The two most important areas for future research as suggested by the expert panel turned out, not surprisingly, to be two of the most controversial and difficult areas of clinical pre-hospital practice: the role of advanced practitioners in pre-hospital critical care and advanced pre-hospital airway management. Despite having received much attention, both of these areas remain unresolved.
The effect and efficiency of physician manning of pre-hospital EMS has been the subject of several studies in recent years [5, 6, 44–52]. This subject has been, and remains, controversial and is linked to the debate between advanced and basic life support . The results have been contradictory to some extent, even in similar studies. For example, in two retrospective comparisons of blunt trauma patients treated by physicians or paramedics, Garner et al.  found a lower mortality in the physician group, whereas Iirola et al.  found a trend towards higher mortality in the physician group. However, both studies found that although treatment in the physician groups was more extensive, it did not delay arrival at the hospital. One review article comparing pre-hospital treatment by physicians or paramedics supported the inclusion of physicians in EMS for pre-hospital trauma care , whereas another found increased survival in trauma patients and a trend towards increased survival in cardiac arrest, myocardial infarction and respiratory distress . Both reviews found that the number of articles was few and that the quality of the studies was variable. The topic has remained controversial due to study design and a lack of consistent results in addition to confounding factors such as publication bias in favour of physician-based services, comparisons of different transport methods and differences in interventions performed by the respective groups [5, 49].
Pre-hospital airway management is one such advanced intervention. Despite being the topic of several articles, including review articles and a Cochrane review [37, 53–55], a recent all-time literature review concluded that the data presented in studies focusing on pre-hospital airway management in adults were deficient and inconsistent , making the majority of studies non-conclusive and invalid. In the last two years, several large, well-designed pre-hospital studies have been initiated, and importantly, the first prospective, randomised, controlled clinical trial of pre-hospital intubation in adult patients with severe head injury, was published in 2010 . Nevertheless, while this study represents a milestone in pre-hospital airway management research, it focuses on paramedic intervention, and the authors conclude that more research is needed. Our knowledge of the crucial factors associated with a good outcome of pre-hospital intubation remains poor, and more high-quality studies are needed. In 2009, an Utstein template for documenting and reporting pre-hospital advanced airway management was published . This template may increase the validity of future studies on this subject .
The third priority was the timing and necessity of critical interventions. A recurring controversy concerns which advanced interventions should be performed in the pre-hospital setting and which can and should wait until hospital arrival. Established time concepts such as 'the golden hour of trauma' have been challenged due to the lack of scientific evidence [58–60]. Concerning time in pre-hospital critical care, there has been a major debate in recent decades between the main approaches of 'scoop and run' vs. 'stay and play' . Connected to the subject of staffing in the literature, the apparently improved outcome of trauma patients treated by basic life support teams compared with advanced life support teams has been explained partly by the extra time spent on the accident site by the latter, as mentioned in several review articles [27, 30, 61]. While the focus of the debate has been on treating the patient at the accident scene in contrast to quick transport to definitive care, studies that focus on which procedures can be omitted at what times while considering all variables are scarce. There are several aspects to time questions, such as which conditions/clinical situations really cannot wait, how long it takes to arrive at definitive care in different systems/geographical areas, and the risks vs. benefits of interventions on-scene in particular situations. The questions posed by the group address these issues in specific ways.
Incidentally, time consumption has been one of the major concerns regarding the implementation of the expert panel's priority number four, pre-hospital ultrasound (US) . New technology continues to make in-hospital standards for monitoring and diagnosis applicable in the pre-hospital setting, but the potential benefit to the patient remains to be proven. The expense, time consumption and added weight of taking new technology to scenes can be considerable. The need for education and training to maintain the skills necessary to operate the equipment is also important. The group chose pre-hospital US as the highest-priority technological tool to be evaluated. Small, battery-powered US machines have proven feasible in the pre-hospital field [63, 64]. However, despite receiving broad attention in the past decade , a review article from October 2010 concluded that there is currently no evidence in the literature to support that pre-hospital US of the abdomen or thorax improves the treatment of trauma patients , and a review article of echocardiography in cardiac arrest from June 2010 concluded that no studies so far have shown an improved outcome through the use of this imaging modality . Conversely, a prospective trial concerning ultrasound in cardiac arrest, published in November 2010, found that application of advanced life support-compliant echocardiography in pre-hospital care is feasible, and alters diagnosis and management in a significant number of patients. However, further studies into its effect on patient outcomes were warranted . The clinical situations in which US should be examined as a useful tool are expanding in number, and publications on its use in brain trauma , airway management , differentiating chronic obstructive pulmonary disease exacerbation from heart failure  and rapid treatment of fractures  mirror the huge expansion and availability of US in emergency medicine.
