- Research article
- Open Access
- Open Peer Review
Pharmaceutical information systems and possible implementations of informed consent - developing an heuristic
© Ploug and Holm; licensee BioMed Central Ltd. 2012
- Received: 26 April 2012
- Accepted: 17 October 2012
- Published: 16 November 2012
Denmark has implemented a comprehensive, nationwide pharmaceutical information system, and this system has been evaluated by the Danish Council of Ethics. The system can be seen as an exemplar of a comprehensive health information system for clinical use.
The paper analyses 1) how informed consent can be implemented in the system and how different implementations create different impacts on autonomy and control of information, and 2) arguments directed towards justifying not seeking informed consent in this context.
Results and Conclusion
Based on the analysis a heuristic is provided which enables a ranking and estimation of the impact on autonomy and control of information of different options for consent to entry of data into the system and use of data from the system.
The danger of routinisation of consent is identified.
The Danish pharmaceutical information system raises issues in relation to autonomy and control of information, issues that will also occur in relation to other similar comprehensive health information systems. Some of these issues are well understood and their impact can be judged using the heuristic which is provided. More research is, however needed in relation to routinisation of consent.
- Health information
- Health information systems
- Informed consent
Information systems holding data on all patients in a given health care system are being developed worldwide. The most ambitious plans envisage a system holding the complete patient record with all attendant data (lab results, imaging data, genetic data etc.) within the system, but such a system has not yet been implemented in any large scale health care system. Such information systems raise many ethical and legal issues. In the present paper we analyse some of these issues, especially issues concerning informed consent. The starting point is a nationwide pharmaceutical information system that has already been implemented in Denmark. By focusing on a concrete, already existing system we hope to be able to provide an analysis based more in clinical and administrative reality than in speculation about the future.
There is a substantial literature on the use of routinely collected patient data for research purposes, but the literature on the ethical issues raised by the collection and use of comprehensive patient data for treatment and diagnostic purposes is more limited. The so far unsuccessful efforts of the English National Health Service to share patient information electronically across all services has generated some ethical debate, [1–4] as has similar initiatives in other countries . In this paper we hope to deepen this debate by showing that issues of consent in this context may be considerably more complicated than they may initially appear. There are many differences between the research context and the clinical context and many of them are relevant in relation to the kind of information systems we analyse here. Although the information in these systems is very useful for research purposes, the primary stated aim of implementing the systems is not to provide knowledge, or to provide general public benefit, but to provide direct benefit to individual patients. These systems involve people’s health related interests much more directly than most research databases do. Another very important difference is that the main purpose of implementing the system necessarily involves using the information it contains, and the inferences that can be drawn from this information in a person-identifiable way. Anonymity is not an option and privacy interests can therefore not be protected by anonymising. This entails that some of the analyses provided in the literature on research databases may not be transferable to the present, clinical context.
The paper falls in 3 sections. We first briefly describe the comprehensive Danish pharmaceutical system and the reasons given for implementing it. In the second part we analyse issues of autonomy and informed consent raised by the introduction of the system and provide a heuristic for classifying these issues in a three dimensional space of relevant considerations. In this second part of the paper we also discuss some arguments raised by the Danish Council of Ethics and show how they can be encompassed by our heuristic. The final part of the paper then considers whether the tension between respect for autonomy and the reasons for introducing the system generate a true ethical dilemma, and discuss further problems created by a possible routinisation of informed consent.
The paper provides a heuristic for analysing these issues which is of value for anyone who thinks that autonomy is important, because the heuristic ranks possible models for informed consent in relation to their impact on autonomy. We do not purport to ‘solve’ the ethical issues since that would require a determination of exactly how important personal autonomy is in relation to all other relevant ethical considerations and values. Such a determination is outside the scope of this paper. Our heuristic is never the less of value, since it provides a detailed account of the effect of different consent arrangements on one side of the balancing between autonomy and other considerations. This should allow decision-makers to have a clearer view of what they are sacrificing in choosing consent arrangements that do not fully protect autonomy.
