Potentially unnecessary short peripheral catheter insertion in the emergency department: a retrospective,...

To estimate the frequency of potentially unnecessary short peripheral catheter (SPC) insertion among adults discharged from an emergency department (ED) and examine associated patient and clinical factors.

Retrospective, single-centre observational study using systematic time-based sampling. Logistic regression used cluster-robust standard errors at the sampled-day level.

The ED of Hospital Manacor, a secondary-care hospital in Mallorca, Spain.

Adults with a recorded SPC who were discharged from the ED during 47 sampled calendar days in 2023.

The primary outcome was potentially unnecessary SPC insertion, defined as no documented intravenous medication, fluids or other intravenous treatment during the complete ED episode. Secondary analyses examined an expanded classification that additionally included patients receiving only one selected intravenous treatment with potential non-intravenous alternatives and examined factors associated with the primary outcome.

Among 1488 patients, 388 (26.1%) had no documented intravenous treatment; the expanded classification included 616 (41.4%). Increasing age was associated with higher odds of the primary outcome (Model 1 OR 1.03 per year, 95% CI 1.02 to 1.04), whereas female sex was associated with lower odds (OR 0.73, 95% CI 0.58 to 0.91). A one-level decrease in clinical urgency was not associated in Model 1 (OR 1.18, 95% CI 0.98 to 1.42), but was associated after adjustment for triage presentation syndrome (OR 1.24, 95% CI 1.03 to 1.50) or discharge diagnosis syndrome (OR 1.21, 95% CI 1.02 to 1.45). Each syndrome variable was associated with the outcome in its respective model (global p<0.001 for both).

Approximately one-quarter of discharged patients with a recorded SPC had no documented intravenous use. Potentially unnecessary SPC insertion was associated with age, sex and clinical presentation, while its association with triage acuity depended on adjustment for clinical context. This operational outcome does not establish the appropriateness of individual insertions or support causal inference.