Digital dystopias: will the electronic health record ever fulfil its potential?

Abstract

‘Where is the wisdom we have lost in knowledge? Where is the knowledge we have lost in information?’ T.S. Eliot, 1934 The Danish Anaesthesia Database was established in 2004 with the aim of collecting data on every patient undergoing anaesthesia in Denmark 1. Analyses serve a number of purposes including quality assurance, quality improvement, audit and research. For hospitals in Denmark, reporting is mandatory, with national coverage around 70% and increasing. The completeness and quality of data are also increasing due to efforts to improve interoperability between local and national IT systems. Just like several well-established UK anaesthesia databases, such as the National Emergency Laparotomy Audit (https://www.nela.org.uk) and the Peri-operative Quality Improvement Programme (https://pqip.org.uk), local clinicians are responsible for inputting data and markers of quality can be recorded, continually assessed and compared between institutions and over time. In a recent issue of Anaesthesia, Thomsen et al. illustrated the value of this database with their analysis of > 650,000 general anaesthetics from 2008 to 2012. They found that supraglottic airway devices were seldom used for difficult airway management (12.4%) despite featuring in difficult airway guidelines and with a recorded success rate of 62.8% for ‘cannot intubate, cannot oxygenate’ situations 2. In their editorial, Ahmad and El-Boghdadly discuss the generalisability and clinical applicability of this finding in the context of the challenges of difficult airway research, and propose a roadmap for the development of a difficult airway research strategy 3. Although this process begins with systematically analysing published research, we argue emergent themes from routinely collected clinical data can likewise establish trends and generate hypotheses. The Australian government are currently implementing ‘My Health Record’, a national electronic health record that already has a 90.1% national participation rate, over 15,000 connected health professionals and 11.5 million clinical document uploads (https://www.myhealthrecord.gov.au). Such systems have great potential to improve patient care, capture clinical data and accelerate continuous gradual improvements in healthcare. This editorial aims to set out what clinicians might want from an electronic health record and at the same time explore the possibility that such systems might one day replace national anaesthesia databases. Pursuit of complete digitisation of patient health records is by no means a new aim for modern healthcare organisations. Few would argue against several laudable objectives such as the provision of accurate, up-to-date and accessible clinical data at the point of care; the ability to efficiently share information with other professionals; and reduced financial costs associated with paperwork, duplication and storage. Such systems promise to reduce inefficiencies; reduce the time required for diagnosis and treatment; increase the accuracy of coding 4; enhance patient safety; allow for safer prescribing; and, as any attempt to retrieve and input data are traceable to individuals within an organisation, make for safer and more private patient data (Table 1). That said, there are many questions to address, including: the type of information that should be available; the specifics of what anaesthetists should record and how they should record it; how such data might be used; the reasons for resistance from healthcare workers and patients; and why we have not been able to achieve it, yet. 1. Easy, quick electronic access to accurate and up-to-date clinical data for patients and healthcare professionals. 2. Clinicians can provide patients with electronic information relevant to their treatment. 3. Patients and healthcare providers accessing information can be registered and verified. 4. Patients can have control over their own data. 5. Better informed clinicians with the aim of providing better patient care. 1. One integrated nationwide system presents enormous logistical issues for organisations. 2. There would be a significant financial impact on countries wishing to design and implement such systems. 3. The temporary disruption of workflow for healthcare professionals. 4. Challenging transition periods. 5. The implementation of any change at a national level is always difficult. 6. Assuring interoperability of systems requires regular involvement and communication with the software industry. 7. Keeping the system up-to-date and secure presents many technical and financial challenges. Anaesthetists require basic clinical information, as does any doctor caring for a patient. Basic clinical information may include: patient characteristics such as height, weight and BMI; communication preference; next of kin; current General Practitioner; ASA physical status; past investigations; and details of medical history, including the year of diagnosis and severity grading. Surgical history may include a list of operations in the past, the most recent intervention and notes on any previous complications during anaesthesia. Other items may include: hereditary medical problems; smoking; alcohol and illicit drug use; exercise tolerance graded in the form of metabolic equivalents; grading of any gastro-oesophageal reflux and a list of medications and allergies. Previous airway assessments and alerts may also be easily accessible. This can all be provided remotely, regularly updated and accessed well in advance of the surgery by all those requiring it. There is currently no consensus for what these preferred datasets should look like, and anaesthetists and their representative bodies should ideally be involved with their design and approval. Replacing ‘pre-operative assessment’ with what we now know to be ‘pre-operative optimisation’ promises to improve surgical outcomes, offer better value for money and improve public health 5. A recent proposal for a new peri-operative pathway suggests pre-operative bundles, including: risk stratification; surgery school; high-risk clinics; optimisation of comorbidities, such as anaemia, cardiovascular, respiratory and renal conditions; and comprehensive geriatric assessment, which should all be completed before final preparations in the pre-operative clinic 6. Such a pathway requires coordination between hospitals and primary care which, in the UK at least, is