Email updates

Keep up to date with the latest news and content from BMC Health Services Research and BioMed Central.

Open Access Highly Accessed Research article

A classification of diabetic foot infections using ICD-9-CM codes: application to a large computerized medical database

Benjamin G Fincke12*, Donald R Miller12 and Robin Turpin34

Author Affiliations

1 Center for Health Quality Outcomes and Economic Research (CHQOER), Bedford VA Medical Center, 200 Springs Road, Bedford, MA 01730 USA

2 Boston University School of Public Health, Department of Health Policy and Management, 715 Albany Street, Boston, MA 02118 USA

3 Merck & Co., One Merck Drive, Whitehouse Station, NJ 08889-0100 USA

4 Department of Health Policy, Jefferson Medical College, 1015 Walnut Street, Philadelphia, PA 19107 USA

For all author emails, please log on.

BMC Health Services Research 2010, 10:192  doi:10.1186/1472-6963-10-192

Published: 6 July 2010

Abstract

Background

Diabetic foot infections are common, serious, and varied. Diagnostic and treatment strategies are correspondingly diverse. It is unclear how patients are managed in actual practice and how outcomes might be improved. Clarification will require study of large numbers of patients, such as are available in medical databases. We have developed and evaluated a system for identifying and classifying diabetic foot infections that can be used for this purpose.

Methods

We used the (VA) Diabetes Epidemiology Cohorts (DEpiC) database to conduct a retrospective observational study of patients with diabetic foot infections. DEpiC contains computerized VA and Medicare patient-level data for patients with diabetes since 1998. We determined which ICD-9-CM codes served to identify patients with different types of diabetic foot infections and ranked them in declining order of severity: Gangrene, Osteomyelitis, Ulcer, Foot cellulitis/abscess, Toe cellulitis/abscess, Paronychia. We evaluated our classification by examining its relationship to patient characteristics, diagnostic procedures, treatments given, and medical outcomes.

Results

There were 61,007 patients with foot infections, of which 42,063 were classifiable into one of our predefined groups. The different types of infection were related to expected patient characteristics, diagnostic procedures, treatments, and outcomes. Our severity ranking showed a monotonic relationship to hospital length of stay, amputation rate, transition to long-term care, and mortality.

Conclusions

We have developed a classification system for patients with diabetic foot infections that is expressly designed for use with large, computerized, ICD-9-CM coded administrative medical databases. It provides a framework that can be used to conduct observational studies of large numbers of patients in order to examine treatment variation and patient outcomes, including the effect of new management strategies, implementation of practice guidelines, and quality improvement initiatives.