieving that need in many circumstances, furthering medical research becomes a moral obligation’ (Harris, 2005: 242).
The second argument is, broadly speaking, based on an appeal to fairness, although one could also characterize it as an appeal for solidarity. The notion here is that as we have all benefited from medical research, and will continue to do so, and have accepted these benefits then we ought to support them. As Harris notes, ‘Since we accept these benefits [the benefits of medical research], we have an obligation in justice to contribute to the social practice which produces them [i.e. medical research]’ (Harris, 2005: 243). To not do so is, claims Harris, to be guilty of being a ‘free-rider’, someone who enjoys a particular benefit without contributing to it. For example, a man who deliberately avoids paying his taxes yet continues to utilize the public services they pay for could be said to be unjustly free-riding on the goodwill of everyone else.
Therefore, if one takes seriously the idea that we have a moral obligation to contribute to medical research, then the plan to make our medical records accessible, under the NHS–Life Sciences partnership, would appear to be an ideal way to discharge that duty 4 with minimal cost to ourselves.
Why might these proposals be a cause for concern?
From the brief overview above the arguments I have presented, with their appeals to enlightened self-interest, civic duty, economic benefit, and improving the care of the sick make a compelling case in favour of allowing access to patient records under the NHS–Life Sciences partnership. However, when the proposals were announced they drew criticisms from a number of quarters. The central objections to the proposals to grant access to patient records arise from a concern for patient’s rights to confidentiality. For example, in an interview with the BBC a spokesman for the campaign group patient concern, Roger Goss, stated that the proposals were ‘[T]he death of patient confidentiality’ (BBC, 2011).
The objections can be classified as follows: 1) patients’ information will no longer be confidential; and 2) the proposal for a system of presumed consent is not acceptable.
1. Patients’ information will no longer be confidential
The thought here is that granting access will lead to a loss, or indeed end of, patient confidentiality. While currently third-party access to an individual’s medical records is restricted and controlled, the new proposals would allow many more people to access your patient records (Walsh, 2011).
This claim has been strongly rebuked by the government, which has insisted patient records will be made anonymous so as to safeguard patient confidentiality 5 (Walsh, 2011). However, concern remains as it is thought that the current laws and regulation surrounding data protection are not robust enough to ensure patient confidentiality will be preserved under the new access proposals. It is also feared that it might be possible to ‘reverse identify’ anonymous data and thus learn the identity of individual patients. The concern here is not with the rationale which underpins the proposals (that the information held by the NHS could assist medical research) but rather the proposed method by which access would be facilitated.
Certainly, for the proposals to work without jeopardizing patient confidentiality there would need to be robust protections in place. Details of how exactly the proposals could be implemented are beyond the scope of this article, and I will assume that such problems could be solved with appropriate regulation.
2. The proposal for a system of presumed consent is not acceptable
The default consent method in research is the classic tri-partite model in which a voluntary and competent patient, having received full and clear information, decides whether or not to participate. 6 The NHS–Life Science partnership proposals would operate a system of presumed consent. That is, unless an individual explicitly opts out it will be assumed that they are happy to participate and thus their records will be made available under the scheme. Why might people object to a system of presumed consent for the purposes of facilitating access to patient records under the government’s proposed scheme?
Objections to the use of presumed consent for the purposes of the government’s record access scheme may well be similar to objections that are levelled against the use of presumed consent in organ donation and which, according to Cohen (1992), can be classified as either being a) mechanical objections, b) sociologic objections or c) moral objections.
Mechanical objections are those that focus on the administrative, legal and regulatory difficulties of orchestrating a presumed consent system. As was noted earlier, concerns regarding regulation fall outside the remit of this article and I will assume that such problems could be solved with appropriate regulation.
In the case of sociologic objections Cohen, albeit from the perspective of the organ donation debate, characterizes these as being, ‘Objections [that are] rooted in the fear of attitudes or practices that the new system will (allegedly) promote’ (Cohen, 1992: 2171). In the case of the patient record access proposals the principal fear is that in the future access to patient records might be extended further either by granting access to non-medical research organizations, such as health insurance companies, and/or allowing data to be accessed in a non-anonymous form, allowing medical companies to, for example, directly market to patients.
