Transcription

1 Volume 14 Issue 2 May

2 Table of Contents EDITORIAL Moving closer to paramedic ACS rule-out in the field Marc Colbec 3 Pre-hospital healthcare in developing countries requires a holistic approach to ensure services meet the diverse needs Jaime Wallis 6 REVIEW The pre-hospital and healthcare system in Malang, Indonesia Suryanto, Malcolm Boyle, Virginia Plummer 8 RESEARCH Accuracy of stroke identification by paramedics in a metropolitan pre-hospital setting: a cohort study Teresa A Williams, David Blacker, Glenn Arendts, Emily Patrick, Deon Brink, Judith Finn 17 Evaluating the knowledge base and current training of paramedics in the southwest United States inassessing and managing toxic alcohol exposures Christopher Bare, Oliver Grundmann 27 Evaluation of the health and physical activity characteristics of undergraduate paramedic and nursing students Peter S Micalos, Alexander J MacQuarrie, Brian A Haskins, Eileen Barry, Judith K Anderson 38 2

3 Volume 14 Issue 2 Editorial Pre-hospital healthcare in developing countries requires a holistic approach to ensure services meet the diverse needs Jaime Wallis Griffith University, Queensland 1

4 Wallis: Pre-hospital care systems in developing countries Editorial Pre-hospital healthcare in developing countries requires a holistic approach to ensure services meet the diverse needs Jaime Wallis BEH, GradCertHlthProfEduc, is a lecturer at Griffith University, Gold Coast, Queensland When you think about travelling in Asia or throughout the Pacific, do you consider the pre-hospital care provided in these countries? Irrespective of your answer, do you believe in health equality for all and more specifically, a reliable and efficient pre-hospital care service? Unfortunately, almost all low to middle-income countries in Asia and the Pacific face significant challenges when it comes to providing pre-hospital care. One of the first questions that come to mind when looking into the pre-hospital care system in developing countries is: What is the educational system supporting it? More specifically, what level of training do the pre-hospital care providers have and is the education specifically tailored to the conditions found in their out-of-hospital environment? Traditionally in these countries, nurses are assigned to emergency vehicles with a driver and are asked to perform patient assessment and management with little training, if any, for the pre-hospital setting. Other important questions are what funding is available and how are the services sustainable? This translates into staffing, training of staff, availability of pre-hospital equipment (including vehicle and contents), available pharmacological interventions and practice guidelines that support pre-hospital staff. At the foundational level it is also necessary to note what systems are in place to coordinate and manage such services. This is fundamental. Alas, most of these countries don t have a central point of contact for calling an ambulance, which is seemingly disproportionate to the communication network availability, hence the dispatch system or constant communication systems can be lacking. Also of importance is an understanding and respect towards the different cultural and religious sensitivities that exist in developing countries. For instance, in some countries the societal culture has an understanding or a belief that incidents that occur in the emergency pre-hospital setting are at God s will that they are acts of fate. This can have a significant impact on the level of importance the public perceive the ambulance service to have. It can also determine the cost at which the pubic are willing to outlay for an ambulance service. These points need careful consideration when it comes to the different geographic locations and settings terrain, urban, rural or remote when reflecting on the provision of pre-hospital care in developing countries. I strongly recommend you take the time to read the seminal work by Suryanto and colleagues in this issue of the Australasian Journal of Paramedics (1). Their research encompasses a holistic analysis and well developed approach to the pre-hospital care progression in the metropolis of Malang in East Java, Indonesia, and provides insight into the issues of healthcare provision, including pre-hospital care, in a developing country. Jaime Wallis Griffith University, Queensland Reference 1. Suryanto, Boyle M, Plummer V. The pre-hospital and healthcare system in Malang, Indonesia. Australasian Journal of Paramedicine 2017;14(2). 01

5 Volume 14 Issue 2 Article 1 The pre-hospital and healthcare system in Malang, Indonesia Suryanto Brawijaya University, Indonesia Malcolm Boyle Griffith University, Queensland, Australia Virginia Plummer Monash University, Victoria, Australia 1

6 Suryanto: The pre-hospital and healthcare system in Malang, Indonesia Review The pre-hospital and healthcare system in Malang, Indonesia Suryanto PhD candidate and Lecturer, School of Nursing 1,2 ; Malcolm Boyle PhD, Academic Lead in Paramedic Education 3 ; Virginia Plummer RN, PhD is Associate Professor of Nursing Research 1,4 Affiliations: 1 Monash University, Victoria, Australia 2 Brawijaya University, Indonesia 3 Griffith University, School of Medicine, Queensland, Australia 4 Peninsula Health, Victoria, Australia Abstract Introduction Malang is the second largest city in East Java and consists of three districts: Malang Regency, Malang City and Batu City. Malang has the most advanced emergency care education in East Java, and possibly Indonesia. The recent launch of the 119 Emergency Medical Service may influence pre-hospital care in Malang and assist in developing a more organised emergency medical system in the area. Objective This paper describes the pre-hospital care and healthcare structure in Malang and focuses on the health facilities, pre-hospital care service and healthcare providers. Discussion There are two major types of healthcare facilities in Malang, hospitals and primary healthcare centres called puskesmas. Ambulance service provision in Malang is either hospital-based or puskesmas-based. There is no organised emergency medical system in Malang. In the case of an emergency, the patient or bystander calls the hospital emergency department or puskesmas. The use of ambulances in Malang is low due to the perceived prohibitive cost of the service. Ambulance service fees for emergency cases outside of the hospital or puskesmas are not covered by the universal health insurance in the country. There is no specific pre-hospital trained staff in Malang. Nurses are responsible for staffing the ambulance, and there is no formal prehospital care education for those nurses working in an ambulance. Keywords: healthcare; health system; prehospital; ambulance; nurse; developing country; Indonesia Corresponding author: Suryanto, suryanto.s@monash.edu; suryanto.s@ub.ac.id 01

