The use of first responders to attend urgent medical incidents, including cardiac arrest, commenced in the mid-1980s in the US (Weaver et al, 1986) and in the late 1980s in Canada (Shuster and Keller, 1993). Firefighters, and other emergency personnel, were used as first responders to decrease the time from collapse to defibrillation (Weaver et al, 1986; Mosesso et al, 1998; Shuster and Keller, 1993), which is associated with an improvement in patient outcomes.
In Australia, the Australian Resuscitation Council defines a first responder as:
‘A person competent in advanced first aid (including oxygen administration and the use of an automated external defibrillator). A first responder may be a salaried/non-salaried member of a fire/police/ambulance service, a community based organization, or any other organization having in place the necessary systems’
First responders strengthen the ‘chain of survival’ concept described in the seminal paper by Cummins et al (1991) as they decrease the response time to the incident, which therefore decreases time to defibrillation. The use of first responders trained in basic life support (BLS) aims to decrease response times and instigate resuscitation procedures, including the use of automatic external defibrillators (AED), prior to the emergency medical service (EMS) arriving. This decrease in response time ultimately increases the chances of patient survival (Hawkins et al, 2007).
Since the inception of firefighter first responders, there has been a gradual increase in the types of first responders to include security guards in casinos and medical centres (Valenzuela et al, 2000; Anonymous, 2001), police (Ross et al, 2001; Kooij et al, 2004; Sayre et al, 2005), community groups such as Hatzolah, a Jewish volunteer first responder service (Chan et al, 2007), and volunteer organizations such as St John Ambulance (Wassertheil et al, 2000).
The primary aim of a first responder programme is to deliver a person capable of providing BLS skills and an AED to a patient in cardiac arrest in as short a time frame as possible and well before the arrival of EMS. First responder programmes (FRP) were the forerunner to public access defibrillation (PAD), where AEDs are placed at strategic locations in places where large numbers of people congregate or pass, e.g. railway stations (Weisfeldt et al, 1996). PAD programmes cannot replace the role of FRPs as locating AEDs in locations other than those that frequently cater to large numbers makes them a costly and inefficient proposition (Winkle, 2010).
There has been no previous study investigating the use of first responder programmes in Australia or their effect on survival from cardiac arrest in the out-of-hospital setting. Therefore, the objective of this study was to identify scientific studies of Australian first responder programmes and report the effect of FRP on survival rates from OHCA.
Methodology
Study design
This study is a review of the literature involving a search of specific medical electronic databases to identify studies of Australian first responder programmes.
Procedures
The Ovid Medline and EMBASE databases were searched from January 1980 until the end of May 2011. The following MeSH heading, emergency medical services, and keywords first responder, prehospital, out of hospital, and Australia were used in the search. The MeSH heading and keywords were used individually and in combination. Articles of any study design and those reporting on Australian out-of-hospital first responder programmes and their outcomes were included. Articles were excluded if they were not written in English or were hospital-based. The reference lists of the retrieved articles were also reviewed to determine if any articles had been missed in the initial electronic database search.
The selection of articles was conducted in a two stage process. First, all titles and abstracts were reviewed by two of the authors (SH and MB) to ascertain if they met the inclusion criteria. Second, where studies were deemed relevant, the full article was retrieved and underwent further review.
Any disagreements regarding the inclusion of articles were reconciled through consensus between the two authors (SH and MB). The third author, SB, was available to resolve any disagreement between the other authors but was not required to do so.
‘This study is the first to identify and review the scientific literature on first responder programmes in the out-of-hospital environment within Australia within Australia’
Results
The search located 674 articles with seven articles meeting the inclusion criteria. Two of these studies were subsequently excluded, leaving five studies available for review and analysis. There were no additional articles identified in the reference list review.
The first excluded study was one that describes the possible use of general practitioners in responding to major incidents as part of the medical disaster plan within Victoria, Australia and did not report on a specific programme or any patient outcomes (Somers et al, 1997). The second excluded study reported on the Metropolitan Fire and Emergency Services Board (MFESB) first responder programme. The study reported cardiac arrest numbers; however, it did not report on the survival outcomes (Boyle et al, 2010).
