We found that an in-transit critical event occurred in 1 in every 20 urgent air–medical transports of adult patients in a transport system covering a large geographic area. This rate is low but clinically important given the serious nature of these events and the challenging out-of-hospital environment in which they occurred. In-transit critical events were independently associated with female sex, use of mechanical ventilation or hemo-dynamic instability before transport, transport in a fixed-wing aircraft, duration of transport, on-scene calls and type of crew.
The incidence of critical events found in our study cannot be compared directly with rates reported in the literature, given the heterogeneity in the criteria for inclusion, reporting of data and definitions of outcomes in other studies.3,11–15
The rate we observed is similar to that reported in a study of out-of-hospital transport by fixed-wing aircraft with comparably defined outcomes12
and higher than the rate in a smaller series from a single-centre organization.16
The rate of 1 critical event per 12.6 hours of transit time in our study is comparable to rates in other acute care fields of medicine.4,17–19
Critical events comprised mainly hemodynamic deterioration and administration of major resuscitative procedures during transport. Procedures for airway management were frequently administered, and the rate of successful intubation we observed is consistent with rates reported in other studies.20–24
We identified several patient- and transport-level factors that were independently associated with in-transit critical events. Female sex was a significant factor even after adjustment for patient- and transport-level confounders. This association persisted in the sensitivity analysis that included transports with primary care paramedic crews. Evidence is growing that disparities exist between men and women in treatment, utilization of health care resources and outcomes across a wide range of diseases.25–30
In our study, this observation may be attributable to differences between men and women in how diseases present, how they are treated and patterns of referral and transport-related requests. Systemic biases during assignment of transports could lead to delayed transport of women, with resulting physiologic instability and a higher risk of critical events. Alternatively, the criteria used to triage patients may underestimate the risk of deterioration among female patients. We observed that men were more often transported by critical care paramedic crews even though their unadjusted rate of in-transit critical events was not significantly different from that of women. Although conclusions about appropriate matching of patients to resources cannot be drawn, better understanding of the root causes of this observation may help optimize allocation of transport resources according to levels of acuity and transport risk of patients.
Although some critical events are likely unpredictable (e.g., unanticipated sudden cardiac arrest), many of the critical events we observed were related to airway management. Some of these events might have been prevented with better elective airway management before transport. In our study, elective intubation of a patient before transport was not a standardized or protocol-driven decision. Further study of this issue could inform policy development and crew training. Our data suggest that patients who have hemodyamic instability before transport should be given special consideration for pre-emptive airway management, given the strong association with in-transit critical events in this group.
Use of fixed-wing aircraft was associated with in-transit critical events. We hypothesize that this association was attributable to preferential assignment of these aircraft to patients in remote areas and to residual confounding in this patient population rather than to characteristics of the aircraft itself. Supporting this hypothesis was the reversal of this association in short-distance transports in the sensitivity analysis, in which sicker patients with high-risk injuries were more often transported by helicopter. We also observed a 2% increase in the odds of critical events with every 10-minute increase in duration of transport, which would be expected with increasing flight times. The association between higher levels of paramedic skill and higher rates of critical events was likely related to the assignment of patients who were at highest risk of deterioration to more advanced crews. The scope of practice of advanced care para-medics is similar to that of critical care paramedics, and we rarely observed the few procedures carried out uniquely by critical care paramedics (e.g., needle cricothyroidotomy constituted 0.3% of all observed procedures; n = 2). The low rate of critical events among primary care transports may be related to the assignment of crew to low-risk patients or to ascertainment bias. The limited scope of practice of these basic paramedics censored them from many procedures included in our outcome measure.
Issues specific to jurisdiction and context are notable in our study. The geography of Ontario requires longer air–medical transports than those typically reported in the United States31
In isolated areas, the levels of skill and training of paramedic transport crews may exceed that of the available local health providers. Thus, crews may deliver primary care and resuscitation “on scene” when picking up a patient. Requests for transport are almost never declined by Ornge because of acuity of illness; when indicated, even patients with the most unstable conditions are transported with aggressive resuscitation before and during transport.
Strengths and limitations
Our study has several strengths. It was population-based and involved the largest published cohort of transported patients. The procedures and clinical protocols used by Ornge Transport Medicine are similar to those of providers in other jurisdictions, enhancing the generalizability of our findings.
Our study was limited by the use of retrospective data. We were unable to obtain a patient-level score of severity of illness that would allow direct comparison of the incidence of critical events in our study to those reported for other acutely ill populations. Hemodynamic instability and need for assisted ventilation before transport provide only crude surrogate measures for severity of illness. We excluded almost one-quarter of the transports in the data set to limit the cohort to patients who were at highest risk of in-transit clinical deterioration. This large proportion of low-acuity transports is primarily related to a greater need for transport among patients in isolated communities to provide basic care and consultation. Indeed, none of the 10 644 patients designated as low risk had baseline hemodynamic instability or were receiving assisted ventilation before transport; only 1.8% experienced an in-transit critical event, and no deaths occurred.
We found that a critical event occurred in transit in 1 in every 20 air–medical transports of adult patients. Most often, these events were the administration of resuscitative procedures or hemodynamic deterioration. Female sex, mechanical ventilation or hemodynamic instability before transport, transport in a fixed-wing aircraft, duration of transport, on-scene calls and type of crew were independently associated with in-transit critical events. These findings may be useful in the refinement of protocols for the optimal alignment of transport resources with patients at high risk of requiring in-transit resuscitation. Our results may also inform the training of paramedic transport crews and continuing education to address the most common scenarios they face. Better understanding of the patient-level predictors of adverse events may also allow for improvements in the preparation of patients by hospital staff before transport.