The shared medicine profile
Since the passing of the law on personal, electronic, medicine profiles in 2003, the Danish Medicines Agency – an agency under the Ministry of Interior and Health – has registered the personal “use” of prescription medicine among Danes on the basis of reports from all Danish pharmacies (see below for a discussion of “use”) [6, 7]. The resulting database feeds information into two separate personal records, the Medicine Profile and the Shared Medicine Profile.
The Medicine Profile is an electronic record containing information about the medicine prescribed for and bought by an individual within the last two years. Medicine prescribed or given during in-patient stays is also registered. The registration is automatic and compulsory. When the law was passed in 2003, pharmacists, doctors and their assistants, the patient and the Danish Medicine Agency were granted access to the records. The Medicine Profile was introduced with the primary purpose of providing doctors, patients and pharmacies with an overview of the medication prescribed for or given to an individual thereby enabling an improved and consistent use of medication leading to benefit for the individual and a reduction in public spending [6, 8]. The Medicine Profile was implemented in 2004.
The Shared Medicine Profile is an electronic record collecting and keeping up to date information about a patient’s medical treatment within the last two years across electronic health records in hospitals and general practitioner practices, electronic municipal records of care for the elderly and the Medicine Profile. The Shared Medicine Profile thus contains information about all prescribed medication and any medication provided by health care personnel along with information about the indication on the basis of which the medication is prescribed or provided. It also includes information about any medication bought by the individual apart from over-the-counter medication, the prescribed daily intake and the date of termination of the medical treatment. The health care personnel’s’ instructions of use are also registered along with notes concerning any known non-adherence to treatment, intolerance or allergy. Patients can also themselves register the use of any non-prescription medication through their electronic access to the Medicine Profile. The Medicine Profile and the Shared Medicine Profile both attempt to provide a record of the medication that is used by a particular patient, but because the vast majority of use in the community is unsupervised there will be a, potentially significant discrepancy between what is prescribed and bought and what is actually used. This is only partly alleviated by the possibility of patients themselves adding information to the Shared Medicine Profile, so the official claim that the two profiles show medication “use” is somewhat dubious. The Shared Medicine Profile has further extended the groups of health care personnel with access to the patient’s medication records. Access is now given to doctors, nurses, midwives, health visitors, social and health care assistants, workers in care for the elderly, dentists, certain pharmacists employed at hospitals as well as pharmacists and pharmacy assistants in community pharmacies, as well as staff at The National Board of Health and the Medicines Agency. Subject to the Ministry of Health issuing more specific rules and regulations the access to the Shared Medicine Profile may also in the future be extended to other persons directly involved in distributing medications to individuals in their homes.
The Shared Medicine Profile was introduced with the explicit purpose of sharing information about a patient’s medication between hitherto separated information systems and between health care personnel in different settings. And the aim is to achieve better coordination of medication within and across sectors, faster correction of medication errors,early identification and consideration of potential drug interactions, and a more effective use of health care resources through reducing multiple, identical entries in multiple parallel information systems . The Shared Medicine Profile will be fully implemented by the end of 2012.
Limits to the requirement of informed consent
In the previous section we introduced a heuristic for developing and evaluating different possible ways of implementing a requirement of informed consent in relation to the Shared Medicine Profile. Using the heuristic we ranked a number of possible models of informed consent. Assuming the value of personal autonomy, the previous section in effect leaves us with the question of what reasons there are for implementing any model but the one that provides the patient with the strongest protection of his or her personal autonomy, i.e. <1>. As already mentioned, there are possible reasons for implementing a weaker model of informed consent, and it is to the analysis of these we now turn.
There are insurmountable practical difficulties associated with obtaining consent
The patient lacks the competence required to provide consent,
There are other ethical concerns trumping the principle of respect for autonomy,
The practice of obtaining informed consent in itself undermines the protection of personal autonomy.
