Patient safety incidents reported by patients and witnesses in prehospital emergency medical services: a multicentre prospective study
Juan Antonio Péculo-Carrasco, Carlos Romero-Olóriz, Carmen Gutiérrez-García, Lorena Cuesta-García, Cristina López-López, María José Luque-HernándezBackground and importance
Patient safety incidents – events or circumstances that could have resulted, or did result, in unnecessary harm to a patient, and referred to hereafter as incidents – fall into four categories: reportable circumstances, near misses, no-harm incidents, and harmful incidents (‘adverse events’). In prehospital emergency medical services (EMS), most research has addressed only harmful incidents, leaving the other three categories poorly characterized.
Objectives
The primary objective was to determine the frequency of incidents reported by patients and witnesses after prehospital EMS care.
Design
Multicenter prospective study, following the Strengthening the Reporting of Observational Studies in Epidemiology statement.
Settings and participants
The eight provincial EMS services of Andalusia, Spain (April 2021–March 2022). Recruitment was consecutive. Eligible participants were patients transported to the hospital by EMS teams, or a witness if not possible, interviewed by telephone 15–20 days after the episode (the EMS attendance and hospital transport).
Outcome measures and analysis
The primary outcome was incident reporting frequency; secondary outcomes were reported harm, sociodemographic characteristics, and a single-item safety self-rating (a four-point scale: ‘very safe’, ‘safe’, ‘fairly safe’, or ‘unsafe’). Rather than relying on spontaneous or passive reporting, data were collected through a systematic, structured telephone interview in which participants were explicitly asked whether four specific incidents – error, fall, harm or injury, or another incident described in their own words – had occurred. Frequencies were reported with confidence intervals (CIs); secondary associations used bivariable analyses and, for sociodemographic characteristics, binary logistic regression, with CIs unadjusted for multiplicity.
Main results
Among 3156 episodes (1756 patients and 1400 witness respondents; response rate 40.4%), an incident was reported in 166 (5.3%; 95% CI = 4.55–6.08). Reported harm was identified in 58 of these incidents (34.9%), leaving two-thirds without reported harm. During the same interview, incident reporting rose from 2.8% among respondents who felt ‘very safe’ to 26.3% among those who felt ‘unsafe’.
Conclusions
In this prospective study, systematic follow-up of patients and witnesses of EMS transportation identified a declared safety incident in 5.3% of episodes, most of them without reported harm.