Our findings add context and perspective to the clinical decision-making involved in determining whether or not a pregnant patient is safe for transport between two hospitals. The findings are amenable to formulating triage algorithms, such as Fig. 1, whereby clinical teams can rapidly assess the best course of action based on the evidence presented. In particular, the findings demonstrate that preterm labor can progress rapidly, and high-risk features for precipitous delivery include rupture of membranes, cervical dilation beyond 3 cm, and frequent contractions that are regular and strong.
It is important to note that unanticipated labor or delivery in the patient population requiring interfacility transport is largely different than the controlled progression of labor in a modern hospital labor and delivery unit. The effects of anesthesia and perinatal interventions common on labor and delivery units augment cervical change such that it is less precipitous than would be expected without any such interventions in an unplanned and potentially precipitous delivery encountered in an emergency 6,7. Furthermore, precipitous deliveries, defined as those occurring in less than 3 hours, have a higher likelihood of happening outside of the controlled environment of a labor and delivery unit 7. As such, findings from in-hospital labor and delivery units are not necessarily representative of the high-risk and unanticipated deliveries encountered in the prehospital and interfacility transport environments. Our findings help fill in this gap and provide better context for clinical decision-making as it relates to the interfacility transport of high-risk obstetric patients.
Another clinical context in which these findings are useful is for clinicians balancing the risk of adverse events during transport, the requirements of the EMTALA in the US, and the potential benefit of a neonate requiring higher-level care being born at an appropriately resourced hospital capable of delivering that care. In this study, nearly all the neonates born in the group with the primary outcome would benefit from advanced neonatology capabilities, as all had gestational age less than 35 weeks, and the majority had gestational ages less than 30 weeks.
As healthcare resources within the US continue to consolidate and centralize, accessing prenatal care and obstetric services will remain a challenge. In this dataset, we found an association between high-risk deliveries and absence of prenatal care. Lack of access to care in obstetric patients prevents medical optimization and planning for high-risk pregnancies. Without improved access, adverse events such as unexpected preterm deliveries, could become more frequent with higher morbidity and mortality. Furthermore, patients with difficulty accessing any type of healthcare could also conceivably present later in the course of an acute health crisis, further decreasing the odds for a favorable outcome.
Lastly, the results imply that with adequate fetal, uterine, and maternal clinical assessment capabilities, there is limited utility in an obstetric transport nurse program if an appropriate triage mechanism is in place. Furthermore, deployment of the obstetric transport nurse led to a mean increase in response time to the patient of over twenties minutes compared to a standard transport team configuration (data not shown). This was due to the standard transport team having to divert to pick up the obstetric nurse from a centralized labor and delivery unit that was away from the point of ambulance dispatch. In the era of staffing shortages exacerbating a lack adequate nursing coverage within hospitals in the US, these findings may inform more efficient deployment of valuable and limited nursing resources.
Limitations
This study had several limitations worth consideration. First, there was implicit selection bias in that the obstetric transport team configuration was being deployed at the behest of the requesting physician at the sending hospital for pregnant patients presumably perceived to be high risk. The patient population therefore contained a preponderance of preterm patients with more comorbidities than would be expected in a more generalized sample. Secondly, the outcome was rare and could be subjectively influenced by the clinical decision-making of clinicians at both the sending and receiving hospitals. Practice bias in either group will lead to non-random findings as it relates to the outcome and physiological parameters. For example, the presence of prolonged, recurrent late decelerations often led to cesarean delivery if that capability was present. The same is likely true of varying degrees of cervical dilation and the nature of contractions. Another potential limitation is the speed at which patients who stayed at the sending hospital were delivered. While many of the cases were taken directly to the OR, there is some uncertainly in how precipitous the labor was, although the rate of cervical change in all but three of the cases implies precipitous delivery. Missing data existed in this data set for patients that did not have a cervical exam performed or who did not have contraction frequency measured. These observations were not included in the logistic analysis. This diluted the power of the study and ability to precisely measure the impact of these factors. Lastly, the study took place within a single region of the United States with the least favorable quartile of morbidity and mortality for both maternal and neonatal patients in the perinatal period. Thus, the patient population described may have worse outcomes than would be expected in other regions with better healthcare resources and/or availability thereof.