Sharing electronic records: separating out the signals from ethical white noise

Why this matters to me

As a GP doing out-of-hours and A&E work, I have often found that patients seemed more frustrated that I did not have access to their healthcare details, than worried that I might abuse them. As a part-time ethics teacher, I recognise that electronic shared records present ethical issues such as confidentiality and patient safety, and merits new discussions.

 

Abstract

The development of shared electronic records has been galvanised by initiatives to improve the information technology infrastructure of healthcare and by fragmentation of healthcare provision. This report uses a public discussion by the RCGP Ethics Committee, Patient Participation Group and the Informatics Group to review key issues relating to shared electronic healthcare records. These records may be a social necessity, and ethical and legal safeguards may allay some public and professional anxiety about their appropriate use. Many of the specific ethical concerns, such as confidentiality of healthcare information, are not new but how these concerns arise in context may be. Function creep was a key concern in the discussion.

Introduction

My own introduction to the controversies around shared electronic healthcare records was presented at the 2009 RCGP Primary Care Conference in Glasgow. A discussion, entitled ‘Whose records?’, was led by representatives of RCGP Ethics Committee, Patient Participation Group (PPG) and Informatics Group. It attracted a number of interested parties, and accordingly the discussion reflected a number of values, agendas and ways of looking at ethics and policy, concepts and practicalities. This paper distills the key themes from that discussion.

The possibility of different types of shared electronic record has arisen from the technical ability to store and retrieve increasingly complex and increasing amounts of information. It has been galvanised by recent initiatives to modernise the NHS information technology infrastructure. It has also been galvanised by social necessity. At the meeting, it was acknowledged that outside the UK, when care is provided either privately or via charities and health insurance, some sort of accessible record is required to allow continuity of care. At the RCGP Annual Primary Care Conference in 2009, the New Labour Minister for Health, Mike O'Brien had publicly announced that general practice boundaries would be abolished, so it might well have been argued that electronic continuity of care was high on the agenda at the meeting.

Policy: concept vs. detail

Perhaps unsurprisingly, the RCGP Informatics Group was primarily concerned with what shared electronic healthcare records might look like and how best to implement them. The RCGP Ethics Committee and PPG were concerned with whether to implement them (or not) and what issues might be raised by their implementation.

The potential utility of a shared record was bolstered by examples of useful shared information, such as special patient notes (SPNs). However, it was pointed out that different healthcare professionals require different types of data and ways of retrieving it. A chest X-ray picture may be as useless to a psychiatrist as a child's drawing in a psychologist's report may be to a vascular surgeon.

Those at the meeting classified levels of patient data as follows:

• anonymous data for the purposes of public health monitoring, government statistics and healthcare research

• summaries of health data such as test results, lists of conditions and current and recent medication

• a complete record including consultation notes and correspondence, dating back to first use of healthcare services or birth.

Ensuring appropriate confidentiality was, unsurprisingly, a dominant theme of discussion. It is a perennial concern for doctors and especially general practitioners (GPs).^1,2 The meeting was preceded by a debate conducted in the British Journal of General Practice specifically about summary care records.^3–5 Function creep – the idea that a tool designed for one purpose might be ‘misused’ for another – was a related concern. Could a comprehensive record be as accessible to those who might deny benefits and entitlements as to those who might facilitate them? Might police or employers be able to gain access to information, and might this deter patients from seeking medical attention? While these issues exist regardless of any electronic healthcare record, the presence of the record can enable fast, remote access to databases.

Doctors' concerns

Doctors' narratives mainly took a position of patient advocacy. Shared electronic records should take into account the most vulnerable patient, who is least equipped to manage that record themselves, or who might be incapable of expressing preferences and giving information in a crisis. The examples were given of the very elderly, the illiterate and people with psychiatric illness. There were suggestions that it was also important that there should be robust and secure shared records for people with particularly private and sensitive details in their records. Examples of such sensitive information might include: terminations of pregnancy, illicit drug use, disorders of sexual preference and details pertaining to criminal convictions.

Perhaps surprisingly, censorship of healthcare records (as a good thing) was a theme in discussion. There was an argument that patients should not have the ability to censor their healthcare data. The rationale for this is that details might affect safe clinical decision-making. Participants had come across requests in practice for total privacy, that healthcare data should be seen by no one without explicit written or verbal consent. This was clearly problematic in the context of unforeseen and unscheduled care.

There was also the suggestion that clinicians might suppress their own entries in health records – the example was given of medico-legal advice to destroy paper records (upon a general practice surgery going paperless) in order to avoid the ramifications of inappropriate or misleading historical comments.

Patients' concerns

Patient representatives at the meeting shared narratives about themselves, their friends and relatives. These narratives coalesced into three themes. There were stories about inappropriate access to, for example, mental health records by medical or surgical specialists. Opinion was divided over whether there ought to be a way of differentiating levels of access or whether healthcare workers should police themselves better. There were also narratives about the errors in the system. Examples of this included incorrect coding of data, or data entered into the notes of a different patient with the same name. Whilst this was used as an argument for patient-editable records, it is equally an argument that clinicians should investigate discrepancies between patient story and patient record rather than automatically opting to follow one or the other. There was an altogether more positive theme about how good records allowed effective handover between carers and how copying the patient into correspondence could lead to a fuller discussion about healthcare decisions. This reflects how records were historically a correspondence that was exclusively between clinicians and which now the patient is increasingly (and rightly) admitted to the conversation.

Conclusions

This discussion was of interest because the three groups leading it were able to inform one another's perspective. However, it illustrated the problems of separating clear signals from ethical white noise. The meeting amplified a problem arising from any public consultation, namely that there is a wide spectrum of agreement and dissent – vocal minorities may potentially dominate a discussion. However, dissenters need to be informed by the enthusiasts, and enthusiasts must listen to the dissenters concerns.

The key lessons which I drew from the discussion were:

• the possible fragmentation of healthcare services and continuity of care in parallel with multiple route of access to healthcare necessitates an electronic continuity of record. The electronic GP record in the NHS and special patient notes are examples of this generally working well

• many of the ethical concerns regarding shared notes, such as confidentiality and sensitive data, are not new. If commissioners and clinicians have a grasp of the basic concepts then these can be applied in the context of new technologies. Furthermore, it may be important that some sensitive information is shared, either in the patient's, or in the public interest

• particular concerns need to be addressed, for example if electronic records will be screened by government agencies seeking to detect undocumented migrants. This may deter patients from using out-of-hours and emergency services. A more everyday concern is that clinicians will only access relevant data. This is simpler with basic records but may be more of an issue as, and when, more comprehensive records become accessible.