It is, for instance, a combination of these considerations that is usually taken to justify the common practice of sharing information without consent within the immediate group of health care personnel involved in the patient’s care. However, the information sharing in relation to the Shared Medicine Profile is much wider and requires separate analysis and justification.
The practical difficulties associated with the implementation of informed consent clearly hinges on the specific model of informed consent. The models providing the strongest protection of personal autonomy require that consent is obtained on many occasions, whereas the models providing the weakest protection only require a single valid consent. The practical difficulties will also depend on the character of the informed consent process. If it is handled as other consent processes in clinical practice then the health care professional must – on the basis of having informed about the Shared Medicine Profile – direct a request to the patient to provide consent to the registration and use of the patient’s personal medication information. This request could alternatively be put to the patient electronically. The Shared Medicine Profile provides the patient with web access to the entries in his or her profile. An electronic request of consent could be transmitted via the Shared Medicine Profile – and in case the consent is given the transfer of information to the profile and between information systems could be automatically initiated. Whichever way the requirement of informed consent is implemented it seems unreasonable to characterise the practical difficulties as insurmountable. Significant, but surmountable practical difficulties cannot as such rule out any implementation of informed consent, although they may be relevant to further ethical considerations of whether or not to prioritise the implementation of a given model of informed consent.
In the opinion of some members of the Danish Council of Ethics the registration and use of personal medication information must be made compulsory for the simple reason “” that many patients will find it difficult to grasp the advantages and disadvantages of not being included in the Shared Medicine Profile. In particular this would hold in relation to the group of patients (…) undergoing treatment with a variety of medicines (…).”  In the first general part of this statement the opinion seems to be that people in general do not satisfy standard criteria of decisional competence [27–30] in that they are considered unable to comprehend the consequences for them of being included in the Shared Medicine Profile. Unfortunately there is no further elaboration on what particular features of the Shared Medicine Profile that patients are supposedly, unable to grasp. There seems to be little difference between the nature of the intervention into a patient’s life constituted by the implementation of the Shared Medicine Profile and other medical interventions for which informed consent is required.
As we briefly discussed above, there are strong ethical reasons for implementing the Shared Medicine Profile. SMP improves conditions for providing medical care in situations of emergency, and more generally leads to better coordination of medication within and across sectors, faster correction of medication errors,earlier identification of potential interactions, and reduced waste of resources in the Health Sector. The question remains, however, if the satisfaction of the principles of beneficence and justice in the case of the Shared Medicine Profile are inconsistent with respecting and protecting personal autonomy through the implementation of a requirement of informed consent, i.e. if we are here faced with a true dilemma.
In order for the Shared Medicine Profile to present a true dilemma two mutually intertwined empirical conditions must be satisfied. Firstly, it has to be the case that no single patient can reap the benefits of the Shared Medicine Profile without registering all personal medication information. If there are individual health benefits and reduced waste of resources to be gained from only partial access to the patient’s personal medication information, then the implementation of a requirement of informed consent resulting in a limited registration of information or limited access to use the registered information, will not pose a hindrance to the improvement of health care. In this case then, the implementation of a requirement of informed consent will not be inconsistent with satisfying other principles such as beneficence and justice. The question therefore is, if there are benefits that may be achieved without full access to the patient’s information. It seems that partial access will – at least to some extent – allow for better coordination of medication within and across sectors, faster correction of medication errors, earlier identification of potential interactions, and also lead to reduced waste of the resources going into the parallel development and maintenance of separate information systems in the health care sector. Partial access to personal medication information may be a problem in terms of discovering potential drug interactions solely on the basis of studying medication records in the SMP, but some cases of interactions may be detected on the basis of access to less than complete personal medication information.
