Citation

Sorock GS, Lombardi DA, Hauser RB, Eisen EA, Herrick RF, Mittleman MA. Am. J. Ind. Med. 2001; 39(2): 171-179.

Affiliation

Liberty Mutual Research Center for Safety and Health, Hopkinton, Massachusetts, USA. gary.sorock@libertymutual.com

Copyright

(Copyright © 2001, John Wiley and Sons)

DOI

unavailable

PMID

11170159

Abstract

BACKGROUND: Acute hand injury is the leading cause of occupational injury treated in United States' hospital emergency departments (e.g., laceration, crush or fracture). To identify risk factors for traumatic hand injuries, we conducted a case-crossover study of transient exposures (e.g., being rushed) for acute occupational traumatic hand injury. METHODS: The case-crossover method, which uses subjects as their own controls, was used to identify risk factors for occupational hand injury. Two hundred and thirty-two subjects were recruited from 17 occupational health clinics in New England and interviewed by telephone a median of 1.2 days after their injury. The a priori hazard period was defined as 10 min before the injury. Two control periods were used: one was 60-70 min prior to the injury (matched-pair interval analysis); the other was the total work-time exposed, on average, in the previous month (usual frequency analysis). RESULTS: In the usual frequency analysis, the relative risk for using malfunctioning or different-from-usual equipment or tools in the hazard period was 25.5 (95% confidence interval = 18.4-35.2). Relative risks were also significantly elevated for performing a task using an unusual work method, doing an unusual task, being distracted, or being rushed. Wearing gloves appeared to be protective (relative risk = 0.8, 95% CI = 0.5-1.2). Matched-pair interval analysis, where appropriate, provided similar findings but had much wider confidence intervals. CONCLUSIONS: This study demonstrates that the case-crossover design is a feasible and efficient method for studying transient risk factors for sudden-onset traumatic occupational hand injury. The usual frequency analysis proved more useful than the match-pair approach to control period selection.

Language: en