The final area of interest to the group is less related to hospital medicine and more to the core business of EMS systems. The group recognised that accurate dispatch is pivotal to success, considering the relatively scarce resources and the importance of early attendance at incidents with time-critical injuries. A systematic review concerning physician-staffed EMS dispatches in 2009 found 34 studies that met the inclusion criteria. However, the study concluded that there are few studies describing the validity of criteria defining an appropriate physician-staffed EMS dispatch and that the results from these studies lack general applicability . A study published in 2010 that provided an overview of dispatch criteria related to physician-staffed EMS organisations in Europe found a lack of uniformity in the use of these criteria for trauma assistance on a national and international level. The study concluded that future research should aim to identify a general set of criteria with the highest discriminating potential. The group recognised that accurate dispatch is pivotal to success, considering the relatively scarce resources and the importance of early attendance at incidents with time-critical injuries.
The recent survey of the 999 EMS Research Forum determined the most important priority topic to be the 'Development of EMS performance measures other than response times for use in performance management, audit and research' . This subject also generated considerable discussion at the consensus meeting, as the value of pre-hospital research is limited in the absence of appropriate outcome measures that reflect the impact of pre-hospital interventions. Current evaluations of the quality of pre-hospital care are frequently as crude as the time from incident to arrival at hospital [73, 74]. Pre-hospital interventions that delay transfer time will be difficult to introduce in such systems unless improvements in clinical condition or outcome are measured and demonstrated. When survival to discharge from hospital is used as a measure of pre-hospital care it can be confounded by the entire chain of in-hospital treatment, this makes the effect of pre-hospital interventions difficult to demonstrate. The group discussed alternative measures for evaluating the quality of pre-hospital care, including isolated pre-hospital physiologic parameters and patient satisfaction, as additional outcome measures. The evaluation of the quality of pre-hospital care is of crucial importance to all of the mentioned areas in this exercise, but it did not emerge as a separate research suggestion in the initial rounds. In an article by Jones et al., in 1991, this topic was listed as one of the key questions in pre-hospital research . Twenty years later, the question seems to be as essential as ever.
The scientific value of Delphi Surveys and NGTs has been questioned [75, 76]. However, consensus methods continue to be a useful tool for assessing the extent of agreement on matters in which hard evidence is difficult to obtain, and the results can serve as input for other processes [40, 77, 78]. Some critics of group meetings have argued that verbally skilled participants can monopolise the group with arguments over wording  and that 'strong' members of the group may take control of the consensus process to defend their own viewpoints . The project group tried to avoid this problem by anonymising the three first stages and by completing the final ranking prior to the consensus meeting.
The prioritisation system used in our study generated discussion at the meeting as to whether it accurately reflected the expert panel's views. Methods to achieve consensus and methods to prioritise suggestions have been widely discussed, but there is still no method that is considered a gold standard . We gave points to reflect how often a particular subject was suggested in addition to the points obtained by priority. Whether this approach provides a more accurate description of the group's opinions can be questioned; however, this did not affect on the resulting top five areas in this particular case.
Nineteen experts participated in our process. A larger number of researchers from more countries may have improved the scientific value of the project. However, physician-based emergency services continue to be a common supplement to pre-hospital medicine in only a few parts of the world [3, 5, 6, 45, 81]. We sought to include countries with well-developed and well-integrated physician-staffed EMS similar to the Scandinavian model. A majority of the delegates (Eleven out of the 19) were from the Scandinavian countries and some major European countries were not present. Invitations were sent to German, French and Czech researchers but unfortunately none of these were able to attend this project. The involvement of clinicians from more countries may have yielded a different priority list, for future projects the group will try to increase the number of countries represented in the expert panel.