Back to Annual Meeting

Back to Annual Meeting

APHA Scientific Session and Event Listing

Recorded presentation

Phillip R. Hunt, CIH, ScD, Occupatinal Health Surveillance Program, Massachusetts Department of Public Health, 250 Washington St, Boston, MA 02108, 617-624-5633, phill.hunt@state.ma.us, Holly Hackman, MD, MPH, Injury Prevention and Control Program, Massachusetts Department of Public Health, 250 Washington Street, 4th Floor, Boston, MA 02108, and Letitia Davis, ScD, Occupational Health Surveillance Program, Massachusetts Department of Public Health, 250 Washington Street, 6th floor, Boston, MA 02108.

Electronic emergency department (ED) data represents a potentially valuable source of information for surveillance of work-related injuries. However, the accuracy of attribution of work-relatedness and cause of injury codes (E codes) has not been established. We assessed the accuracy of this information in electronic ED data from a stratified sample of twelve hospitals that submitted records of all injuries (ICD-9-CM N800-N999) to the Massachusetts Department of Health. Probable work-relatedness (PWR) was indicated by workers' compensation (WC) as expected payer. Medical charts were obtained for a stratified random sample of 1002 PWR and 250 probable non-work-related (PNWR) cases. Validation E codes were assigned by an expert E coder. Compared to chart information, the indicators of work-relatedness had a sensitivity of 83.6%, a specificity of 98%, and a predictive value positive (PVP) of 85%. The low PVP is primarily a result of 12% of medical records with no or ambiguous information on work-relatedness, suggesting that charts are less than perfect for ascribing work-relatedness. E codes were present for 98% of validated cases. E codes were accurate to 4 digits for 65% of PWR cases. Using broad cause of injury categories (ICE Matrix), external causes were correct for 84% of PWR cases. Non-specific causes accounted for 44% of broad category misclassifications. Among specific causes, machinery injuries were most frequently misclassified, predominantly as cut/pierce or struck by/against. Work-relatedness and cause of injury were reliably identified in the ED data, however, significant gaps are evident, indicating a need for training of hospital coding staff.

Learning Objectives: At the conclusion of the session, the participant (learner) in this session will be able to

• Develop a plan for evaluating the accuracy of electronic data elements
• Describe the reliability of cause of injury codes and indicators of work-relatedness in electronic emergency department data.
• Discuss the value of Emergency Department data in Occupational Health Surveillance

Keywords: Occupational Surveillance, Injuries

Presenting author's disclosure statement:

Any relevant financial relationships? No

Recorded presentation