XML Europe 2001 logo21-25 May 2001
Internationales Congress Centrum (ICC)
Berlin, Germany

XML and Pathology

Steve Pynegar <spynegar@drimh-tr.trent.nhs.uk>
 PDF version    Latest version   

ABSTRACT

Doncaster Royal Infirmary has been involved in many IT initiatives and implementations. The XML Pathology Requesting project is the Trusts' first step towards incrementally building an XML based Electronic Patient Record (EPR).

The Pathology Directorate, in conjunction with the Computer Services Department at Doncaster Royal Infirmary and the suppliers of their pathology system software, have been one of the pioneers in England of delivering results electronically to their customers, in a form that met national standards (X400 protocols, Read codes,EDIFACT) [1].

This was very much a first generation step towards an electronic form of results reporting. All concerned are now working towards a new approach, which is more acceptable internationally and will bring the development more into line with current thinking in electronic messaging.

The local driver for change has been the need to meet the requirements of the national information strategy, 'Information for Health' [1], published by the NHS Executive in 1998. This set targets for achieving a step type approach to the implementation of an electronic patient record (EPR) and in particular the hospital focussed on achieving what is known as Level 3 by April 2002. Two key features of Level 3 are the ability to place electronic clinical orders and results reporting.

The first step has been to undertake procurement of an order communication system which will also have the ability to expand into a fully operation EPR development which will take the hospital to well above Level 3.

Whilst order communications can be provided for a whole range of services throughout the hospital, it was recognised that the pathology service probably receives the most requests of all the services that are provided. There are about 2000 individual request forms a day received by the department, which can contain requests for tests ranging from one up to about 10. The first stage of the implementation of an EPR was therefore very much focussed on delivering an order communications system that would work for pathology, whilst recognising that it would have to be applicable to other services.

It was recognised very early on in the procurement phase that a system had to be found that not only met a specification for the order communications for a pathology service but that it also had to have the capability of meeting other EPR requirements. Additionally the NHS published a further IT strategy document 'Building the Information Core - Implementing the NHS Plan' [2], which made reference to the adoption of e-GIF (Government Interoperability Framework) standards which include the requirements for all government/public organisations to use XML to communicate with each other. Further guidance to the strategy produced by the NHS Executive defined the implications of e-GIF which were alignment with internet standards, use of browsers as the key interfaces and the adoption of XML for data integration.

The implications of working within a XML environment were not at first apparent but it soon became clear on further investigation just what powerful features it brought with it. The structure it can impose on documents were not only useful in terms of neatness, but it created the possibility of data search facilities which were far quicker and more useful than current traditional methods.

These features are particularly useful when considering the handling of vast amounts of pathology data. The primary output from a pathology service is data either in numerical or text format. At Doncaster in excess of 3million individual data items, excluding demographic patient details, are produced each year. They can be used to facilitate diagnosis or to support and monitor treatment. This data is then stored in written format in patient case notes. At Doncaster some reports are produced in cumulative format to aid visual scanning of results. For immediate reflection this works reasonably well but when it comes to clinical audit or research, identifying and collecting data on a number of patients, the task can be very difficult and time consuming.

Documents structured in XML can be created very easily with pathology results transferred into key areas of a document. Suddenly differentiating between patients who have overdosed on salicylate and those who have salicylate as part of their treatment becomes relatively easy and will save hours on search times.

The structure that XML gives will also make it easier to define report screens, which can operate in ward and GP practice environments. This is particularly important with pathology data, which can come in more than one format. On the screen there may be the actual result, normal ranges, patient demographic information and other additional comments. Additionally links may be introduced to web sites, help files, the pathology handbook held within an intranet and email access back to pathology for further queries. More effective results reporting, particularly in terms of detail and timeliness will also reduce the overhead on the pathology service which comes from operating a telephone enquiry service.

Within the XML environment we can also deliver our initial goal which was an order communications system for pathology in the first instance. Again structured ordering screens can be produced which can use standard windows features such as tick boxes and pull down menus to aid user selection. I believe we have the capability of producing an interface that the user will want to use and additionally will provide him or her with additional information at the time of need. It is too late when the sample has been received or processed to then relay information that should have been given at the point of request. An order communications system working within a web browser, using XML and windows standard functions allows the interchange of data between the two parties to occur in a structured and timely manner.

The benefits of XML to pathology at Doncaster have yet to be fully realised. The potential is there. Within pathology these benefits will include accuracy of data, less transcription errors and timeliness of the delivery of the service. For the users of the service the benefits will be even greater. Instant access to a greater scope of information at the time of use, advice and guidance at the time of use and rapid search and audit facilities when needed. In a climate where clinical governance is the centrepiece of service delivery, XML provides a useful tool in collating and communicating health related data.

Bibliography

[1] NHS Executive (1998) Information for Health Wetherby: Department of Health Publications
[2] Department of Health (2001) Building the Information Core-Implementing the NHS Plan London: Department of Health.

Glossary

EPR

Electronic Patient Record

Biography

Steve Pynegar
Pathology Business Manager; Clinical Chemistry Department
Doncaster Royal Infirmary
Doncaster
United Kingdom
Email: spynegar@drimh-tr.trent.nhs.uk

Steve Pynegar - Worked as a Biomedical Scientist for 23 years at Doncaster Royal Infirmary, attaining all the necessary professional qualifications. In 1992 became the Pathology Business Manager and in 1995 managed the procurement and installation of a new pathology computer system. Keenly interested in the use of application software within a hospital environment and the development of user interfaces, which encourage the use of information technology. In 2000 successfully completed a Masters Degree dissertation on the implementation of the NHS IT strategy in a district general hospital and currently project manager for the introduction of an order communications system at Doncaster Royal Infirmary, using Graphnet who are XML software providers.