7 Suryanto: The pre-hospital and healthcare system in Malang, Indonesia Introduction Malang is the second largest city in East Java Province, the second most populated province in Indonesia after West Java Province (1). East Java Province covers a 47, km2 wide area consisting of more than 60 islands and a population of 41,437,769 people in 2014 spread over 38 districts: 29 regencies and nine cities. These districts are further divided into 662 sub-districts which consist of 782 urban villages and 7,741 rural villages (2,3). The majority of people in the province are from the Java tribe and are Muslim (3). Even though the population of East Java Province is the second largest after West Java province, the number of healthcare providers in East Java Province is the greatest at 108,220 health workers including 31,099 non-health workers (4). Those healthcare providers are distributed in two major health facilities, the hospitals and the primary healthcare centres called puskesmas. The Malang area, commonly called Malang Raya (The Greater Malang) is one of the largest areas in East Java Province and consists of three districts: Malang Regency, Malang City and Batu City. Batu City was previously part of Malang Regency, as one of the sub-districts, however, since 2001, based on the Republic of Indonesia Law No. 11 of 2001, it has become an independent district under the East Java Province (5). Malang City and Batu City cover less than 10% of the Greater Malang, which is mostly made up of Malang Regency. Malang City and Batu City have a more rapidly developing economy than Malang Regency, thus people tend to live in the urban areas for lifestyle and employment reasons. The economic development in the area also influences the healthcare services where more hospitals with specialty services are available in Malang City (6-8), hence, healthcare workers in the Malang City area account for a higher number compared to the other two areas (9). Previously (around the year 2000), there was an ambulance system known as 118 Emergency Ambulance Service, which was functional in 18 Indonesian cities including Jakarta, Palembang, Yogyakarta, Surabaya, Makassar, Denpasar and Malang (10). The system was terminated within 5 years due to a lack of funding support from the government (10). A breakthrough followed 10 years later in the Indonesian prehospital care system with the launch of the 119 Emergency Medical Service by the Ministry of Health of Indonesia in July 2016 (11). The launch of the new pre-hospital care system may enhance the pre-hospital care services in the area as Malang has the most advanced emergency care education in East Java (and possibly Indonesia) and an emergency medicine specialist medical program and Master of Nursing program majoring in emergency nursing (12). Methods The information and data in this paper are mostly obtained from the grey literature including the website of the Indonesian Ministry of Health both at a national level and provincial level. Google Scholar, MEDLINE, EMBASE and CINAHL were used to obtain the latest research on pre-hospital care in Malang and the development of pre-hospital care in other countries. Discussion Health facilities There are 315 hospitals in East Java Province consisting of 233 general hospitals and 82 specialty hospitals (13). General and specialty hospitals are classified in Indonesia from Class A to Class D, with Class A the highest hospital class and Class D the lowest. This classification is based on the health services provided by the hospital (14). Most of the hospitals in East Java Province are Class C hospitals (116 hospitals), 55 hospitals are Class D hospitals, 41 are Class B hospitals, and five are Class A hospitals, with 98 hospitals not yet certified by the Ministry of Health (13). The geographic distribution in Malang has an effect on the health facilities in the area. As for other areas in Indonesia, there are two major types of healthcare facilities in Malang, the hospitals and puskesmas. Based on urban development, compared to the other two districts, there are more hospitals in Malang City. There are 25 hospitals in Malang City, 17 hospitals in Malang Regency and three hospitals in Batu City (6-8). On the other hand, due to the large less-urban areas in Malang Regency, it has more puskesmas available. There are 39 puskesmas in Malang Regency, 15 puskesmas in Malang City, and five puskesmas in Batu City (15). Batu City was the area with the least number of healthcare facilities due to the city previously being a sub-district of Malang Regency. Furthermore, there were more Class C hospitals and specialty hospitals in Malang City, plus a quantity of general hospitals. Crossing the boundaries to seek healthcare occurs with most of the people in Malang Regency and Batu City tending to go to Malang City hospitals for specialty services. Although there were more puskesmas in Malang Regency, the ratio of puskesmas to the population in Malang Regency was the lowest, 1.4 per 100,000 people. Surprisingly, Batu City, with the least number of puskesmas has the highest ratio, 2.5 per 100,000 people. Malang City, with the highest number of hospitals, has the highest ratio of hospitals to population, 2.9 per 100,000 people. The detail of ratio of healthcare facilities in the three districts is shown in Table 1. 02

8 Suryanto: The pre-hospital and healthcare system in Malang, Indonesia Table 1. Ratio of healthcare facilities in Malang (9) Characteristics Malang Regency* Malang City* Hospitals Puskesmas *per 100,000 people Batu City* The ambulance service in Indonesia is either puskesmasbased or hospital-based. Not all emergency departments (ED) of a puskesmas or hospital can provide emergency care, which includes an ambulance service. Based on the Indonesian Ministry of Health Law No. 856 of 2009, there are four levels of ED in Indonesia. Level I ED is the lowest ED level and Level IV ED is the highest level (16). The Level IV ED includes the availability of a ventilator, high care units, resuscitation ward and intensive care unit; the availability of an on-call service for all sub-specialties and on-site for four major specialist doctors (surgeon, obstetrician, paediatrician and internist). On the other hand, a Level I ED provides only stabilising and evacuating of patients with no specialist doctor services available (16). Level IV ED is the minimum standard for a Class A hospital, Level III ED is the minimum standard for a Class B hospital, Level II ED is the minimum standard for a Class C hospital, and Level I ED is the minimum standard for a Class D hospital (17). More than 80% of hospitals in East Java, including 223 (89.4%) general hospitals and 98 (86.7%) specialty hospitals in East Java Province are able to provide Level I emergency care (Table 2) (17). Importantly, 469 (91.4%) of the puskesmas are also able to provide Level I emergency care (15). There is no published data about the number of hospitals and puskesmas with Level II to IV emergency care in East Java Province. The annual report on health of East Java Province shows a distribution of health facilities in Malang compared to those elsewhere in East Java (15). All general hospitals and inpatient puskesmas both in Malang Regency and Malang City are able to provide emergency care services, but one out of three hospitals and all puskesmas in Batu City are unable to provide these services (15). The majority of specialty hospitals cannot provide emergency care services in those three districts, only two out of 22 specialty hospitals are able to provide the emergency services (Table 3). Pre-hospital care setting Currently, there is no organised emergency medical system in Malang. The previous emergency medical system for the Malang area was the implementation of the 118 service that was operational in around Malang was one of the seven cities that implemented the 118 service, others included Jakarta, Palembang, Yogyakarta, Surabaya, Makassar and Denpasar (10). The ambulance service in Malang is either hospital-based or puskesmas-based. In the case of an emergency, the patient or bystander would call the hospital ED or puskesmas where they want to be admitted. Then, depending on the availability of an ambulance, the patient may be picked-up by an ambulance, but in most cases the patient tends to use a private vehicle instead of using an ambulance. Table 2. Healthcare facilities with Level I emergency care in East Java Province (15,17) Health facilities Quantity Availability of Level I ED n % General hospitals Specialty hospitals Puskesmas with in-patient facility Other health facilities Table 3. Health facilities in Malang providing emergency care (15) Districts n General hospitals Specialty hospitals Puskesmas with in-patient facility Providing emergency care n Providing emergency care n Providing emergency care Malang Regency (100%) 6 1 (16.7%) (100%) Malang City 9 9 (100%) 14 0 (0%) 4 4 (100%) Batu City 3 2 (66.7%) 2 1 (50%) 3 0 (0%) 03

