This paper presents a set of potential QIs for quality measurement of P-EMS. Using a modified nominal group technique an international expert panel achieved consensus on these QIs that describe six quality dimensions and include structure-, process- and outcome-indicators.
Quality measurement of pre-hospital services has been identified as a high-priority research area and pivotal to achieve improvement in care [
12,
13,
23,
24]. However, identifying valid quality indicators that are feasible to collect in the operational context of pre-hospital services has been a challenge [
25]. We deliberately asked the experts to propose quality indicators themselves, not simply selecting from a pre-defined list. The rationale behind this was to make this process as open as possible in order to achieve a broad selection of proposals. The multidisciplinary composition of the expert panel was partly to facilitate this broad approach.
A premise for this study is that the principles for quality measurement in health care also applies to P-EMS. P-EMS is the practice of medicine outside hospitals, and we find it reasonable to accept this premise. The six core characteristics of quality depicted in Fig.
1 were defined by Institute of Medicine, naming them dimensions of quality [
22]. Each of these is distinct and no one is defined more important than the others. When measuring quality, all six quality dimensions should be appraised. We argue that this multidimensional approach to quality measurement seems particularly reasonable for services with a highly heterogenic patient population, like P-EMS. Patients cared for by P-EMS differ a lot: Neonates vs. elderly patients, medical vs. surgical diagnoses, patients rescued from open water vs. Intensive Care Unit transferals [
2,
5]. What is considered high quality care for each patient will be context-specific. With this complexity of cases, treatments and operational contexts, we argue that adequate quality measurement of P-EMS should be multidimensional.
Quality dimensions
Timely care is about reducing needless and potentially harmful delays before the patient receives specialized care from the P-EMS. Traditionally, attempts on quality measurement of pre-hospital services, have been limited to data on time-variables corresponding the quality dimension
“timeliness” [
13,
26]. Studies of EMS have shown that response time affects outcome only for a small group of patients [
27,
28]. Moreover, time variables describe the logistics, but not the provided care. Response time of P-EMS is measured in QI 2 “Time to arrival of P-EMS” and is indeed important for some time critical conditions such as cardiac arrest and major trauma [
29]. However, the importance of short response times cannot be generalized to all emergency responses [
30]. In selected situations, too much emphasis on timeliness is misleading in respect of what really represents quality for the patient. In the United Kingdom paramedics criticized the use of a time target structure measure (eight-minute response time for 75% of category A or emergency calls) as the main performance indicator in EMS. They argued this QI was “too simplistic and narrow” and that it could also increase risk for patients and ambulance crews [
31]. An example may illustrate the limitation of time-variables as the sole QIs in P-EMS: Performing an ultra-sound scan of the traumatized patient may prolong the time on scene slightly. However, the examination can result in changes in treatment or triage [
32], hence making the extra time spent on scene well worth.
The quality dimension
“safety” focus on safety issues related to P-EMS responses for patient, EMS-staff or others. The safety issues can be medical, technical or operational. P-EMS operates rapid response cars and helicopters, all activities associated with operative risks for patients, bystanders and crew [
33]. Additionally, P-EMS care for severely injured or ill patients without access to safety initiatives as seen in hospitals e.g. senior assistance or access to patient history. Moreover, the pre-hospital environment can be associated with hazards like extreme temperatures, traffic and difficult access requiring application of rescue techniques [
34].
The quality dimension “
efficiency” is about avoiding medical waste; including waste of use of P-EMS personal, equipment and energy. Advanced major incident management reduce over-triage and is an example of how to prevent waste of resources [
35]. This issue is covered in QI 22, which measures the proportion of P-EMS doctors who have completed a major incident management program.
“Equity” is about ensuring that quality of care is provided equally regardless of the patient’s gender, ethnicity, geographic location and socioeconomic status. P-EMS contributes to equitable care by reducing transportation times (when using a helicopter) and by bringing the hospital competencies to the pre-hospital environment. This role of P-EMS can also be defined a governmental objective [
36] as an initiative to give people living in scattered spread populations specialized care within due time. Thus, a more equitable access to centralized medical treatments like neurosurgery and invasive cardiology can be provided. The expert panel argued that the involvement of a physician or a paramedic from P-EMS in the dispatch decision would secure the most correct use of P-EMS, thus contributing to equitable care. This is addressed in QI 9 «P-EMS involvement in dispatch decision».
“Effectiveness” is about ensuring that provided treatment is evidence-based. Care proven effective should be provided, thereby preventing undertreatment. Similarly, care proven ineffective should not be provided, thereby preventing overtreatment. There is some evidence that the use of physicians in EMS for selected patient groups, improve outcome or proxy outcomes such as physiological variables [
1,
37]. However, the current documentation on the impact of P-EMS initiatives is controversial and, therefore, effectiveness QIs are difficult to derive from the literature. The expert panel combined existing evidence with the experience and considerations of all panel members. One of the resulting QIs, QI 12 “Advanced Treatment”, addresses care considered indicated, but not feasible without the competence of P-EMS. Please note that withholding unethical or unnecessary treatment by the P-EMS physician also was defined as “advanced treatment” by the expert panel. Thus, critical decision making as illustrated for pre-hospital advanced airway management by Rognås et al.[
38], is recognized as a part of quality care.
