Introduction

Prehospital endotracheal intubation (ETI) is a vital procedure for patients in life-threatening conditions, with first-pass success rates for physician-performed ETI reported between 86.2% and 96.6% [1–3]. Given the high risk of complications such as hypoxia, hypotension, and aspiration, optimizing success is essential [4]. ETI is a complex process that requires appropriate patient assessment, medication selection, environmental control, and communication among team members. Previous studies have shown that not only procedural techniques but also provider experience scene environment, and the use of checklists and protocols contribute to improving first-pass success and patient outcomes [2,5–10].

Standardized protocols for prehospital ETI and anesthesia have been developed and evaluated in regions such as North America, Europe, and Australia. These protocols typically cover preparation, induction, intubation, and post-intubation management, and have been associated with improved first-pass success rates and a potential reduction in early mortality [2,9–11] They also improve educational outcomes and facilitate continuous quality improvement through the collection of uniform data [12,13].

In Japan, physician-staffed prehospital care via rapid response cars and helicopter emergency medical services (HEMS) has expanded. However, several challenges hinder the standardization of physician-performed prehospital ETI. Practice settings, provider training, and available resources vary across regions, and no national guidelines currently exist [14]. As a result, outcome data related to physician-performed prehospital ETI remain limited and heterogeneous, making systematic evaluation and generalization within the Japanese context challenging.

This study aimed to explore expert perspectives on standardizing physician-performed prehospital ETI in Japan. Given the lack of formal domestic guidelines and operational variability, we employed a structured consensus process using the Delphi method [15]. Here, we present the first-round findings, comparing perceived appropriateness and implementation tendencies with internationally reported practices.

Materials and methods

Identification of Delphi items

Initial candidate items for the Delphi process were identified through a combination of literature review and a pilot study. S1 Fig. outlines the overall study process.

For the literature review, PubMed/MEDLINE was searched in November 2024 to identify studies published between 2015 and 2024. Eligible studies addressed prehospital care in combination with ETI or anesthesia. As the objective of the review was to inform item generation for the Delphi process, studies relevant to the standardization of prehospital ETI were considered, and heterogeneity in study design, target populations, and outcome measures was accepted. Studies were excluded if they (1) were conducted exclusively in emergency department settings, (2) were limited to intensive care unit settings, or (3) involved paramedic-performed ETI. Articles were screened by title and abstract by one reviewer (H.T.) and assessed for eligibility by a second author (K.O.). A total of 177 studies were included for item generation. From these sources, 79 preliminary items were extracted for the Delphi process.

These items were evaluated in a pilot study conducted in December 2024 by three authors (H.N., H.O., and Y.Y.), who independently reviewed the draft items and provided feedback on content appropriateness and completeness. Based on their input, a supplementary manual search was conducted to include additional topics suggested during the pilot study. Several additional practices were added, including laryngoscope selection according to patient characteristics, assessment of dental stability, equipment preparation such as stethoscope availability, and confirmation procedures after ETI. The final list comprised 90 items. These items were categorized by the authors along a chronological axis—ranging from initial preparation to induction, confirmation, and troubleshooting—and grouped into eight domains (Fig 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Distribution of recommendation items across eight domains.

https://doi.org/10.1371/journal.pone.0346146.g001

Derived from the literature and a pilot study, a total of 90 recommendation items were organized along a chronological axis from preparation to troubleshooting and grouped into eight procedural domains. The number and percentage of items in each domain are indicated, illustrating the relative emphasis across phases of prehospital endotracheal intubation.

Results

Panel characteristics

The survey was conducted from February 24 to March 17, 2025. Of the 58 physicians invited, 40 (69.0%) responded. No protocol deviations occurred during the first round of the Delphi process. Panel characteristics are shown in Table 1. All participants were male emergency physicians, with only 2.5% also board-certified in anesthesiology. Sixty-five percent had more than 15 years of prehospital care experience. Among their affiliated institutions, 27.5% reported access to hands-on training, while 37.5% reported no formal training.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Characteristics of participants in the first-round Delphi survey.

https://doi.org/10.1371/journal.pone.0346146.t001

Consensus rates and domain-level trends

Table 2 summarizes domain-level consensus and agreement rates, while S2 Table presents the full distribution of item-level responses across the 4-point Likert scale for both perceived appropriateness and implementation. Of the 90 items, 72 (80.0%) reached the predefined consensus threshold for appropriateness, while 57 (63.3%) reached consensus for implementation. At the domain level, the highest proportion of items reaching the consensus threshold for appropriateness was observed in the Confirmation domain (100%; 9/9 items), whereas the lowest was observed in the Environment during endotracheal intubation domain (25%; 1/4 items).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Domain-level summary of consensus and agreement rates.

https://doi.org/10.1371/journal.pone.0346146.t002

At the domain level, agreement rates further characterized the alignment between perceived appropriateness and actual implementation. Across all domains, agreement rates for implementation were lower than those for appropriateness. Such differences were observed in domains characterized by high agreement on appropriateness but lower agreement on implementation, including peripheral equipment and medications and dosage considerations (median appropriateness agreement, 82.5% for both; median implementation agreement, 67.5% for both). Similar patterns were also observed in domains involving role assignments, troubleshooting, and environmental considerations.