9 Suryanto: The pre-hospital and healthcare system in Malang, Indonesia Compared to other lower-middle income countries, Indonesia is not the only country that does not have an organised EMS system. Other countries include India, Morocco, Vietnam, Armenia, Nigeria, Ghana and Sri Lanka (18). The establishment of the Tulungagung Emergency Medical Service (TEMS) initiated the launch of the current 119 Emergency Medical Service in Indonesia. The 119 service has been implemented in 27 locations in the country, mostly on Java Island (11). Similarly, India initiated the establishment of an EMS system in 2002 (19) with the system then adopted in the nine states of India in 2011 (20). In Nigeria, initially, there were seven ambulance stations during the piloting of an EMS system in the country in 2004 and within 10 years, the system was available in every district in Nigeria (21). There is no published data about the coverage target of the 119 service in Indonesia. The launch of the 119 service was in conjunction with the establishment of the National Command Centre (NCC) in Jakarta and Public Safety Centres (PSCs) in the 27 locations involved in the system (11). The 27 locations involved in the pilot project were chosen because the areas have national government hospitals (22). Based on the Indonesian government regulation, PSCs must be available gradually in every district/city in Indonesia and function as the centre of emergency care services in the area (22). Currently, the PSCs utilise a local phone number and it is projected that the PSCs will have an integrated service with the police and fire department in every district (23). With the current EMS system, emergency calls are received by the NCC in Jakarta and then transferred to the PSC in the district closest to the incident. The PSC then deploy an ambulance from either a puskesmas or hospital to the scene. The system does not only manage the ambulance services but also provides the information regarding the availability of emergency units and in-patient rooms in the hospitals. Several challenges in implementing the system include the lack of infrastructure and resources (24), the delay in responding to the calls due to a lack of staff available in the NCC, and the delay in transferring calls from the NCC to PSC (11). The TEMS was established in November 2015 in Tulungagung, East Java Province and integrates the ambulance service with police, army, fire and the disaster management department (25). With the use of the Global Positioning System ( GPS ), the TEMS coordinates the ambulances both from the puskesmas and hospital EDs in Tulungagung (27). The TEMS has become a blueprint model for the 119 service in Indonesia (28). However, due to a lack of information provided to the community, especially the call centre number, the TEMS service is under-utilised (27). Therefore, enhancing public awareness of pre-hospital care is one of the essential keys in developing the pre-hospital system. As the largest hospital in the Malang area, Saiful Anwar Hospital, which is located in the Malang City area, received emergency ambulance calls per month in 2011 (29). With the population of more than 800,000 people in Malang City (30), the numbers show the low utilisation rate of ambulance services in the area. Only 11% (3,160) of patients were transported by ambulance to the Saiful Anwar Hospital ED, 89% (9,758) of pre-hospital patients were brought in by other vehicles (31). As Saiful Anwar Hospital is the largest hospital in the area, the other hospitals may have lower numbers of ambulance utilisation. The implementation of the universal health insurance in Indonesian called Badan Penyelenggara Jaminan Sosial (BPJS), which commenced in January 2014, may also influence the use of an ambulance by the people. By the end of 2015, more than half of the Indonesian population of 157 million people were covered by the new universal health insurance (32). The BPJS covers the ambulance service. The premium fee paid each month includes the ambulance service but this covers services for referral purposes only and not emergency pre-hospital care services. The ambulance service covered includes a condition that both originating and destination health institutions must have a funding agreement with the BPJS (33). Currently, there are 19,969 health institutions that have a funding agreement with the BPJS as primary health facilities, including the puskesmas, general practitioner clinics and medical clinics. There are 1,847 health institutions classified as advanced health facilities including both government and private hospitals (32). Unfortunately, in the case of an emergency outside the hospital or puskesmas, such as traffic accident or heart attack, or any referral between health institutions which have no funding agreement with the BPJS, the ambulance service cost is not covered with the monthly premium fee (33) and hence the patient must pay for ambulance services or use other transport. Traffic accidents have become one of the major pre-hospital concerns in Malang. There were 623 traffic accident cases in Malang Regency in 2012, which resulted in 140 deaths and 796 minor injuries (34). There were 13,383 out of 26,907 (49.7%) trauma cases managed by the Saiful Anwar Hospital ED in 2007, with 72.9% (9,758) being male patients and a 2.8% mortality rate (31). Motorcycle accidents accounted for the highest incidence of traffic accidents in Malang. In 2012, there were 771 motorcycle injuries in the city (34). Unfortunately, most of motorcycle accident patients admitted to the Saiful Anwar Hospital ED with open fractures did not have a driver s licence. Those who did have a licence mostly had never undergone a driving licence test (35) as they had obtained a licence illegally. The most recent traffic accident study in Malang found accidents involving 263 motorcyclists showed that human factors had a significant contribution to a traffic accident. The main causes of the accident included carelessness, drowsiness, drunkenness, fatigue, unskilled rider and indiscipline and speeding (34). 04

10 Suryanto: The pre-hospital and healthcare system in Malang, Indonesia The study also showed that vehicle factors such as brake or headlamp malfunction had no significant contribution to the accidents, but environmental factors including road cornering and rainfall did have a significant contribution (34). Interestingly, road conditions, including slippery or damaged roads, did not significantly contribute to accidents, which were more likely caused by human factors. Human resources Even though the population of East Java Province is the second largest in Indonesia (1), the number of healthcare providers in East Java Province is the greatest at 108,220 health workers, plus 31,099 allied health workers (4). As Malang City has the largest number of advanced health facilities, the number of healthcare providers in Malang City was also the highest compared to the other two districts. In 2014 there were 6,179 healthcare providers in Malang City, 4,340 in Malang Regency, and 910 in Batu City (9). The details of healthcare providers in those three districts can be seen in Table 4. Table 4. Healthcare providers in Malang (9,36) Healthcare providers Malang Regency Malang City Batu City East Java Province Indonesia n Ratio* n Ratio* n Ratio* n Ratio* n Ratio* Specialist doctors , , General practitioners , , Dentists , , Nurses 1, , , , Midwives , , Pharmacists , , Other healthcare Providers , , Total 2,948 N/A 4,035 N/A 626 N/A 77,121 N/A 646,116 N/A *per 100,000 people Based on Table 4, Malang City has the highest number of healthcare providers. Compared with the population of the three districts, Malang City had the highest ratio of healthcare providers among the districts, except for the ratio of general practitioners (GPs) to population. Batu City has the highest ratio of GPs to population, 32 per 100,000 people. Even though Malang Regency has the highest number of GPs and midwives, the ratio of both healthcare providers in the district was the lowest compared with the other two districts. Similar to other areas in Indonesia, there is no specific pre-hospital-trained staff in Malang. Nurses are responsible for staffing the ambulance for both hospital-based and puskesmas-based ambulances. However, there is no formal pre-hospital care education for those nurses working in an ambulance. As the largest hospital in Malang, Saiful Anwar Hospital provides some pre-hospital education for pre-hospital providers including advanced trauma life support, basic trauma life support and ambulance protocol. A study showed that the education had a significant contribution to the differences of pre-hospital treatment, ambulance transportation and communication between field attendance and the trauma centre in Malang (37). The study used a pre-post test design involving both nurses and medical doctors working in the pre-hospital care system in Malang and was conducted over two periods, April to July 1999 (n=1,839) and April to July 2003 (n=1,408). The study revealed that pre-hospital education had a positive outcome for patients where there were significant differences in pre-hospital treatment for patients before and after the study such as intravenous therapy, splinting, medication and wound healing. As not all nurses working in the ambulance had an education in pre-hospital care management, the latest qualitative study involving six puskesmas-based nurses in Malang demonstrates that nurses had a feeling of powerlessness, and had an emotional response to the process of change when caring for victims of traffic accidents (38). Those feelings indicated a lack of knowledge about treating traffic accident patients. Participants also expressed that they felt worried about endangering the patient due to a lack of knowledge about prehospital trauma management. The worry was compounded by a lack of facilities in the puskesmas, a lack of authority and a fear of a lawsuit. Prasetya et.al investigated the nurses experience in prehospital trauma care involving six hospital-based ambulance nurses in Malang found that there were several obstacles that the nurses faced while working on the ambulance. These included the people s culture, environmental safety issues, difficulties in patient fetching and interprofessional collaboration issues (25). The study highlighted a need for special education in pre-hospital care for nurses working in ambulances and the need of an organised EMS in the area. 05