“Patient-centeredness” is about ensuring that care is responsive to individual needs. Although most stakeholders and clinicians in P-EMS presumably put the patient in the center of the care, the study group wanted to secure that the patients were represented in the expert panel. Therefore, a leading representative from a major patient organization was invited to join the expert panel. Developing quality indicators for this quality dimension in P-EMS is challenging, primarily because many of the patients cared for by the service are unconscious or at least not capable of expressing their own needs in their usual manner. This can be due to the clinical condition itself, stressful situation or pharmacological interventions. The needs of the patient’s family, however, can be expressed more easily. Moreover, the term “patient-centeredness” has been argued expanded to “patient- and family-centeredness” [
39]. Patient- and family-centered care is based on the beneficial partnership between patient, family and health care workers, and it can be applied to patients in all ages and in any health care setting [
39,
40]. As a surrogate for measuring the patient’s needs, the needs of the patient’s relatives could be addressed, as defined in QI 15 “Care for relatives”. This QI addresses the relatives’ needs, including the need for practical and emotional assistance.
Types of quality indicators
J. Mainz has reviewed the strengths of structure-, process- and outcome-indicators [
14]. Structure indicators are found most useful when they predict variations in processes or outcomes of care. Process indicators are particularly useful when coping with short time frames, low volume providers and when tools to adjust or stratify for patient related factors are difficult to apply. Comparison of process indicators are generally easier to interpret and more sensitive to small differences than comparison of outcomes data. Based on these characteristics, we consider process indicators particularly suitable for continuous quality measurement of P-EMS. Although necessary to get information about a patient’s final outcome, long-term outcome indicators appear less feasible for measuring the isolated quality of P-EMS. From a patient is admitted to hospital by P-EMS until a long-term outcome is measured, the patient has received care from numerous units, each potentially influencing outcome [
41]. Unless performing risk adjustment and outcome measurement for each of these care intervals, it will problematic to use long term outcome measures as indicators of the isolated quality of P-EMS. Instead, quality indicators from the pre-hospital care interval should be developed for this purpose [
23]. The Institute of Medicine has stated that «quality of care is the degree to which health services for individuals and populations increase the likelihood of desired health outcomes and are consistent with current professional knowledge» [
42]. This definition of quality is a reminder that good quality is not identical to good outcomes. Despite excellent health care is provided, outcome for patients can be poor. Opposite, patients receiving poor quality health care can have good outcome.
Strengths and limitations
Using the professional network of the study group for recruitment of panel members, may have limitations: Colleagues that share our own professional interests may have been easier to identify and invite, than those with other views and mindsets. This practice can possibly lead to an imbalance in the composition of the expert panel. Although the expert panel reflected the inter-disciplinary nature of EMS, we recognize that we did not include a representative from an Emergency Medical Communication Central (EMCC). There was a trade-off between a manageable number of experts and the need for an inter-disciplinary composition of the expert panel. Consensus methodology literature describe an optimal group size of eight to twelve members [
43]. Our efforts in making the panel sufficiently inter-disciplinary resulted in a group size of 18 experts. However, due to a rigorous time schedule throughout the process, the slightly larger expert panel did not lead to any unnecessary delay.
Eight nations were represented in the expert panel; all from developed countries and the majority from Scandinavia. Therefore, we recognize that other areas may have other QIs which should be implemented locally. However, P-EMS as a service is usually only delivered in d eveloped countries. Hence, for these services the nationalities included should be representative.
In the consensus process we used a system of ranking and scoring to identify the QIs supported by the most experts in the panel. There are different methods to prioritize proposals and obtain consensus, and no method is considered clearly superior to the others [
44]. At the consensus meeting, any proposal was omitted if vigorously opposed by one or more of the participants.
The use of a Likert scale is another recognized method for defining the level of consensus. Likert scores are used for QI selection in several studies, including a recent Danish study selecting QIs for hospital-based emergency care [
45,
46]. Whether the use of a Likert scale would have improved our consensus process remains unclear. Moreover, it is methodological important to prevent that verbally skilled panel members dominate the consensus process. This issue also relates to “strong” personalities or experts enjoying a higher reputation than the other panel members [
19]. Therefore, proposals and rankings in stage 1 and 2 were anonymous.
The value of this study is the development of multidimensional quality indicators for P-EMS. This represents a starting point for future studies on measuring and improving quality of P-EMS. The necessary next step should be to test the feasibility and validity of the QIs in a sample of P-EMSs. Thus, a more final set of QIs for P-EMS can be developed.