These discrepancies were also evident at the item level. For example, checklist use was rated as appropriate by 65.0% of respondents, whereas only 10.0% reported that it was routinely implemented.

Discussions

This study examined expert opinions on physician-performed prehospital ETI in Japan via a first-round consensus process. While most items were considered appropriate, only 63.3% were reportedly implemented. Substantial gaps were observed between appropriateness and implementation, particularly regarding checklists, peripheral equipment, and medication protocols.

International efforts to standardize prehospital ETI have advanced in North America, Europe, and Australia [2,8,10,22,23]. These typically address key domains such as environment, assessment, airway preparation, drug administration, and troubleshooting [7,24,25]. In contrast, Japan lacks national guidelines, resulting in fragmented training and practice across institutions. In our study, only 27.5% of participants reported access to hands-on training, and 37.5% reported no formal training, suggesting a potential need for nationwide education frameworks. For example, although 65.0% rated checklist use as appropriate, only 10.0% reported its routine implementation.

This study revealed clear variation across domains in the alignment between perceived appropriateness and actual implementation. Domains such as Patients and Indication, Oxygen and Equipment Preparation, and Confirmation showed consistently high ratings for both appropriateness and implementation, suggesting that these practices are widely accepted and feasible within routine prehospital workflows. These domains likely represent foundational elements of airway management that require minimal additional resources. In particular, the uniformly high agreement observed in the Confirmation domain may reflect the recognition of these items as fundamental safety practices, although such consensus could also be influenced by shared normative expectations among clinicians.

In contrast, lower implementation levels were observed in domains such as Medication administration, Peripheral Equipment, and Role Assignments, despite moderate to high perceived appropriateness. These domains appear more vulnerable to contextual constraints related to training, staffing structure, and resource availability. To better understand these gaps, the observed discrepancies can be broadly interpreted through three interrelated dimensions: educational, structural, and operational factors.

From an educational perspective, substantial heterogeneity was observed in provider experience, training opportunities, and procedural preparedness among prehospital physicians. In Japan, prehospital care is predominantly delivered by physicians with emergency medicine backgrounds, and access to structured airway training varies across institutions. This variability may particularly affect the implementation of practices requiring advanced pharmacologic management or adjunct airway devices. Similar education-related challenges have been reported in other physician-staffed prehospital systems operating under staffing and resource constraint [26–28].

Structural and operational characteristics of Japanese prehospital care may further limit the feasibility of implementing certain airway management practices. Free-text comments indicated that ETI is frequently performed by a single physician, often without designated assistants or clearly defined role assignments, constraining task delegation and team-based approaches. Respondents also described variability in equipment availability, medication stocking, and procedural environments across institutions. Environmental constraints, including limited procedural space and difficulty interpreting environment-dependent concepts such as “360-degree access” in ambulance-based settings, further illustrate challenges in directly applying internationally recommended practices to the Japanese prehospital context.

This study represents the first nationwide and structured effort to explore consensus among experts on physician-performed prehospital ETI in Japan. By evaluating both the perceived appropriateness and implementation status of internationally derived recommendations, the study revealed discrepancies between conceptual agreement and actual clinical practice, as well as contextual and institutional differences. These findings provide valuable groundwork to inform the development of practical and context-specific protocols. Further consensus-building may help refine expert perspectives and inform feasible, evidence-informed guidelines for the Japanese setting. Beyond consensus efforts, future research should focus on evaluating real-world protocol implementation and the role of structured education and simulation-based training within physician-staffed prehospital systems.

Several limitations should be acknowledged. First, the participant pool was demographically and professionally unbalanced, consisting exclusively of male physicians, most of whom were board-certified in emergency medicine, with relatively limited participation from anesthesiologists. This may have introduced a degree of bias in the perspectives captured, and broader representation would strengthen future rounds of the consensus process. In addition, detailed information on institutional characteristics was not collected, which may have influenced reported practice patterns. Second, the study reflects only the first round of a Delphi-style process, and further rounds are needed to achieve more robust consensus. Some variability in item interpretation was observed, which is inherent to early-round Delphi processes, and the influence of subjective judgments cannot be fully excluded. Third, the recommendation items were generated through a broad, structured review of international literature, which necessarily encompassed heterogeneous evidence quality, clinical contexts, and implementation levels. Accordingly, the content validity of individual items should be regarded as provisional and exploratory at this stage. In addition, the use of a relatively modest consensus threshold (≥70.0%) was intended for first-round screening rather than definitive endorsement, with item refinement and validation anticipated in subsequent Delphi rounds.

Conclusions

This study identified key discrepancies between expert agreement and reported implementation regarding physician-performed prehospital ETI practices in Japan. The findings offer an important foundation for future standardization efforts tailored to the specific characteristics of Japanese prehospital care.

Supporting information