The second empirical prerequisite of a true dilemma concerns the patient’s actual behaviour in terms of providing consent. If the implementation of the Shared Medicine Profile is to pose a true dilemma, then we must be able to rule out the possibility that adequately informed patients voluntarily choose to consent to the registration and use of their personal medication information. However, there seems to be little evidence of widespread refusal of consent – at least in a Danish context. On the one hand, it does seem reasonable to expect that some consent processes may result in the withholding of consent. On the other hand, it does not seem as if patients generally exhibit a restrictive behaviour in other consent processes in the health sector. The extent to which consent will be refused is thus hard to predict. The point still applies, however, that if we cannot rule out the possibility that people voluntarily choose to provide consent, then the implementation of the Shared Medicine Profile with a requirement of obtaining informed consent does not necessarily pose a true dilemma.
Although the above considerations of the empirical preconditions of a true dilemma raise some important questions, they also point to a different sense in which the Shared Medicine Profile poses a dilemma. The implementation of the Shared Medicine Profile along with a requirement of informed consent is de facto incompatible with the maximal satisfaction of the targets of better coordination of medication, faster correction of medication errors,earlier identification of potential drug interactions, and reduced waste of health care resources. In so far as the principles of beneficence and of justice are taken to be requirements of maximal beneficence and justice, then the complete satisfaction of these principles is incompatible with the satisfaction of the requirement of informed consent. If the satisfaction of the principles of beneficence and justice may be gradual or partial, then it clearly is the case that the implementation of the Shared Medicine Profile with a requirement of informed consent is compatible with beneficence and justice.
One might here take the position that both of the empirical conditions are de facto satisfied: 1) the benefits of the implementation of the Shared Medicine Profile can only be realised on condition of full access to the personal medication information and patients would generally choose to withhold consent; and 2) that the principles of beneficence and justice are to be the sole principles guiding action. Consequently, the Shared Medicine Profile must be implemented with a requirement of compulsory registration and full access to the use of personal medication information. However attractive this position may seem, it conflicts with the strong tradition in the health care sector of giving priority to the protection of personal autonomy in situations that exhibit a significant similarity with this one. Thus it is common practice to accept a patient’s refusal of therapy that medically is considered to be appropriate, and instead provide therapy that is considered to be less adequate, but in accordance with the goals, plans and values of the patient – even in cases where this may lead to increased public costs. This leaves a proponent of the position sketched above with two options. Either the ethically relevant difference between clinical practice and the implementation of the Shared Medicine Profile has to be pointed out, or one must argue in favour of restricting the protection of personal autonomy through informed consent in all of these situations.
The fourth and final reason for restricting the use of informed consent in relation to the Shared Medicine Profile is if a requirement of informed consent undermines the very function of the informed consent as a protection of personal autonomy. How is that possible? The answer is that the informed consent may lose its function as a protection of personal autonomy if it is routinised. By routinisation is meant the provision of consent as a routine action, i.e. as an unreflected, habitual act. If the provision of consent becomes an act of routine then it will no longer protect the personal autonomy since the very essence of this protection is about reflecting on the consistency of another’s suggested intervention with one’s goals, plans and values. On the assumption that routinisation may undermine the function of informed consent as suggested here, it evidently becomes of interest to clarify if the possible implementations of informed consent discussed in the previous sections may lead to the routinisation of consent. The process of providing consent before installing computer software may be seen as an analogous case shedding some light on this question. It seems to be a common experience that the consent provided before installing software is routinised – the consent is provided as an unreflective, habitual act. The act of providing consent is simply reduced to a number of steps that have to be taken before access to certain attractive functionalities is acquired. In this case the routinisation seems to be conditioned by four factors. First, the unmanageability of the extensive information on the conditions of use. Second, the high frequency with which consent is to be provided. Third the strong desire to acquire the functionalities constituted by the software. Fourth, the expectation that the conditions of use are not unreasonable. The second factor is highly relevant when considering the possible implementation of informed consent discussed previously. If informed consent is required for every registration and use of personal medication information, then a significant number of people will be asked to provide consent rather often. If this model of consent is implemented such that the consent is to be provided to health personnel in the course of treatment, then the third and fourth factor are also relevant. The provision of consent may thus be influenced by a desire to start or focus upon treatment, and it seems likely that many people will expect the actions suggested by the health personnel – access to the patient’s personal medication information – to be in their best interest. The general conclusion therefore seems to be that at least some of the suggested models of informed consent may lead to the routinisation of the provision of informed consent, and hence to the undermining of the ability of the informed consent to protect personal autonomy. It has to be noted, however, that this argument needs further substantiation. Routinisation ultimately refers to an empirical condition, and therefore has to be vindicated as such. Note also, that although the argument from routinisation undermines extensive and repeated use of informed consent in the implementation of the Shared Medicine Profile, it is wholly consistent with the assertion of the principle of respect for personal autonomy. Thus one may hold that it exactly is because personal autonomy is worthy of protection, that routinisation poses a problem.