This objection can be characterized as a slippery slope and/or an appeal to the precautionary principle. In other words the proposals for patient record access should not be allowed because it could potentially lead to deleterious consequences later on.
The problem with objections that invoke slippery slopes and/or precautionary principles is the uncertainty which surrounds the phrase ‘could potentially lead to deleterious consequences’. The reason why this is a problem is that saying that something ‘could happen’ is not really very helpful unless we can quantify that ‘could’ in some meaningful way. After all, with enough thought we could ascribe almost anything with possible deleterious consequences. However, when deciding whether to allow a certain practice we usually require something more solid upon which to base our decisions. We would also need to agree that the possible deleterious consequences would be so grave that it would be better to forgo the benefits of taking action (in this case introducing the government’s proposals). It has been suggested that ‘slippery slopes’ reasoning can stifle progress and bring about harm by encouraging policy-makers and researches to be ultra conservative and overly cautious (Harris, 2005; Hughes, 2006). In this case, being overly cautious would be to miss out on the potential usefulness that the scheme could bring.
Even so, this sociological objection is useful for drawing attention to the outcomes we ought to be careful to avoid in the pursuit of the good associated with the patient record access proposals.
Regarding moral objections relevant to the government’s proposals, there is a ‘critical difference between consenting and not objecting’ (Cohen, 1992: 2171). A system of presumed consent would not be conducive to respect for an individual’s autonomy because it is built on a false assumption, namely that the absence of an objection implies, or can be equivocated with, consent. Or when participation is on the basis of active informed consent from the individual we could be more certain that each individual participating wanted to, and has been given the opportunity to express that desire. As Cohen notes, ‘If persons do in reality object … a system that requires express consent will protect his autonomy more surely than the revised system [i.e. a system of presumed consent]’ (Cohen, 1992: 2172).
Of course, if the government’s system was based on active, expressed consent instead of presumed consent the objection would no longer stand. However, such a change might prove problematic for the government’s proposals, for reasons similar to those in the organ transplant debate. The problem for organ transplantation is that while surveys indicate that the majority want their organs to be used very few formally register their wishes (English, 2007). Consequently, the number of organs available is reduced because under an active/expressed consent system the default option when we do not know an individual’s preferences is to assume that they do not wish for their organs to be used. Likewise, we could envisage a similar problem occurring with the patient records system in so much as, if the majority were in favour, for the reasons cited in the previous section, but they failed to register those preferences the default position, to assume they do not wish to participate may make the number involved so small as to negate any advantage.
In the case of organ transplantation, Cohen argues that given the majority preference for transplantation, a system of presumed consent is in fact more conducive to ensuring that people’s preferences are expedited. As he notes, ‘If we knew that only 1 or 2 persons in 10 would automatically donate their organs, a system that presumed consent, protecting only 10% automatically but obliging the other 90% to register their objections … would be unfair. But if we have good reasons to believe that 7 or 8 of 10 … would in fact choose to donate their organs … a system that presumes the absence of consent (what we have now) similarly protects a minority and obliges the majority to register their views expressly, and it is unfair’ (Cohen, 1992: 2172).
The same could be said of the government’s patient records access proposals. An empirical study would need to be carried out to find out whether operating the government’s scheme on the basis of presumed consent would better fulfil preferences than a system of active consent.
Given this, I am inclined to agree with Cohen’s assertion that rather than casting the argument between presumed and active/expressed consent as a moral one it is in fact an operational argument with moral consequences. As Cohen notes, ‘The merit of proceeding in the one way or the other depend largely upon what we believe to be the general inclination’ (1992: 2172).
Summary
I have given an overview of the principal arguments in favour and against the government’s proposals to allow access to NHS patient records as part of the wider NHS–Life Sciences partnership.
The arguments in favour are compelling and there is a strong case that pursuing such a scheme would be in our interests (both as individuals and for the country as a whole). However, as the objections illustrate, the case in favour of pursuing the government’s proposals on patient record access is only as strong as the operational and regulatory systems required to facilitate it.
Footnotes
Acknowledgements
I would like to thank Dr Zuzana Deans (University of Bristol) for her helpful comments.