For the purposes of this study, improvement was defined as having completed incidents in terms of process measures, i.e., a measure of the time expended during the execution of the call. Particular focus was placed on what are regarded by METRO EMS as the most important indicators: dispatch, response and mission times. The results indicate that between 2005 and 2008, the service had made significant improvements in its performance across all incident categories as is reflected by the improvement in the 'all other P1' categories.
The reason for this performance is likely to be found in the improved resources both in terms of staffing and of equipment. In addition, general improvements in operational structures, management capacity and personnel management may have contributed to this improvement. Improvements observed in the general ambulance operations may also reflect improvement throughout the health system as a whole. Notwithstanding, the introduction of the Flying Squad programme has resulted in significant improvements in the performance of obstetric and neonatal incidents.
The process adopted in the introduction of the Flying Squad model is significant for understanding its success. The engagement of key stakeholders early on, as well as their continued involvement throughout the implementation and evaluation phase, was crucial . Therefore, ownership of the programme was ensured and facilitation of meeting its principle objectives would be the responsibility of all involved.
It is in the analysis of the findings concerning the dispatch that the greatest impact of the Flying Squad programme is most apparent. The dramatic increase in the percentage of calls dispatched within 4 min is responsible for the bulk of the performance improvement.
Another aspect of the process that influences the success is that while it focussed on ring fencing of the ambulance and its staff, it had in fact ring fenced the dispatch of the resources as well. Efficiency of resource utilisation is built on the dispatcher's decision-making acumen, which in turn is determined by the quality and the accuracy of the information obtained. By determining the manner in which an incident is captured and evaluated (i.e. the development of predetermined criteria for dispatch), the programme has resulted in a more accurate and less vague form of communication. The result is greater dispatcher confidence and more accurate, rapid and appropriate dispatches stemming from clearly defined triage categories. This is evident in the substantial improvement in both the neonatal and maternity percentages of dispatched incidents in less than 4 min.
Perhaps one of the key questions that the study raises, and one that needs to be explored in later studies, is the question concerning dispatcher bias. The dispatcher, alone, determines which resource to activate and which P1 incident to dispatch; using his/her experience and judgement to prioritise P1 incidents before dispatching them. It is suggested that in the case of neonatal and maternity incidents, the fact that these patients are already accommodated at a health facility has led many dispatchers to defer their dispatch in favour of a primary response (such as to a road accident or patient's home).
Response time was the second process measure that was examined, and it demonstrated a significant improvement across all the incident categories evaluated. While an improvement was observed in the 15-min response performance for maternity incidents, the improvement in the neonatal category was not found to be statistically significant.
Neonatal transfers achieved a mean response time of 56 min in 2008 (118 min in 2005). The lack of available benchmarking as well as the vague definition of response times renders any meaningful comparison with times observed in other studies difficult . Whilst both Kempley et al. and Abdel-Latif have reported median response times for neonatal transfers of 85 and 75 min respectively, this cannot be used as a comparison for performance achieved in this study [8, 9]. Both used 'response time' as a measure from the initial discussion with the receiving facility to what they referred to as the 'first look'. They do however provide an indication of the time frames involved in executing these transfers.
Neonatal transfers have very specific requirements where safety is as important as speed of transfer. Specialised equipment is needed in terms of incubators, transport ventilators, medication and infusion pumps, etc. . The Advanced Life Support (ALS) skills required to perform these transfers safely are also in high demand, further hampering a speedy execution of the transfer request. This aspect, together with the high incidence of adverse events, has meant that services need to adopt a 'stay and play' policy when dealing with these incidents. It is on this basis that the Flying Squad included in its strategy a differential response for maternity and neonatal incidents. This evolved into the use of two intermediate life support (ILS) crews to perform the maternity transfers, while the ALS crew was reserved to attend to all the neonatal and critical obstetric transfer requests.
Greater efficiency was not only seen in the dispatch and response time performance, but is also evident in the analysis of the mean status times for each of the case type categories. Most notable is the status time of the neonatal incidents in which the longest mean 'on-scene' time was observed in both 2005 and 2008. This occurred despite the significant improvement in response time performance. Kempley et al. and Abdel-Latif also made this observation in their analysis of retrieval teams and their performance [8, 9]. This is likely due to the specialised nature of the neonatal calls. In the Flying Squad programme, this was addressed by cultivating an appropriate skill set among the crew during their 6-month rotation.
A further initiative was to ensure that each ALS crew had the necessary equipment to execute the neonatal transfers. Therefore, ambulances avoided wasting time in an attempt to locate a neonatal ventilator or working incubator. The allocation of a dedicated ambulance ensured that vehicle downtime was minimised as the crew had a greater sense of ownership and therefore took greater care. These measures ensured greater efficiency and culminated to reduce mean 'pre-hospital time'.
Although the study demonstrates a significant improvement in dispatch and response times, the absence of patient outcome measures has limited the conclusions that can be made. Teams may have executed these transfers more efficiently, but the appropriateness of the dispatch or the quality of the clinical management cannot be determined. It is therefore not known whether the introduction of the Flying Squad programme provided a better level of care (which was one of the programme's key objectives).
Secondly, the Flying Squad programme does not 'stack' calls. This means that when two requests are received simultaneously, only one is allocated: a second resource is then utilised from the general ambulance operations in order to service the second call. This is part of the operating procedures for the Flying Squad, and a measure of its impact on the level of service provided is desirable. However, the frequency with which this occurs is not recorded, and therefore the impact that this has on the improvement in performance cannot be measured.
A third limitation lies in the failure to determine the time spent at the hospital during handover. In so doing, a critical component of the transfer process was ignored. Therefore, the role that the hospital has to play in enabling greater efficiency was not examined. However, during discussions at Flying Squad meetings, clinicians from referral and receiving facilities anecdotally have made no changes in standard practices when receiving these patients.
Furthermore, as this study is a retrospective analysis, the impact of potential bias on the part of the investigator cannot be ignored. More research is therefore required, examining both the process and patient outcome measures, also focussing on the establishment and validation of a morbidity and/or mortality score based on the dispatch criteria of this Flying Squad programme.