Factors predictive of subsequent injury in a longitudinal cohort study

Abstract

Objectives: The aims are to: (1) Determine the cumulative incidence of self-reported subsequent injury (of any anatomical site or injury type) occurring between 3 months and 12 months after a sentinel injury among participants in the Prospective Outcomes of Injury Study in New Zealand and (2) Examine the preinjury and injury-related predictors of subsequent injury. Methods: Prospective Outcomes of Injury Study participants (n=2282) were interviewed 3 months, 5 months and 12 months after a sentinel injury event. Data were collected about a range of preinjury and injury-related factors at the 3-month interview and about subsequent injury at the 5-month and 12-month interviews. Poisson regression modelling was used to determine the preinjury and injury-related predictors of subsequent injury. Results: Between the 3-month and 12 month interviews 28% of the participants reported at least one subsequent injury. Subsequent injury was 34% more likely among participants with a prior injury affecting them at the time of the sentinel injury compared with participants without a prior injury affecting them, and more likely among non-workers (31% more likely) and trade/manual workers (32% more likely) compared with professionals. Participants whose sentinel injury was due to assault were 43% more likely to report a subsequent injury compared with those whose sentinel injury was accidental. A subsequent injury was 23% less likely if the sentinel injury was a lower extremity fracture compared with other injuries, and 21% less likely if the sentinel injury event involved hospitalisation. Conclusions: Among general injury populations it may be possible to identify people at increased risk for subsequent injury.