The quality of the analysed studies was predominantly low level. Four were retrospective case series studies (O'Rourke et al, 1997; Wassertheil et al, 2000; Smith et al, 2002; Chan et al, 2007) and one a prospective clinical trial (Smith et al, 2001). More information on the study details, first responder response times and cardiac arrest outcomes is reported in Table 1, Table 2 and Table 3 respectively.
| Author/s | Year published | Type of study | Data collection | Primary outcome measures | Study/participant demographics |
|---|---|---|---|---|---|
| Chan et al | 2007 | Retrospective case series | 1995-2005 | Incidence response, response time, cardiac arrest survival | All Hatzolah responses within East St Kilda, St Kilda, Elsternwick, Caulfield, Caulfield North/South, Balaclava, Elwood and Ripponlea |
| Smith et al | 2002 | Retrospective case series | February 2000-February 2001 | Incidence response survival rates, care provided | All EMR trained firefighters (1337, 91%) across 47 fire stations in MFESB service area |
| Smith et al | 2001 | Pilot study/ prospective control trial | 14 July 1998-22 June 1999 | Cardiac arrest response times /survival rates, cost | BLS trained fire-fighters from 7 fire stations covering an area of around 20% of the Melbourne population |
| Wassertheil et al | 2000 | Retrospective case series | December 1989 -January 1998 | Cardiac arrest survival, rates, time to CPR/ defibrillation | Cardiac arrest patients attended to by SJA at the Melbourne Cricket Ground and Shrine of Remembrance |
| O'Rourke et al | 1997 | Retrospective | 65 months for data collection-actual dates not specified. | Cardiac arrest | Cardiac arrest patients in QANTAS planes and major terminals. |
| Author/s | Year | First responder response time | Time until first defibrallation (if required) | Paramedic response time |
|---|---|---|---|---|
| Chan et al | 2007 | <3mins [median] | Not stated | Not stated |
| Smith et al | 2002 | 6.8 mins (95%CI 6.6 -7.0 mins)[mean] | 8.75 mins | 7.1 mins (95%CI 7.0 - 7.3 mins) [mean] |
| Smith et al | 2001 | 5.85 ± 1.66 mins [mean] | 8.62 mins | 7.61 ± 2.94 mins [mean] |
| Wassertheil et al | 2000 | Not stated | <5mins* | 11.14 ± 3.34 mins [mean] |
| O'Rourke et al | 1997 | Not stated | 38 seconds$ | Not stated |
Legend: $= from time AED opened; = no actual time reported,
>5 min considered delayed
| Authors | Year | Number of total responses | Number of cardiac arrests | BLS provided by FRP prior to ambulance | Cardiac arrest survival to hospital discharge |
|---|---|---|---|---|---|
| Chan et al | 2007 | Not stated | n=35 | n=29 (83%) | n=5 (26%) |
| Smith et al | 2002 | n=2942 | n=1,324 | n=226 (52.6%) | n=5 (4%) |
| Smith et al | 2001 | Not stated | n=161 | n=67 (42%) | n=6 (4%) |
| Wassertheil et al | 2000 | n=28 | n=35 | n=26 (93%) | n=20 (71.4%) |
| O'Rourke et al | 1997 | In flight n=81 |
In flight n=27 |
n=109 | In flight n=2 (7.4%) |
This review has highlighted that all but one of the reported programmes is situated in in the south-eastern Australian state of Victoria.
Discussion
This study is the first to identify and review the scientific literature on first responder programmes in the out-of-hospital environment within Australia. The development of the first responder took much longer in Australia than in many similar countries. The first documented first responder programme commenced in 1991 on international QANTAS aircraft and in QANTAS's major Australian terminals (O'Rourke et al, 1997). Other Australian first responder programmes (Chan et al, 2007) followed at least a decade after the US, and several years after QANTAS.
Within Victoria, ambulance response times have been reported as a 9.4 minute average (Bernard, 1998) and a 9.5 minute median (Boyle et al, 2010). A study in the UK found that a pulseless patient receiving no intervention for greater than 12 minutes will not survive; therefore an early response and intervention is imperative in improving patient outcome (O'Keeffe et al, 2010). The reported reluctance of bystanders to commence CPR, including mouth-to-mouth respiration, has often resulted in over half of witnessed collapses receiving no interventions prior to the arrival of the EMS (Spaite et al, 1990; Nolan et al, 2010).
Response times
Response times play a pivotal role in the success of first responder programmes. The three articles reporting on the Metropolitan Fire and Emergency Services Board (MFESB) first responder programme (FRP) all describe a considerable reduction in mean response times by fire fighters acting as first responders (Smith et al, 2001; Smith et al, 2002; Boyle et al, 2010). Smith et al (2001) reports in the MFESB pilot study a reduction in mean response times by 1.60 minutes in the pilot area when compared with the control region, however the study only reports combined EMS response times and not that of the initial first responders. Similarly, Smith et al (2002) notes a mean first responder time 1.1 minutes faster then the average Ambulance Victoria (AV) response time of 8 minutes. In both studies, time to defibrillation in cardiac arrests was reduced from 10.1 to 8.6 minutes and 8.75 minutes respectively (Smith et al, 2001; Smith et al, 2002). While this still exceeds the ideal goal of defibrillation within 6 minutes of collapse, it is a vast improvement on previous standards. Boyle et al (2010) also reports response times consistent with these studies.