11 Suryanto: The pre-hospital and healthcare system in Malang, Indonesia In order to enhance the competency of ambulance nurses in Malang, two projects funded by the Australian Department of Foreign Affairs and Trade through the Australia Indonesia Institute have been implemented (24). The projects involved staff from Monash University, Griffith University and Brawijaya University. The sharing of knowledge and experience by Australian university staff about aspects of pre-hospital care has been received by Brawijaya University staff during the pilot project of the Ambulance Nurse Course involving ambulance nurses from both the puskesmas and hospital EDs in Malang and will also be used as part of the nurse specialist program (Emergency Nursing) at Brawijaya University (24). With great anticipation, the course is expected to be delivered by Brawijaya University with a broader range of participants in the future. The signing of a Memorandum of Understanding (MoU) between Brawijaya University and Monash University was undertaken in December The MoU will maintain the collaborative work between the two universities in pre-hospital education and other health-related education and research. There is an opportunity to enhance the collaboration with the Griffith University for further development of pre-hospital care education in Indonesia. Even though there is no formal pre-hospital care education for healthcare providers working in an ambulance in Indonesia, emergency care education is developing in the country and Malang is the only city in Indonesia that has advanced education in emergency care for both doctors and nurses. Malang has an emergency medicine specialist program, the only one in Indonesia, and has a Master of Nursing program with a major in emergency nursing, the first of two such programs in Indonesia. The two emergency care education programs were implemented by the Brawijaya University, one of the largest universities in Malang, in conjunction with the Saiful Anwar Hospital. The programs were established with assistance from the Singapore International Foundation which provided voluntary emergency doctors and nurses to improve the emergency care in Malang (39), including the setting up of an EMS system in Malang (40). The assistance program initially focused on traumatology, which was an 8 year assistance program starting in 1996 and ceasing in Following 2004, it was revised to a general emergency program which included extensive informal training for medical and nursing staff both in-hospital and for pre-hospital emergency services (39). Both doctors and nurses from Saiful Anwar Hospital and Brawijaya University were involved in the training. The initial management of emergency healthcare education in Malang was assisted by the Singapore International Foundation and the Rotary Club of Malang (12). As the primary focus of the program was traumatology, the committee established the Malang Trauma Service Centre. The committee implemented several preparation courses for traumatology and emergency care including the Basic Cardiac Life Support, Basic Trauma Life Support, ECG and Resuscitation, Triage Officer, Ambulance Protocol, and Trauma Nursing Care. These courses were supported by the SIF who provided the educational staff (12). To enhance the program, doctors and nurses from Malang were sent to Singapore for clinical experience at the Singapore General Hospital, National University Hospital, Tan Tockseng Hospital and Kandang Kerbau Hospital in October 1999 (12). As a result, the Emergency Medicine Specialist Program was established in The Master of Nursing program majoring in emergency care was established in Using the results of the project supported by the Australian government as a blueprint of prehospital care education in Indonesia, the specialist program of Emergency Nursing at Brawijaya University is now under preparation. Conclusion Malang, one of the most populated cities in Indonesia has a developed emergency and trauma care service resulting from significant development of an Emergency Medicine Specialist program and Master of Nursing majoring in emergency care program. However, the 119 Emergency Medical Service has not yet been implemented in Malang, thereby continuing the limited pre-hospital care services for the people. The national health insurance scheme does not cover the pre-hospital ambulance service for urgent and acute incidents therefore it is not surprising that a majority of people use private vehicles for transporting patients to hospital. Even though there is no formal pre-hospital care education for ambulance staff, it is believed that the project supported by the Australian government may initiate the establishment of pre-hospital education for nurses working in ambulances in Malang. Conflict of interest The authors declare they have no competing interests. Each author of this paper has completed the ICMJE conflict of interest statement. Malcolm Boyle is Editor of the Australasian Journal of Paramedicine. References 1. Kementerian Kesehatan Republik Indonesia. Profil Kesehatan Indonesia Available at: go.id/resources/download/pusdatin/profil-kesehatanindonesia/profil-kesehatan-indonesia-2012.pdf [Accessed 9 April 2014 April]. 2. Kementerian Dalam Negeri Republik Indonesia. Peraturan Menteri Dalam Negeri Republik Indonesia Nomor Available at: [Accessed 23 April 2014]. 06

12 Suryanto: The pre-hospital and healthcare system in Malang, Indonesia References (continued) 3. Kementerian Dalam Negeri Republik Indonesia. Profil Daerah: Provinsi Jawa Timur Available at: www. kemendagri.go.id/pages/profil-daerah/provinsi/detail/35/ jawa-timur [Accessed 23 April 2014]. 4. Kementerian Kesehatan Republik Indonesia. Data SDM Kesehatan yang Didayagunakan di Fasilitas Pelayanan Kesehatan (Fasyankes) Available at: depkes.go.id/sdmk/rekap.php [Accessed 28 April 2014]. 5. Republik Indonesia. Undang Undang Republik Indonesia Nomor Available at: bpkp [Accessed 26 April 2014]. 6. Kementerian Kesehatan Republik Indonesia. Data Rumah Sakit Online: Kota Batu Available at: ta+batu&simplesrchfieldscomboopt=kab%2fkota&simp lesrchtypecombonot=&simplesrchtypecomboopt=equals &a=integrated&id=1&criteria=and [Accessed 26 April 2014]. 7. Kementerian Kesehatan Republik Indonesia. Data Rumah Sakit Online: Kota Malang Available at: ota+malang&simplesrchfieldscomboopt=kab%2fkota& simplesrchtypecombonot=&simplesrchtypecomboopt= Equals&a=integrated&id=1&criteria=and [Accessed 26 April 2014]. 8. Kementerian Kesehatan Republik Indonesia. Data Rumah Sakit Online: Kabupaten Malang Available at: lang&simplesrchfieldscomboopt=kab%2fkota&simples rchtypecombonot=&simplesrchtypecomboopt=equals&a =integrated&id=1&criteria=and [Accessed 26 April 2014]. 9. Kementerian Kesehatan Republik Indonesia. Rekapitulasi SDM Kesehatan per Kabupaten: Provinsi Jawa Timur Available at: php?prov=35 [Accessed 28 April 2014]. 10. Pitt E, Pusponegoro A. Prehospital care in Indonesia. Emerg Med J 2005;22: Dzulfiqar F, Riza B. Health Ministry Launched Emergency Service Available at: news/2016/07/01/ /health-ministry-launches- Emergency-Service-119 [Accessed 4 November 2016]. 12. Fakultas Kedokteran Universitas Brawijaya. Ilmu Kedokteran Emergensi: Sejarah Available at: ac.id/em/profil/sejarah/ [Accessed 28 April 2014]. 13. Kementerian Kesehatan Republik Indonesia. Data Rumah Sakit Online: Propinsi Jawa Timur Available at: php?alamat_prop=jawa+timur&submit=find [Accessed 5 December 2016]. 14. Republik Indonesia. Undang-Undang Republik Indonesia Nomor Available from: produk-hukum/2009/10/28/undang-undang-no-44- tahun-2009 [Accessed 9 April 2014]. 15. Dinas Kesehatan Provinsi Jawa Timur. Profil Kesehatan Provinsi Jawa Timur Tahun Available at: www. depkes.go.id/resources/download/profil/profil_kes_ PROVINSI_2012/15_Profil_Kes.Prov.JawaTimur_2012.pdf [Accessed 28 April 2014]. 16. Kementerian Kesehatan Republik Indonesia. Kepmenkes Nomor Available at: go.id/v2010/dokumen/2014/sdk/mibangkes/perundangan/ Bina%20Upaya%20Kes/2009/KMK%20No.%20856%20 ttg%20standar%20igd%20di%20rs.pdf [Accessed 29 April 2014]. 17. Dinas Kesehatan Provinsi Jawa Timur. Profil Kesehatan Provinsi Jawa Timur Available at: dinkes.jatimprov.go.id/userfile/dokumen/profil%20 KESEHATAN%20PROV%20JATIM%202014(1).pdf [Accessed 4 November 2016]. 18. Suryanto, Plummer V, Boyle M. EMS Systems in Lower- Middle Income Countries: A Literature Review. Prehosp Disaster Med 2017;32: Patrick RW. International EMS. An EMS system for India. Emerg Med Serv 2002;31: Jena BN, Umar NS. Emergency medical services (EMS) in India - effectiveness in managing morbidities resulting from accidents, injuries, poisoning, fracture and burns. Indian Emerg J 2011;6: Frimpong J, Dinye RD. Ambulance emergency services and healthcare provision in ghana: a district level performance appraisal. Int J Health Med Inform 2014;3: Biro Komunikasi dan Pelayanan Masyarakat KemKes RI. Layanan 119, Terobosan Baru Layanan Kegawatdaruratan Medik di Indonesia Available at: go.id/article/view/ /layanan-119-terobosanbaru-layanan-kegawatdaruratan-medik-di-indonesia.html [Accessed 2 February 2017]. 23. Biro Komunikasi dan Pelayanan Masyarakat KemKes RI. Public Safety Center (PSC) 119 Siap Digunakan Available at: public-safety-center-psc-119-siap-digunakan.html [Accessed 2 February 2017]. 24. Boyle M, Wallis J, Suryanto. Time to improve pre-hospital care in developing countries. Australasian Journal of Paramedicine 2016;13: Prastya A, Drajat RS, Haedar A, Setijowati N. Hubungan Moda Transportasi dengan Waktu Tanggap/Response Time pada Pasien Henti Jantung di Luar Rumah Sakit yang Dirujuk ke IGD RSUD Dr Iskak Tulungagung. Medica Majapahit 2016;8: Prastya A, Drajat RS, Haedar A, Setijowati N. Hubungan Moda Transportasi dengan Waktu Tanggap/Response Time pada Pasien Henti Jantung di Luar Rumah Sakit yang Dirujuk ke IGD Dr. Iskak Tulungagung. Medica Majapahit 2016;8(2). 27. Sujarwoko DH. Layanan TEMS RSUD Tulungagung Belum Efektif Available at: berita/172843/layanan-tems-rsud-tulungagung-belum-efektif [Accessed 29 November 2016]. 07