poor due to, among other factors, inefficiencies and the limited use of electronic communication 7. Although it seems logical to assume electronic health records may reduce these inefficiencies and improve the quality of communication between hospitals and primary care, there is currently no evidence to support this. Moreover, the lack of an electronic health record might not even be part of the problem. The launch of ‘My Health Record’ in Australia, therefore, presents an opportunity to study its effect on the efficiency and efficacy of electronic communications between primary care and hospitals, pre-operative optimisation processes and postoperative outcomes. A common criticism of electronic health records is a perceived lack of data privacy. Some may fear the consequences of data falling into the hands of outside agencies such as, for example, employers or insurance companies. A recent example from the UK demonstrates the cost of such breaches for all parties (https://www.bbc.co.uk/news/uk-england-manchester-46419646). In Australia, stronger laws now protect the rights of patients, with punishments including fines and custodial sentences (https://www.myhealthrecord.gov.au/about/legislation-and-governance/summary-privacy-protections). Access codes can be provided to individual medical practitioners and patients are informed if their record has been accessed (when, by whom, where, etc.). More generally, any misuse or abuse of electronic records is taken seriously by employers and regulators in all hospitals with electronic records. That said, the inevitable conflict is that on the one hand, clinicians want fast and reliable access to accurate, usable data but on the other, we expect our own data and those of our patients to be as private as possible. This has not yet been fully resolved and it is difficult to envisage how it might be. Transferring from paper records to an electronic system is without doubt with the best of intentions. Regardless, it is always wise to consider the possible unintended negative consequences. Obvious problems may include slow and ‘clunky’ interfaces, which consume time and take healthcare staff away from direct patient contact. Some may feel such systems make their working life difficult and may even see electronic systems as a direct threat to care quality, productivity and patient safety 8. Other, ‘non-informational’ consequences might be somewhat less visible. For instance, an online theatre booking system for emergency operations, which is available from anywhere in the hospital, may appear to be helpful. This might replace a paper-based system, where booking forms are placed in slots in a board on the theatre corridor. This physical board acts as a gathering-place for discussion between staff about priorities, timings and personnel, and this face-to-face coordination is lost if it becomes a placeless, depersonalised system. There might be many other similar unintended consequences that, over time, lead to clinical staff resenting such systems. How electronic systems change our working practices beyond what is immediately visible must, therefore, be carefully contemplated by all. No nation has implemented an integrated nationwide electronic patient health record system, yet. In the UK in 2002, the National Programme for IT (NPfIT) project attempted to do just that with a budget of £6.2 billion 9. What was and probably remains the world's largest ever civil IT project, that promised to solve all the problems around interoperability, was dismantled in 2011, and there have been many harsh criticisms from various stakeholders 10. Last year, the UK Secretary of State launched a new vision for better digitisation in the NHS and appointed Ben Goldacre as chair of the new ‘Health tech’ advisory board (https://www.gponline.com/dr-ben-goldacre-lead-new-tech-advisory-board-nhs/article/1494314). Things seem to be changing in the UK, but few envisage another national IT project for the NHS anytime soon. The UK is not alone in these failings and there have been significant setbacks for those implementing interconnected IT projects in the Netherlands 11, France 12 and elsewhere across Europe and the rest of the world 13. Analysis of routinely collected clinical data may allow us to generate hypotheses and make sense of trends in clinical practice 3, 14, 15. The quality and completeness of such data is usually poor, at least in the UK, and probably more so in hospitals without electronic records. Nations with advanced anaesthesia databases have shown us time and again how useful such enquires can be 2, 16, 17. A nationwide interoperable and interconnected patient health record system would, perhaps, ensure usable data are collected in sufficient detail to learn from the past and improve the way services are delivered. That said, there is no consensus on how these data might be accessed by those outside the clinical team, more so when patients have not provided consent and institutions advise against formal submission of ‘non-research’ projects to research ethics committees. The blurred boundaries between what can be described as research, quality improvement and service evaluation persist 18. Patients trust us with their data when they are at their most vulnerable and we all have a duty to protect it and use it ethically. How this can be achieved in the era of ‘big-data’ is another conflict that has not yet been resolved 19. For all doctors including anaesthetists, relevant, accurate, accessible, shareable and well-kept information enhances our ability to make clinical decisions, which may improve patient outcomes. An ideal electronic health record might better facilitate this, and it may also act as a continually running combination of every available prospective database currently used to answer specific questions about patient care, such as NELA, PQIP and many others. Despite this enormous potential, there are unresolved issues around the impact such systems might have on current working practices, the risks and benefits of implementation, data privacy, and the ethics of utilising recorded data for purposes other than direct clinical care. The eyes of the world are now on Australia as they prepare to launch ‘My Health Record’ with the aim of succeeding where everyone else has failed. Will the electronic health record ever fulfil its potential? We might be about to find out. MC is Social Media Editor of Anaesthesia. No external funding or competing interests declared.