Identifying priorities in methodological research using ICD-9-CM and ICD-10 administrative data: report from an international consortium
1 Department of Community Health Sciences, University of Calgary, Calgary, Canada
2 Centre for Health and Policy Studies, University of Calgary, Calgary, Canada
3 Information Services, Healthcare Information Group, Edinburgh, UK
4 British Columbia Cardiac Registry, Vancouver, Canada
5 Institut Universitaire de Médecine Sociale et Préventive, University of Lausanne, Switzerland
6 Centre for Health Evaluation and Outcome Sciences, University of British Columbia, Vancouver, Canada
7 Health Division, Statistics Canada, Ottawa, Canada
8 Manitoba Centre for Health Policy, Department of Community Health Sciences, University of Manitoba, Winnipeg, Canada
9 Public Health College, Second Shanghai Medical University, Shanghai, China
10 University of California Davis School of Medicine, Davis, USA
11 Department of Human Services, Melbourne, Australia
12 Institute for Clinical Evaluative Sciences, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, Canada
13 Canadian Institute for Health Information, Toronto, Canada
14 Department of Medicine, University of Calgary, Calgary, Canada
BMC Health Services Research 2006, 6:77 doi:10.1186/1472-6963-6-77Published: 15 June 2006
Health administrative data are frequently used for health services and population health research. Comparative research using these data has been facilitated by the use of a standard system for coding diagnoses, the International Classification of Diseases (ICD). Research using the data must deal with data quality and validity limitations which arise because the data are not created for research purposes. This paper presents a list of high-priority methodological areas for researchers using health administrative data.
A group of researchers and users of health administrative data from Canada, the United States, Switzerland, Australia, China and the United Kingdom came together in June 2005 in Banff, Canada to discuss and identify high-priority methodological research areas. The generation of ideas for research focussed not only on matters relating to the use of administrative data in health services and population health research, but also on the challenges created in transitioning from ICD-9 to ICD-10. After the brain-storming session, voting took place to rank-order the suggested projects. Participants were asked to rate the importance of each project from 1 (low priority) to 10 (high priority). Average ranks were computed to prioritise the projects.
Thirteen potential areas of research were identified, some of which represented preparatory work rather than research per se. The three most highly ranked priorities were the documentation of data fields in each country's hospital administrative data (average score 8.4), the translation of patient safety indicators from ICD-9 to ICD-10 (average score 8.0), and the development and validation of algorithms to verify the logic and internal consistency of coding in hospital abstract data (average score 7.0).
The group discussions resulted in a list of expert views on critical international priorities for future methodological research relating to health administrative data. The consortium's members welcome contacts from investigators involved in research using health administrative data, especially in cross-jurisdictional collaborative studies or in studies that illustrate the application of ICD-10.