13 Suryanto: The pre-hospital and healthcare system in Malang, Indonesia References (continued) 28. Arif S. 911 Tulungagung Jadi Percontohan Nasional Available at: php?r=5&n=14&date= [Accessed 29 November 2016]. 29. Prasetya E, Winarni I, Haedar A. Studi Fenomenologi: Pengalaman Perawat di Lingkungan Tempat Kejadian Trauma Akut di Pre-Hospital Kota Malang. Jurnal Medika Respati 2015;X: Kementerian Dalam Negeri Republik Indonesia. Profil Daerah: Kota Malang Available at: go.id/pages/profil-daerah/kabupaten/id/35/name/jawa-timur/ detail/3573/kota-malang [Accessed 23 April 2014]. 31. Haedar A. Pre-Hospital System in Indonesia Available at: presentations/ _bangkok/pre-hospital%20 System%20in%20Malang%20-%20PAROS%20(ALI%20 HAEDAR%20-%20ACEM).pdf [Accessed 8 November 2015]. 32. Badan Penelitian dan Pengembangan Kesehatan. Laporan Pengelolaan Program Tahun 2015 dan Laporan Keuangan tahun 2015/2016. Available at: bpjs/index.php/unduh/index/572 [Accessed 30 November 2016]. 33. Badan Penelitian dan Pengembangan Kesehatan. Peraturan Badan Penyelenggara Jaminan Sosial Kesehatan Nomor 1 Tahun Available at: bpjs/index.php/unduh/index/10 [Accessed 30 November 2016]. 34. Marsaid, Hidayat M, Ahsan. Faktor yang Berhubungan dengan Kejadian Kecelakaan Lalu Lintas pada Pengendara Sepeda Motor di Wilayah Polres Kabupaten Malang. Jurnal Ilmu Keperawatan 2013;1: Maharani T, Haedar A. The Influence of status and the patterns of driving license ownership toward the gradation of open fractures according to sardjito scoring system suffered by motorcycles accident patients in emergency department of saiful Anwar General Hospital from April to June Prehosp Disaster Med 2011;26(Suppl 1):s Kementerian Kesehatan Republik Indonesia. Profil Kesehatan Indonesia Available at: go.id/resources/download/pusdatin/profil-kesehatanindonesia/profil-kesehatan-indonesia-2015.pdf [Accessed 4 November 2016]. 37. Soemarko M. Evaluation on the results of training and education for the pre hospital care to the injured patients. Jurnal Kedokteran Brawijaya 2004;XX: Kusumaningrum BR, Winarni I, Setyoadi, Kumboyono, Ratnawati R. Pengalaman Perawat Unit Gawat Darurat (UGD) Puskesmas dalam Merawat Korban Kecelakaan Lalu Lintas. Jurnal Ilmu Keperawatan 2013;1: Teo J. Traumatology specialist project in Malang, Indonesia. Australas Emerg Nurs J 2005;8: Lateef F, Anantharaman V. Emergency medical services in Singapore. Can J Emerg Med 2000;2: Lateef F, Anantharaman V. Emergency medical services in Singapore. ibid. 2000;2(4). 08

14 Volume 14 Issue 2 Article 2 Accuracy of stroke identification by paramedics in a metropolitan pre-hospital setting: A cohort study Teresa A Williams Curtin University, Australia David Blacker Sir Charles Gairdner Hospital, Western Australia Glenn Arendts Fiona Stanley Hospital, Western Australia Emily Patrick St John Ambulance, Western Australia Deon Brink St John Ambulance, Western Australia Judith Finn Curtin University, Australia 1

15 Williams: Accuracy of stroke identification by paramedics Research Accuracy of stroke identification by paramedics in a metropolitan pre-hospital setting: A cohort study Teresa A Williams PhD, RN, ICU Cert, is Senior Research Fellow 1,2 ; David Blacker MBBS, FRACP, is Neurologist and Stroke Physician 3 and Medical Director 4 ; Glenn Arendts PhD, MBBS, MMed, FACEM, is an Emergency Physician 5 ; Emily Patrick BSci, GradDip Paramedicine, MMed Res, is a paramedic 2 ; Deon Brink NDip, EmergMedCare, is Executive Manager Clinical Governance 2 ; Judith Finn PhD, RN, ICU Cert, FAHA, is Director 1 and Adjunct Research Professor 2 Affiliations: 1 Prehospital Resuscitation and Emergency Care Research Unit, Curtin University, Australia 2 St John Ambulance, Western Australia 3 Department of Neurology, Sir Charles Gairdner Hospital, Western Australia 4 Western Australian Neuroscience Research Institute 5 Fiona Stanley Hospital, Western Australia Abstract Introduction Acute stroke is a medical emergency. Identifying patients suffering a stroke is crucial if paramedics are to maximise delivery of appropriate management. One suggested screening tool to identify stroke is FAST (Face, Arms, Speech, Time), but the accuracy of identifying pre-hospital stroke is unknown. Objectives We aimed to: a) examine how well stroke is identified by paramedics using the emergency department (ED) discharge diagnosis as the comparator; and b) assess compliance with ambulance clinical practice guidelines. Methods A retrospective cohort study was conducted in the Perth metropolitan area in Western Australia between July 2012 and June 2014 using linked data from ambulance and ED databases. Patients aged 45 years and over, transported to the ED by road ambulance and assigned the ambulance problem code or ED discharge diagnosis of stroke, were selected. Positive predictive value (PPV), negative predictive value (NPV), specificity and sensitivity were calculated. Text fields were examined for documentation that patients met FAST criteria. Results There were 2217 patients identified as stroke by paramedics. Of the 1834 patients identified as stroke in the ED, 958 were also identified as stroke by paramedics; 876 were not identified as stroke. Sensitivity for identification of stroke was 52.2%. Of the 2096 patients identified as stroke by paramedics and who had an ED record, PPV was 958/2096 (45.7%), NPV 99.5% and specificity 99.4%. Paramedics recorded two or three stroke signs and symptoms in 1137 (51%) patients. Conclusion Systematic assessment and documentation is needed to better identify patients with stroke in the pre-hospital setting. Keywords: stroke; paramedic; stroke scales; FAST; ambulance; identification Corresponding author: Teresa A Williams, teresa.williams@curtin.edu.au 01

