THE LINKS, NOT VOLTS OR THE BYTES, ARE WHAT MATTER*

Among the career achievements of the British physician Sir Donald Acheson (1926-2010) were his contributions as chief medical officer of the UK (1983-1991), to clinical epidemiology and medical education reform when as professor and dean at Southampton University (1963-1978), assessing and redressing the effects of social inequalities on health and especially his pioneering work at Oxford (1957-1968) on the linkage of medical records with other health data. He was a gracious humanitarian and his contributions were made with modesty and effectiveness.

Record linkage, pursued with great energy and creativity in Western Australia by Michael Hobbs (an Acheson protégé) and his colleagues over several decades especially in their monitoring trends in cardiovascular disease in Australia, provides us with an understanding of what happens to patients as they traverse episodes of illness and care, whether in hospital or in the community.  This enables patient-centred care, so extolled by quality gurus such as American Don Berwick.  We know so much more from a longitudinal record of a patient than we can gather by cutting and pasting the records of unconnected episodes of care. 

So whereas we often have separate medical records for inpatients and outpatients in our hospitals, and indeed separate systems for recording their data from pathology, imaging, the ED and the ward, a linked record enables us to see what has been done with and for a patient in all these settings and assess the value of the interventions.  Recently, in an attempt to assess what happens to patients with severe chronic problems managed using coordinated care in a NSW hospital, we had to figure out their use of services both within and out of hospital from eight unrelated electronic data files. Electronics are not enough.

The achievement of the linked medical record should be understood as the fundamental reason for the current interest and investment in electronic health records.  We have immense data processing and transmitting advantages over those that Acheson, Hobbs and others had available to them when they started their linkage projects. ‘Electronic’ the record will be, but that simply describes one of its qualities, not what it is for.

By using the word ‘linked’ we put right up front the purpose for the construction of these records.  We want the advanced orders, discussed sensitively and carefully in the general practice, to be there when the ambulance responds to an emergency call and the patient is seen in ED by a team to whom he or she is not known. 

We want linkage so that the general practitioner can see, even without asking or having to dither around changing portals, logins and passwords to access another web site, what went on with a patient during his or her recent sojourn in hospital.

The literature about electronic records is immense and there is recognition in parts of it of the value of having the patient involved in the construction, quality surveillance and oversight of the information it contains.  While privacy issues tend often to be overstated, a patient surely should be able to determine whether elements of their health journey are not to be made freely available through this medium.

A managerial appreciation of the value of a linked record derives from what information collated from hundreds or thousands of patients treated for a condition about their fate and experience especially when treated with different modalities for different conditions achieves and at what cost. Outcomes can be factored into the assessment, something that cannot be done at present with unlinked hospital data as are used to determine activity-based funding.  

The recent upsurge of interest in electronic records, so essential if we are rationally to order our investment in health services, has come to be called ‘effectiveness research’ now espoused in several places in Australia. The Scots, who by national persuasion are interested in wise use of money, define it thus: ‘Clinical effectiveness is the extent to which specific clinical interventions do what they are intended to do, i.e., maintain and improve the health of patients securing the greatest possible health gain from the available resources.’

A completely linked data system is a wonder to behold. The agencies for the care of veterans in Australia and (especially) the US have such systems and use them to advantage.  Managed care organisations such as Kaiser Permanente have ‘cradle to grave’ linked record systems.  In such an environment preventive programs can be targeted to those most likely to benefit at times such as pregnancy, post-operatively or upon diagnosis of diabetes when they are searching for preventive help.  Instant understanding is available for clinical managers about length-of-stay, complication rates, medication errors, diagnostic results, and anything you as a clinician or manager of clinical service may wish to know about both the well-being of your patient, the performance of yourself and your fellow clinicians and the performance of the system. 

So why are we stuck with paper?  The problem that seems to defy us is making the change from charts and an independent electronic data system to a linked system is management.  These changes can only be effected by attending at the micro level to processes of getting the system running in the clinical setting and then helping the clinician to adapt.  Medicine, for excellent reasons, is conservative: clinicians will change if convinced of the value of the change and if supported in making the change. Hectoring, incentivising with little dobs of money, or otherwise ignoring the magnitude of the reordering needed to move to a linked record have no evidence in their favour.

Kaiser, for example, in its multibillion dollar switch to electronic systems, spent HALF that investment on change management. I am deeply puzzled when I see how little we invest in managing the necessary changes and how little we understand how critical change management is in achieving a fully functional linked, patient-controlled electronic record. 

The history of IT in health care led Donald Acheson to say about 30 years ago, “The electronic medical record is today exactly where it was 20 years ago: just around the corner.”  I do not think that things are quite that desperate today.  It is not the electronics that are holding us up, not the volts or the bytes, but change management.


*Published in Aus Doc