Studying psychoactive substance use in injured patients: does exclusion of late arriving patients bias the results?

Abstract

Background: Most studies of the prevalence of psychoactive substances in injured emergency department patients have excluded those who arrive more than 6h after injury. This may cause a selection bias. The aim of this study was: (1) to describe the characteristics of patients who arrive more than 6h after injury, compared to patients who arrive sooner (2) to examine whether self-report can add to the assessment of alcohol use when the patient is assessed more than 6h after injury.

Methods: Blood sample analysis and self-report data were used to assess the prevalence of psychoactive substances in injured patients admitted to an emergency department within 48 h of injury (n=1611). Discriminant function analysis was used to assess group differences.

Results: The patients who arrived more than 6h after injury differed significantly from those who arrived earlier in several respects. They more often screened positive for hypnotics; they were older, they were more likely to have had a fall and they were more often injured at home and at night. Self reported use of alcohol showed good consistency with blood sample screening within 6h of injury and could therefore be used to assess alcohol use more than 6h after injury.

Conclusions: Patients who arrive more than 6h after injury differ significantly from those who arrive earlier. Future studies on the prevalence of psychoactive substances in emergency departments could expand the inclusion window.