16 Williams: Accuracy of stroke identification by paramedics Introduction Emergency ambulance services play an integral role in the time-critical management of acute stroke victims. Paramedic identification of patients suspected of suffering a stroke is crucial to enable priority transport of the patient to an acute stroke centre (1). In addition, pre-hospital notification enables rapid referrals from emergency department (ED) staff to stroke teams and rapid access to imaging (2-4). The time from onset of symptoms to definitive diagnosis and treatment is important: less than 3 hours has been shown to improve patient outcomes (4-9). Patients transported by ambulance are more likely to arrive at an ED within 3 hours compared to those brought to the ED by other modes of transportation (5,10,11). For example, Kothari et al. (10) found transport of patients by ambulance (OR 4.0, 95% CI ) and those of Caucasian race (OR % CI ) were independently associated with ED arrival with 3 hours of symptom onset.. Transport by ambulance was also important in the Iosif et al. study (5) that reported 41% of 907 patients were transported by ambulance and 12% of those arrived within 3 hours at the hospital from the time of symptom onset. Other factors significantly associated with early arrival at the hospital were age, living with a friend and educational level (5). In Australia in 2013, 80% of patients with suspected stroke were transported to hospital by ambulance: 49% arriving within 3 hours and 58% arriving within 4.5 hours of stroke onset (12). Of patients not transported by ambulance only 34% arrived within 4.5 hours (12). Identification of stroke can be difficult in the pre-hospital setting because it is not consistently defined in clinical practice or research, the wide variation in stroke presentation and the lack of discriminatory ability of pre-hospital diagnostic tools (13,14). For example, stroke scales use different selection criteria to define stroke (13,15-18). Nevertheless, the introduction of stroke screening tools, such FAST (Face, Arms, Speech, Time), have improved paramedic identification of stroke (13,16,17,19). Few studies have examined how well stroke is identified by ambulance paramedics in Australian pre-hospital settings and compliance with stroke clinical practice guidelines (17,20). We aimed to: a) examine how well stroke is identified by paramedics with the ED discharge diagnosis used as the comparator; and b) assess compliance with the ambulance clinical practice guidelines. This information is important to inform pre-hospital clinical practice. Methods Study design A retrospective cohort study of prospectively collected ambulance data from 1 July 2012 to 30 June 2014 was linked to an ED database using probabilistic and deterministic data linkage and manual checking (21). The study was approved by St John Ambulance-Western Australia (SJA-WA) Research Advisory Group and the University s Human Research Ethics Committee (RA 128/2013). Setting All road-based emergency ambulance services in Western Australia (WA) are provided by a single ambulance service provider SJA-WA. There are nine public EDs in the Perth metropolitan area, one of which is dedicated to paediatric patients, and one private ED. Of these, three tertiary hospitals and one secondary hospital have stroke units. The emergency ambulance service is staffed by paramedics trained in providing a range of clinical interventions including advanced life support. Paramedics record clinical data including patient details, clinical assessment and treatment given, on an electronic patient care record (epcr) that is uploaded in real time to the ambulance server. The computer-aided dispatch data is also uploaded and combined with the epcr data to form a single ambulance database. Paramedic-assigned problem urgency is based on the Australasian Triage Scale (ATS) (22). The five ATS categories are: 1 resuscitation (lights and sirens), 2 emergency, 3 urgent, 4 semi-urgent and 5 non-urgent. An ambulance bespoke problem code is assigned by paramedics for what they believe to be the principal presenting problem, eg. problem code 312 for stroke. The SJA-WA clinical practice guidelines (CPG) (19,23,24) guide paramedics clinical practice. The CPG require paramedics to use the FAST test to assess for signs and symptoms of stroke (19). The FAST test has three criteria: Facial movements to assess symmetry, Arm movements to assess strength difference and arm drift, and Speech assessed for new disturbances. The T in FAST signifies the importance of time for patients having a stroke. Patients fulfilling two or three FAST criteria are considered FAST positive and to be suffering a stroke (19). Patients should be fast-tracked onto acute stroke bypass, ie. transported to a facility with an acute stroke service and the receiving ED pre-notified if (a) at least two of the three FAST criteria are identified; (b) the time from onset of symptoms to arrival at an acute stroke service would be less than 3 hours under priority 2 conditions (emergency, no lights and sirens) and (c) the patient s blood sugar level is 4 22 mmol/l (1). Where bypass criteria are not met, paramedics transport the patient to the nearest ED. Stroke bypass is disregarded if there is seizure activity or airway compromise because transport to the nearest ED is vital (19). Population All emergency patients aged 45 years and over (25,26) identified from the SJA-WA database were included if transported by a road ambulance to one of the eight public adult EDs in the Perth greater metropolitan area. Patients with suspected stroke were those assigned the problem code 312 by paramedics. Transient ischaemic attack (TIA) had a 02

