Intravenous Access During Out-of-Hospital Emergency Care of Noninjured Patients: A Population-Based Outcome Study

Study objectiveAdvanced, out-of-hospital procedures such as intravenous access are commonly performed by emergency medical services (EMS) personnel, yet little evidence supports their use among noninjured patients. We evaluate the association between out-of-hospital, intravenous access and mortality among noninjured, non–cardiac arrest patients.MethodsWe analyzed a population-based cohort of adult (aged ≥18 years) noninjured, non–cardiac arrest patients transported by 4 advanced life support agencies to one of 16 hospitals from January 1, 2002, until December 31, 2006. We linked eligible EMS records to hospital administrative data and used multivariable logistic regression to determine the risk-adjusted association between out-of-hospital intravenous access and hospital mortality. We also tested whether this association differed by patient acuity by using a previously published, out-of-hospital triage score.ResultsAmong 56,332 eligible patients, half (N=28,078; 50%) received out-of-hospital intravenous access from EMS personnel. Overall hospital mortality for patients who did and did not receive intravenous access was 3%. However, in multivariable analyses, the placement of out-of-hospital, intravenous access was associated with an overall reduction in odds of hospital mortality (odds ratio=0.68; 95% confidence interval [CI] 0.56 to 0.81). The beneficial association of intravenous access appeared to depend on patient acuity (P=.13 for interaction). For example, the odds ratio of mortality associated with intravenous access was 1.38 (95% CI 0.28 to 7.0) among patients with lowest acuity (score=0). In contrast, the odds ratio of mortality associated with intravenous access was 0.38 (95% CI 0.17 to 0.9) among patients with highest acuity (score ≥6).ConclusionIn this population-based cohort, out-of-hospital efforts to establish intravenous access were associated with a reduction in hospital mortality among noninjured, non–cardiac arrest patients with the highest acuity. Reasons why this occurred (cause and effect) could not be determined in this model.