tional sources of information that could be consulted include emergency medical services data, school health services records, community and private health care providers data, etc. However, there are serious technical and practical limitations to integrating these sources.
External Cause of Injury codes (E-codes) were developed by the World Health Organization (WHO) as a supplemental code for use with the International Classification of Diseases (ICD). These codes provide a systematic way to classify diagnostic information that health care providers have entered into the medical record. They are standardized internationally and thus permit comparisons of data among communities, states, and countries (Educational Development Center, 1999). Since 1999 mortality data in the United States has been coded using the 10th Revision of the ICD (ICD-10), while morbidity data is coded using a clinical modification of the 9th Revision of the ICD (ICD-9CM) (Annest et al., 1998). ICD-10CM is currently being developed, and is expected to improve the specificity and accuracy for descriptions of non-fatal injuries (Annest et al., 1998). Currently, 26 states either mandate or have rates over 90 percent for use of E-codes in their Hospital Discharge Data Systems (ICRIN, 2001), and 11 states require their use in Emergency Department Data Systems (Annest et al., 1998).
The usefulness of E-codes for a surveillance system rests on the consistency of their use, and technical concerns regarding the compatibility of the format and type of different systems (CDC, 1995). For example, the number of permitted fields on reporting forms would need to be standardized since more than one field allows much more detailed and informative coding. The CDC is currently pilot-testing its National Electronic Disease Surveillance System (NEDSS), an initiative that will standardize public health data systems for infectious disease to allow integrated and electronically compatible national, state, and local surveillance systems. NEDSS also will support surveillance of other public health concerns including causes of injury (CDC, 2001b). In the future, use of a NEDSS compatible system will be a requirement for CDC surveillance funding of infectious diseases (CDC, 2001a). An ongoing international effort seeks to develop a new multi-axial classification system for external causes of injury which is intended to be used in both mortality and morbidity databases (Annest and Pogostin, 2000). Such a system needs to be compatible with existing data coding systems in order to maintain consistency of monitoring and to increase the feasibility of large-scale implementation (IOM, 1999).