In this article we have argued that the implementation of a comprehensive health information system such as the comprehensive, nationwide Danish pharmaceutical information system known as the Shared Medicine Profile poses an ethical challenge, namely the challenge of balancing efforts to effectively gather and disseminate information in order to prevent harm to the individual and promote a just distribution of health care resources while at the same time respecting the patient’s autonomy.
Constructing possible models of informed consent, and
Evaluating these models with respect to their impact on personal autonomy
That there are no insurmountable practical difficulties involved in implementing a requirement of informed consent
That it seems unlikely that patients should lack the competence required to provide consent in relation to the registering and use of personal medicine information
That the implementation of pharmaceutical information systems do not necessarily poses a ‘strong’ ethical dilemma, and
That the implementation of a model of informed consent requiring the patient to provide frequent consent may lead to the routinisation of informed consent
With respect to any future research agenda, the fourth conclusion is of special importance. As argued above the routinisation of informed consent undermines the very ability of informed consent to protect personal autonomy, and it seems as if routinisation poses a serious threat to a significant sub-set of the models of informed consent considered in this article. As far as we know the notion has not received any attention in the literature on medical ethics as well as within the broader framework of ethics and moral philosophy.
- Cross M: England sets up task force to hasten use of electronic records. BMJ. 2006, 332 (7556): 1467-View ArticleGoogle Scholar
- Watson N: Patients should have to opt out of national electronic care records - For. BMJ. 2006, 333 (7557): 39-42. 10.1136/bmj.333.7557.39.View ArticleGoogle Scholar
- Halamka JD: Patients should have to opt out of national electronic care records – Against. BMJ. 2006, 333 (7557): 39-42. 10.1136/bmj.333.7557.39.View ArticleGoogle Scholar
- Baird G: Confidentialiy: what everyone should know, or, rather, shouldn’t. Br J Gen Pract. 2008, 58 (47): 131-133. 10.3399/bjgp08X277131.View ArticleGoogle Scholar
- Halford S, Obstfelder A, Lotherington AT: Beyond implementation and resistance: hot the delivery of ICT policy is reshaping healthcare. Policy & Politics. 2009, 37 (1): 113-128.View ArticleGoogle Scholar
- Ministry of Health: Lov om ændring af lov om offentlig sygesikring (Law on changing the law on National Health Service). 2003, Copenhagen,https://www.retsinformation.dk/Forms/R0710.aspx?id=9727,Google Scholar
- Ministry of Health: Sundhedsloven (The Health Law). 2010, Copenhagen,https://www.retsinformation.dk/Forms/R0710.aspx?id=130455&exp=1,Google Scholar
- The Danish Council of Ethics: Det fælles medicinkort. 2010, Copenhagen: The Shared Medicine ProfileGoogle Scholar
- Beauchamp TL, Childress JF: Principles of Biomedical Ethics. 2001, Oxford: Oxford University PressGoogle Scholar
- Feinberg J: Autonomy. The Inner Citadel – Essays in Individual Autonomy. Edited by: Christman J. 1989, Oxford: Oxford University Press, 27-53.Google Scholar
- Ploug T: The Ethical Significance of Expectations and the Case of Microsoft Office Accounting. Studies in Ethics, Law, and Technology. 2010, 4 (2):Google Scholar