Hatzolah and St John Ambulance (SJA) report impressive response times. Chan et al (2007) identifies a median cardiac arrest response time by Hatzolah of less than 2 minutes and a 90 percentile response time of 5-6 minutes to other medical incidents, resulting in their arrival on scene in vastly better response times than AV. These response times are most likely reflective of the close geographical proximity where Hatzolah members are located within the communities they serve (Chan et al, 2007).
Wassertheil et al (2000), while reporting impressive times until SJA defibrillation, was inconclusive as to whether these short times were due to lay responder, SJA first responder, or EMS interventions.
The QANTAS training dictates that once a passenger is identified as potentially in cardiac arrest, the time from identification, removal from the seat to the door bay, and subsequent defibrillation, if required, should be within six minutes. For QANTAS in-flight cardiac arrests, the average time to defibrillation for a shockable rhythm was 38 seconds from the time that the AED was opened (O'Rourke et al, 1997). There was no report of response time for cardiac arrests in major QANTAS Australian terminals. In QANTAS's major Australian terminals there were 19 cardiac arrests with 17 (89%) being in a shockable rhythm (O'Rourke et al, 1997).
‘Survival rates are indicative of the effectiveness of a first responder model, as increased survival rates correlate linearly with early intervention’
Cardiac arrest survival rates
Cardiac arrest survival rates are available for 4 of the 5 studies reviewed. Survival rates are indicative of the effectiveness of a first responder model, as increased survival rates correlate linearly with early intervention (Valenzuela et al, 2005). Smith et al (2001) recognized that while there was no significant difference in survival until discharge, the survival rate of witnessed shockable rhythms was 23% and 36% in the control and pilot areas respectively. The likely explanation for this difference is that the decreased response time allowed early defibrillation, in all cases prior to the arrival of paramedics. However, when looking at all cardiac arrests, the survival rates are only marginally better than the EMS rates. Smith et al (2002) reports appreciably greater survival rates, with 7.2% of cardiac arrests patients on whom resuscitation was started by the MFESB FRP surviving until hospital discharge.
Although Boyle et al (2010) provides no cardiac arrest outcome data, the findings that firefighters provided 54.3% of initial care for a mean time of 4.62 minutes prior to paramedic arrival illustrates that earlier access to BLS and defibrillation improves cardiac arrest survivability.
Both Chan et al (2007) and Wassertheil et al (2000) also report improved survival rates following cardiac arrest well above the expected norm for OHCA, with a hospital discharge survival rate of 26% and 71.4% respectively. For QANTAS in-flight cardiac arrests, only six of 27 were in a shockable rhythm, with two of the six (33%) surviving to hospital discharge (O'Rourke et al, 1997). There was no reporting of cardiac arrest survival rates that occurred in QANTAS's major Australian terminals (O'Rourke et al, 1997).
Previously, the major focus of measurable outcomes for FRPs in Australia has been cardiac arrest. However, the expansion of the FRPs to include responses to other medical and non-medical incidents is indicative of the beneficial role that FRPs can play in supporting EMS in responding to out-of-hospital incidents. Three of the programmes in this review reported on first responders attending to other medical and nonmedical incidents (Wassertheil et al, 2000; Smith et al, 2001; Chan et al, 2007).
The additional incidents responded to by the first responders in the three articles include: falls, chest pain, respiratory distress, drug overdose, seizures, road traffic accidents, a collapse of unknown origin, traumatic incidents including burns and lacerations, drowning and near drowning, and electrocution. Culturally, the Hatzolah programme has provided improved community education regarding medical care that has resulted in greater community access to the ambulance service (Chan et al, 2007). Similarly, SJA's continued contribution in community education and public health are clearly demonstrated (Wassertheil et al, 2000).
We are aware of other first responder programmes, especially within Victoria: the Community Emergency Response Team (CERT) model has been in place for several years, but to date its effectiveness has not been reported in the scientific literature. Further research is needed to report of the efficiency and effectiveness of other first responder programmes within Victorian and Australia generally.
This study is potentially limited by the inability to accurately compare each study, as there is inconsistent reporting of the results. The restriction of English-only articles is unlikely to have missed any relevant articles, but that possibility remains.
Conclusion
This study suggests that first responder programmes within Australia demonstrate better response times to OHCA than that of the ambulance service, and that the survival to hospital discharge is better for those patients attended by first responders than those that are attended by the ambulance service alone. The results of this study also suggest the survival from OHCA attended by first responders in Australia is, in some studies, comparable with other similar international studies.