17 Williams: Accuracy of stroke identification by paramedics different problem code. Rural patients were excluded because epcrs were not used and ambulances are usually staffed by volunteer ambulance officers, not paramedics. Patients transported by air medical services, or transfers, eg. medical practitioner referral or clinic appointments were also excluded. Procedures Two data sources were used: the SJA-WA database contains patient, service and clinical information from the epcr; and the ED Information System (EDIS) data (27). The EDIS is available from seven of the eight adult, metropolitan, public hospital EDs (93% of the SJA-WA transports to Perth public hospital EDs, unpublished data). Only those that used the EDIS were included in the comparisons of paramedic and ED-identified stroke, as per a previous study (28). The EDIS contains patient demographic information, admission, transfer and discharge destination and time-tracking information. The principal clinical diagnosis, mapped to International Classification of Disease Version 10-Australian Modification (ICD 10-AM) codes (29), is a mandatory EDIS field and is entered by ED medical staff on patient discharge from ED. The ICD 10-AM codes used for the ED discharge diagnosis of stroke were I60.9, I61.9, I62.0, I62.9, I63, I64 and I66.9 but not G45.4 (transient global amnesia) or G45.9 (TIA) (30). Outcome measures The primary outcome was sensitivity, ie. the proportion of ambulance service patient episodes with an ED discharge diagnosis of stroke who were correctly identified as stroke by the paramedics. Secondary outcomes were PPV, the proportion of patients identified as stroke by paramedics who also had an ED discharge diagnosis of stroke; negative predictive value (NPV), the proportion of patients for whom stroke was not recorded by paramedics and not identified in EDIS as stroke; and specificity, the proportion of patients without an EDIS diagnosis of stroke who also did not have stroke recorded by paramedics. Compliance with SJA-WA CPG were assessed by the proportion of cases assessed by FAST and transport to a stroke unit. Data analysis Cohort characteristics were described as appropriate for the data: mean and standard deviation (SD) or median and interquartile range (IQR) for continuous variables and proportions for categorical variables. Continuous variables were compared by t-tests if normally distributed or Mann Whitney tests if not normally distributed, categorical variables with chi-square. The variables included: pre-hospital variables pre-hospital triage, paramedic triage assessment (problem urgency), problem codes identified by paramedics, pre-hospital observations and treatments received; ED variables ED discharge diagnosis (ICD-10 AM), ED discharge destination. The ED discharge diagnosis was used as the gold standard for stroke. A two by two table was used to calculate PPV, NPV, sensitivity and specificity. Cases transported to ED by ambulance and identified as stroke in EDIS were used for the calculation of sensitivity. Paramedic-identified stroke cases in the ambulance database were used for the calculation for PPV. The calculation of the specificity and NPV included all ED cases transported by the ambulance service in the denominator not just those with stroke. Text descriptions of the patients signs and symptoms recorded by paramedics were extracted from the SJA-WA database using statistical software and examined for documentation of FAST criteria. The IBM SPSS Statistics for Windows, Version 21.0 (IBM Corp. Released Armonk, NY: IBM Corp.) was used for the statistical analysis. Results During the study period 2217 patients met the selection criteria (Figure 1). Of these, 51% were female. The median age was 80 (IQR 71 86) years. Seizures were reported by paramedics for 61 cases (3%). Falls were reported in 423 patients (19%). Overall 1742 (79%) were transported to hospital as problem urgency 1 or 2. The median time from SJA-WA call-takers answering the call to arrival at ED was 51 (IQR 42-62) minutes. Seventy-one percent of patients (n=1,578) were transported to one of the three tertiary hospitals, 163 (7%) to the secondary hospital with an acute care stroke unit, 467 (21%) to other secondary hospitals and nine to private hospitals (0.4%). There were 121 (5%) patients that could not be linked to EDIS: 110 patients transported to the public hospital with no EDIS facility, two to a tertiary hospital and not linked to EDIS, and nine transported to private hospitals. Age was slightly older (median 82, IQR years) and there were more females (69%) than patients with EDIS records. The ED discharge disposition for 2096 patients identified as stroke by paramedics and who had EDIS data: 87% of patients were admitted, 3% transferred to another health care facility, 9% discharged/left at own risk and 0.7% died in ED. ED-identified stroke During the study period 1834 patients who were transported to the ED by ambulance had an ED discharge diagnosis of stroke. Of these patients identified as stroke in the ED, 958 patients were also identified as stroke by paramedics (ambulance problem code 312), ie. a sensitivity of 52.2% (95% CI: %). The PPV was 45.7%, NPV 99.5% and the specificity 99.4%. Of the 876 patients not identified as stroke by the paramedics, 129 patients (7.0%) were identified as TIA and 284 (15.5%) as having another neurological condition. The problem codes assigned by paramedics to the EDidentified stroke cases are shown in Table 1. Problem urgency was assessed by paramedics as 1 or 2 in 88% of patients identified as stroke by both paramedics and in ED and 55% in those identified as stroke in ED only. For patients identified as stroke by paramedics and in ED, 86% were transported to a hospital with a stroke unit. For those identified only in ED, 75% were transported to a hospital with a stroke unit. 03

18 Williams: Accuracy of stroke identification by paramedics Figure 1. Flow chart of patient selection Table 1. Ambulance problem codes assigned by paramedics to 1834 patients with an emergency department discharge diagnosis of stroke Paramedic-assigned problem code Description Number of patients Percent Stroke and transient ischaemic attack CVA/stroke Transient ischaemic attack Other neurological Convulsion-other Altered consciousness Headaches Convulsion-epilepsy Status epilepticus Syncope Unconscious-unknown Trauma Cardiac Respiratory Infection Other Paramedic-identified stroke Of 2096 patients with an EDIS record and transported to ED by SJA-WA with suspected stroke (problem code 312), 958 also had an ED discharge diagnosis of stroke (PPV 45.7%, 95% CI: %). In addition, 334 patients (16%) identified by paramedics as stroke had an ED discharge diagnosis of TIA (ICD-10-AM code G45.9). The ED discharge diagnoses for patients identified as stroke by paramedics are shown in Table 2. Other cerebrovascular conditions and neurological conditions accounted for another 10% of the ED discharge diagnoses. Problem urgency was assessed by paramedics as 1 or 2 in 88% of patients identified as stroke by both paramedics and in the ED and 72% for patients not identified only by paramedics. Patients identified as stroke by paramedics and in ED were transported to a hospital with a stroke unit in 86% of cases while 80% of those identified only by paramedics were transported to a hospital with a stroke unit. 04

19 Williams: Accuracy of stroke identification by paramedics Table 2. Emergency department discharge diagnosis for 2096 patients identified as stroke by paramedics ED discharge ICD-10-AM code ICD-10-AM description Number of patients Percentage Cerebrovascular stroke I60.9 Subarachnoid haemorrhage, atraumatic 12.6 I61.9 Intracerebral haemorrhage, atraumatic I62.0 Subdural haemorrhage, atraumatic 9.4 I62.9 Intracranial haemorrhage, atraumatic 19.9 I63 Cerebral infarction 0 0 I64 Cerebrovascular accident, cause unknown I66.9 Cerebral thrombosis/embolism/infarction Cerebrovascular other I67.4 Hypertensive encephalopathy, hypertensive; malignant hypertension 2.1 I67.8, I67.9 Cerebrovascular insufficiency I69.8, Z86.7 Cerebrovascular accident with/without deficit, old [AGED?] Neurological seizures, transient ischaemic attack, other G40.1 Focal seizure G40.3 Generalised tonic/clonic seizure G41.0, G41.9 Grand mal epilepsy, status epilepticus G45.9 Transient ischaemic attack; transient global amnesia/ischaemia; cerebrovascular insufficiency, acute with focal signs C71.9, C79.3, C80, D32, D43.2, R Neurological other Neoplasm brain Malignant neoplasm, cerebral-metastasis/ meningioma, intracranial space occupying lesion Cardiovascular condition Mental health condition Respiratory tract condition Injury Infection other than respiratory Other