- Berofsky B: Liberation From Self: A Theory of Personal Autonomy. 1995, Cambridge: Cambridge University PressView ArticleGoogle Scholar
- Dworkin G: The Concept of Autonomy. The Inner Citadel – Essays in Individual Autonomy. Edited by: Christman J. 1989, Oxford: Oxford University Press, 54-62.Google Scholar
- Frankfurt H: Freedom of the Will and the Concept of a Person. The Inner Citadel – Essays in Individual Autonomy. Edited by: Christman J. 1989, Oxford: Oxford University Press, 63-76.Google Scholar
- Kant I: Groundwork of the Metaphysics of Morals. 1997, Cambridge University Press, CambridgeGoogle Scholar
- Hill TE: The Kantian Conception of Autonomy. The Inner Citadel – Essays in Individual Autonomy. Edited by: Christman J. 1989, Oxford: Oxford University Press, 91-108.Google Scholar
- Ong LML, De Haes JCJM, Hoos AM, Lammes FB: Doctor-Patient Communication: A Review of the Literature. Soc Sci Med. 1995, 40 (7): 903-918. 10.1016/0277-9536(94)00155-M.View ArticleGoogle Scholar
- Kreuter MW, Bull FC, Clark EM, Oswald DL: Understanding how people process health information: A comparison of tailored and non-tailored weight-loss material. Health Psychol. 1999, 18 (5): 487-494.View ArticleGoogle Scholar
- Koo MM, Krass I, Aslani P: Factors Influencing Consumer Use of Written Drug Information. Ann Pharmacother. 2003, 37: 259-267.View ArticleGoogle Scholar
- Zarcadoolas C, Pleasant A, Greer DS: Understanding health literacy: an expanded model. Health Promot Int. 2005, 20 (2): 195-203. 10.1093/heapro/dah609.View ArticleGoogle Scholar
- Holm S, Madsen SM: Informed consent in medical research – A procedure stretched beyond breaking point. The Limits of Consent. Edited by: Corrigan O, McMillan J, Liddell K, Richards M, Weijer C. 2009, Oxford: Oxford University Press, 11-24.View ArticleGoogle Scholar
- Appelbaum PS, Roth LH, Lidz CW, Benson P, Winslade W: False hopes and best data: consent to research and the therapeutic misconception. Hastings Cent Rep. 1987, 17: 20-24.View ArticleGoogle Scholar
- Elbourne D, Snowdon C, Garcia J: Informed consent Subjects may not understand concept of clinical trials. Br Med J. 1997, 315: 248-249.View ArticleGoogle Scholar
- Faden RR, Beauchamp T: A history and Theory of Informed Consent. 1986, Oxford: Oxford University PressGoogle Scholar
- Featherstone K, Donovan JL: Random allocation or allocation at random? Patients' perspectives of participation in a randomised controlled trial. Br Med J. 1998, 317: 1177-1180. 10.1136/bmj.317.7167.1177.View ArticleGoogle Scholar
- Featherstone K, Donovan JL: Why don't they just tell me straight, why allocate it?" The struggle to make sense of participating in a randomised controlled trial. Soc Sci Med. 2002, 55: 709-719. 10.1016/S0277-9536(01)00197-6.View ArticleGoogle Scholar
- Buchanan E, Brock DW: Deciding for Others – The Ethics of Surrogate Decision Making. 1990, Cambridge: Cambridge University PressView ArticleGoogle Scholar
- Brock DW: Decisionmaking Competence and Risk. Bioethics. 1991, 5: 105-112. 10.1111/j.1467-8519.1991.tb00151.x.View ArticleGoogle Scholar
- Buchanan A: Mental capacity, legal competence and consent to treatment. J R Soc Med. 2004, 97: 415-420. 10.1258/jrsm.97.9.415.View ArticleGoogle Scholar
- Freedman B: Competence, Marginal and Otherwise – Concepts and Ethics. Int J Law Psychiatry. 1981, 4: 53-72. 10.1016/0160-2527(81)90020-0.View ArticleGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1472-6939/13/30/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.