International Data Standards

A draft International Classification for External Causes of Injury (ICECI) will be released for comment during the 4th World Conference on Injury Prevention and control in Amsterdam in May 1998.

The International Classification of Diseases (ICD) and its ancestors have served for a century as the most widely used basis for classifying diseases, injuries and their `external causes.' These classifications have provided an enormously important foundation for statistical comparability over time and between places.

For injury prevention, the most important part of the ICD is the Supplementary Classification of External Causes of Injury and Poisoning, often known as the "E-code" classification. Injury researchers and others with a need to count or monitor injuries and the ways in which they occur have long been concerned by its limitations.

Limitations of the E-code classification largely reflect its origin as a tool designed for mortality data, and the fact that its basic structure has not altered in more than a century. It is essentially a single list, which has grown enormously and in a rather piecemeal manner over the decades.

The past twenty years or so have seen growing interest in approaches to data collection for injury prevention that go beyond the limits of the External Causes classification. Recognition that much of the burden of injury is due to cases that do not result in death or admission to a hospital has prompted collection of data on ambulatory cases, most commonly in emergency departments. Injury investigators typically want to know more about cases than is revealed by External Cause codes. The system developments required to extend collection into new areas have provided opportunities to try new approaches, embodying contemporary concepts of injury prevention. Increasing availability of computers for data capture and analysis has relaxed some previous constraints on the types of information that it is practicable to handle.

Examples of such developments are the National Electronic Injury Surveillance System (NEISS) operated in the USA since the 1970s, a classification developed by the Nordic Medico-Statistical Committee (NOMESCO) and applied in the Scandinavian region, the European Home and Leisure Accident Surveillance System (EHLASS) in the Western European Region, the National Injury Surveillance and Prevention Project and National Data Standards for Injury Surveillance in Australia, and a Minimum Data Set developed in New Zealand.

The WHO Programme for Safety Promotion and Injury Control (SPIC) convened the WHO-Working Group on Injury Surveillance Methodology Development in 1991, to advance work in this area. Early work of the Group revealed great similarity between developments in several parts of the world. This led to an assessment that it was desirable and practicable to develop a harmonised international classification of external causes of injury, to embody the main characteristics of contemporary approaches. The classification should meet the requirements of injury control practitioners and fit within the family of WHO-classifications for diseases and health-related problems.

The first draft International Classification for External Causes of Injuries (ICECI) was circulated for comment in 1995 and, slightly revised, was published by NOMESCO in 1997 for the Nordic region. The draft International Classification was welcomed by the international community but was also noted to lack demonstrated compatibility with ICD and to suffer shortcomings in definitions, explanatory notes, mutual exclusiveness and consistency. Therefore it was decided that further development and consultation should be undertaken during 1997 and 1998. The Resolution on Violence Prevention adopted by the 1996 World Health Assembly also prompted extension of the work, as it puts high priority on developing an adequate nomenclature and classification of injuries due to violence.

The new draft ICECI is being released for comment in conjunction with the 4th World Conference on Injury Prevention and Control, at Amsterdam in May 1998. The document centres on a Data Dictionary, presently comprising 17 data items. These are grouped into a core set of seven items, and two supplementary modules, one on violence, and the other on transport.

The relationship between the ICECI and the external cause chapter of ICD-10 has been made explicit through coding maps. Certain features of the External Cause classification constrain this mapping, and its validity has yet to be tested in the field. Nevertheless, if adequate validity can be confirmed, the map will enable data collected according to ICECI to be reported as ICD0-10 External Causes codes.

The ICECI is intended to be a tool, applicable anywhere, which helps its users to plan, develop and operate injury surveillance systems. Accordingly, the draft contains a glossary of relevant terms, and introductory chapters on the theory and practice of injury surveillance.

Consultation concerning the draft and its further development and testing will commence at the Injury ICE meeting and at the 4th World Conference. Testing of the system will commence later in 1998. The extent and nature of field-testing will depend on the willingness of organisations to participate in and contribute to the process, as the project is being undertaken essentially on a voluntary basis.

The aim is to have undertaken sufficient testing and revision to enable completion of the first version of ICECI by late 1999. It will then be launched at the 5th World Conference in New Delhi in the year 2000, the Centenary of the first revision conference of what was to become the ICD.

Further information is available from
James Harrison at the RCIS,
Tel: 08 8374 0970;
e-mail: james.harrison@nisu.flinders.edu.au