The imaging systems of COMPUTED TOMOGRAPHY (CT) and magnetic resonance imaging (see MRI) have powerful computer techniques underlying them.
Computerised statistical analysis of study data, population databases and disease registries is now routine, leading to enhanced understanding of the interplay between diseases and the population. And the results of research, available on computerised indexes such as MEDLINE, can be obtained in searches that take only seconds, compared with the hours or days necessary to accomplish the same task with its paper incarnation, Index Medicus.
Medical informatics The direct computerisation of those activities which are uniquely medical – history-taking, examination, diagnosis and treatment – has proved an elusive goal, although one hotly pursued by doctors, engineers and scientists working in the discipline of medical informatics. Computer techniques have scored some successes: patients are, for example, more willing to be honest about taboo areas, such as their drug or alcohol consumption, or their sexual proclivities, with a computer than face to face with a clinician; however, the practice of taking a history remains the cornerstone of clinical practice. The examination of the patient is unlikely to be supplanted by technological means in the foreseeable future; visual and tactile recognition systems are still in their infancy. Skilled interpretation of the result by machine rather than the human mind seems equally as remote. Working its way slowly outwards from its starting point in mathematical logic, ARTIFICIAL INTELLIGENCE that in any way mimics its natural counterpart seems a distant prospect. Although there have been successes in computer-supported diagnosis in some specialised areas, such as the diagnosis of abdominal pain, workable systems that could supplant the mind of the generalist are still the dream of the many developers pursuing this goal, rather than a reality available to doctors in their consulting rooms now.
In therapeutics, computerised prescribing systems still require the doctor to make the decision about treatment, but facilitate the process of writing, issuing, and recording the prescription. In so doing, the system can provide automated checks, warning if necessary about allergies, potential drug interactions, or dosing errors. The built-in safety that this process o?ers is enhanced by the superior legibility of the script that ensues, reducing the potential for error when the medicine is dispensed by the nurse or the pharmacist.
Success in these individual applications continues to drive development, although the process has its critics, who are not slow to point to the lengthier consultations that arise when a computer is present in the consulting room and its distracting e?ect on communication with the patient.
Underlying these many software applications lies the ubiquitous personal computer – more powerful today than its mainframe predecessor of only 20 years ago – combined with networking technology that enables interconnection and the sharing of data. As in essence the doctor’s role involves the acquisition, manipulation and application of information – from the individual patient, and from the body of medical knowledge – great excitement surrounds the development of open systems that allow di?erent software and hardware platforms to interact. Many problems remain to be solved, not least the fact that for such systems to work, the whole organisation, and not just a few specialised individuals, must become computer literate. Such systems must be easy to learn to use, which requires an intuitive interface between user(s) and system(s) that is predictable and logical in its ordering and presentation of information.
Many other issues stand in the way of the development towards computerisation: standard systems of nomenclature for medical concepts have proved surprisingly di?cult to develop, but are crucial for successful information-sharing between users. Sharing information between existing legacy systems is a major challenge, often requiring customised software and extensive human intervention to enable the previous investments that an organisation has made in individual systems (e.g. laboratory-result reporting) to be integrated with newer technology. The beginnings of a global solution to this substantial obstacle to networking progress is in sight: the technology that enables the Internet – an international network of telephonically linked personal computers – also enables the establishment of intranets, in which individual servers (computers dedicated to serving information to other computers) act as repositories of ‘published’ data, which other users on the network may ‘browse’ as necessary in a client-server environment.
Systems that support this process are still in early stages of development, but the key conceptualisations are in place. Developments over the next 5–10 years will centre on the electronic patient record available to the clinician on an integrated clinical workstation. The clinical workstation – in essence a personal computer networked to the hospital or practice system – will enable the clinician to record clinical data and diagnoses, automate the ordering of investigations and the collection of the results, and facilitate referral and communication between the many professionals and departments involved in any individual patient’s care.
Once data is digitised – and that includes text, statistical tables, graphs, illustrations and radiological images, etc. – it may be as freely networked globally as locally. Consultations in which live video and sound transmissions are the bonds of the doctor-patient relationship (the techniques of telemedicine) are already reality, and have proved particularly convenient and cost-e?ective in linking the patient and the generalist to specialists in remote areas with low population density.
As with written personal medical records, con?dentiality of personal medical information on computers is essential. Computerised data are covered by the Data Protection Act 1984. This stipulates that data must:
be obtained and processed fairly and lawfully.
be held only for speci?ed lawful purposes.
•not be used in a manner incompatible with those purposes.
•only be recorded where necessary for these purposes.
be accurate and up to date.
not be stored longer than necessary.
be made available to the patient on request.
be protected by appropriate security and backup procedures. As these problems are solved, concerns about
privacy and con?dentiality arise. While paper records were often only con?dential by default, the potential for breaches of security in computerised networks is much graver. External breaches of the system by hackers are one serious concern, but internal breaches by authorised users making unauthorised use of the data are a much greater risk in practice. Governing network security so that clinical users have access on a need-to-know basis is a di?cult business: the software tools to enable this – encryption, and anonymisation (ensuring that clinical information about patients is anonymous to prevent con?dential information about them leaking out) of data collected for management and research processes – exist in the technical domain but remain a complex conundrum for solution in the real world.
The mushroom growth of websites covering myriad subjects has, of course, included health information. This ranges from clinical details on individual diseases to facts about medical organisations and institutes, patient support groups, etc. Some of this information contains comments and advice from orthodox and unorthodox practitioners. This open access to health information has been of great bene?t to patients and health professionals. But web browsers should be aware that not all the medical information, including suggested treatments, has been subject to PEER REVIEW, as is the case with most medical articles in recognised medical journals.... information technology in medicine
Health Encyclopedia