20 Williams: Accuracy of stroke identification by paramedics Discharge disposition of patients identified as stroke by paramedics and in the ED compared to patients not identified as stroke in the ED was: admissions 91% versus 81%, transfers 5% versus 3%, discharges 2% versus 16% and deaths 1.7% versus 0.2% (p<0.001). Two patients identified as stroke by both paramedics and in ED and one identified only in the ED left at their own risk. No difference was found in seizure activity in the ambulance between patients identified as stroke by both paramedics and in the ED (n=13) and those whose stroke was not identified in the ED (n=44, p=0.24). Compliance with ambulance clinical practice guidelines Examination of the epcr text fields revealed no consistent method used by paramedics to document the signs and symptoms of stroke on the epcr. The word FAST was documented in 42 cases (2%). Paramedics recorded signs and symptoms consistent with the FACE criteria of FAST in 1174 (53%) patients, the ARM criteria in 1,161 (52%) patients and the SPEECH criteria in 1,064 (48%) patients. For patients with no EDIS record, paramedics documented symptoms consistent with FACE (47%), ARM (49%) and SPEECH (39%), lower than those reported for patients overall. There were 1154 (52%) patients with two or three FAST criteria, of whom 81% were transported as problem urgency 1 or 2, and 80% were transported to a hospital with an acute stroke unit. Of the patients identified as stroke in the ED and identified by paramedics, 118 (19.7%) patients had no FAST criteria documented on the epcr by paramedics, 262 (52.8%) had one FAST criterion and 578 (78.5%) had two or three FAST criteria documented (Figure 2). For those patients not identified as stroke by paramedics, 484 (80.3%), 234 (47.2%), 158 (21.5%) patients had zero, one, two or three FAST criteria documented respectively. Sensitivity was 19.6% (95% CI: %), 52.8% (95% CI: %), 78.5% (95% CI: %) for patients with zero, one, two or three FAST criteria, respectively. Figure 2. Percentage of patients identified as stroke by paramedics and identified as stroke in the ED compared to those only identified as stroke in the ED by number of FAST criteria There was no difference in the number of FAST symptoms zero or one versus two or three in transfers to a hospital with a stroke unit (52%) or no stroke unit (48%, p=0.22) for paramedic-identified stroke. For those not transported to stroke unit hospitals, 45% were identified as stroke by the ED physicians. The most common symptom documented was facial symptoms (87%). For patients identified as stroke by paramedics and in the ED, 121 (33.6%), 273 (41.4%) and 564 (52.4%) had zero, one or two to three FAST criteria documented by paramedics respectively compared to 239 (66.4%), 387 (58.6%) and 512 (47.6%) in those identified as stroke only by paramedics (Figure 3). The PPV was 33.6% (95% CI: %), 41.4% (95% CI: %), 52.4% (95% CI: %) for patients with zero, one or two to three FAST criteria. 06

21 Williams: Accuracy of stroke identification by paramedics Figure 3. Percentage of patients identified as stroke by paramedics and identified as stroke in the ED compared to those only identified as stroke by paramedics by number of FAST criteria Discussion Over a 2-year period, 1834 cases had an ED discharge diagnosis of stroke and were transported by the ambulance service to the ED. Of these, 958 (46%) were also identified as stroke by paramedics. A further 7% of patients were identified by paramedics as TIA and 15% as other cerebrovascular conditions or neurological conditions. For 2096 patients assigned the problem code for stroke by paramedics, 1138 (52%) patients were not identified as stroke in the ED. Identifying stroke patients is difficult in the pre-hospital setting and the rates of pre-hospital stroke identification are variable. Studies have reported a higher accuracy of stroke identification by paramedics (31,32) than those found in this study while others report lower rates (33,34). Conditions that mimic the symptoms including TIAs were often included in those studies that may account for some of this variability. The ICD-10-AM codes and the paramedic-assigned problem codes for stroke do not include TIAs. Studies have also found up to 25% of patients with atypical stroke symptoms have a stroke mimic. Furthermore, identifying stroke within the first few hours is particularly difficult if signs and symptoms are unclear, eg. patient is comatose or has dysphasia, confusion, atypical or changing signs and symptoms, or is unable to illicit a past medical history (35,36). A balance is needed between under- and over-triage. Not identifying cases of stroke (undertriage) may impact on administration of time-critical treatment and lead to poorer outcomes. Over-triage may impact on ambulance and other health service resources not being available to be used most effectively. Stroke scales have been shown to improve outcomes but they use different selection criteria such as age restriction (26,37), patients wheelchair bound or bedridden at baseline (26), history of seizures or epilepsy (26), onset of symptoms >24 hours (26), Glasgow Coma Scale score <10 (16), and blood sugar levels (4 22 mmol/l, <4 mmols/l) (16,26). This study was restricted to patients aged 45 years and older based on other stroke scale exclusion criteria (26,37) and because the documentation in the ambulance data to detect stroke mimics was inconsistent. A systematic approach to clinical assessment and documentation improves paramedic identification of stroke (20) and ultimately patient outcomes. Several pre-hospital stroke triage tools (15-18,26) have been developed to identify those patients most likely to benefit from acute stroke care. Implementing education programs and strategies such as mandatory fields for documentation of the stroke signs and symptoms of the specific FAST criteria in a drop down menu of the epcr, a strategy that SJA-WA proposes to introduce shortly, are likely to lead to improvement in documentation in this area (17). The ambulance service is also introducing the Rapid Arterial occlusion Evaluation (RACE) scale (18) that will be used for patients who test positive for FAST to improve the identification of stroke (16,26,37). Accurate times of the onset of symptoms were poorly described on the epcrs making it difficult to assess the 3-hour time window. Patients whose time of onset is unknown cannot have thrombolysis due to the uncertainty of whether they are within the treatment time-window (3,38). The majority of patients were transported as high priority and most patients were transported to a hospital with an acute stroke care service. Three percent of patients were transferred from the initial hospital to another facility with higher levels of care. This management is consistent with stroke guideline recommendations (1,2,19). 07

22 Williams: Accuracy of stroke identification by paramedics This study has several limitations. It was a retrospective review of prospectively collected data. A strength of the study was the single ambulance service providing emergency road ambulance services for all Western Australia that facilitates consistency in clinical practice. As in previous studies (28,39), the limitations of using the ED discharge diagnosis as the comparator is acknowledged but access to the hospital discharge diagnosis was unavailable. However, the ED discharge diagnosis is often based on diagnostic tests that are simply not available in the pre-hospital setting. Notwithstanding this, the ED discharge diagnosis does not always concord with the final hospital diagnosis. An important limitation is having only one ED discharge diagnosis and one paramedic problem code for each patient care episode, as it relies on clinician judgement as to which is the most important condition. The FAST criteria to identify stroke were only reported in half the cases. When evaluating the reporting of FAST, it cannot be assumed that patients did not have relevant signs or symptoms: they may not have been documented by paramedics on the epcr or FAST may not have been used and this should be considered when interpreting the study results. Further, the limited data on FAST may affect the interpretation of the sensitivity and specificity. Conclusion It is unclear whether problems of paramedic assessment are failure to apply FAST, failure to detect positive FAST signs, or failure to record FAST findings. Systematic assessment and documentation using a stroke scale is likely to improve documentation that may lead to better management of these patients. Acknowledgement Professor Ian Jacobs, who was a co-investigator of this project and Clinical Services Director of the ambulance service, died before the final draft of this manuscript was completed. We acknowledge his support and contribution to this study. We also acknowledge St John Ambulance-WA for providing access to the ambulance data. Conflict of interest The authors declare they have no competing interests. Each author of this paper has completed the ICMJE conflict of interest statement. St John Ambulance WA played no role in the study design, conduct or interpretation of the results. References 1. Government of Western Australia Department of Health. Model of Stroke Care Perth: Health Networks Branch, Department of Health, Western Australia National Stroke Foundation. Clinical Guidelines for Stroke Management. Melbourne: National Stroke Foundation; Sheppard JP, Mellor RM, Greenfield S, et al. The association between prehospital care and in-hospital treatment decisions in acute stroke: a cohort study.emerg Med J 2015;32: Patel MD, Rose KM, O Brien EC, Rosamond WD. Prehospital notification by emergency medical services reduces delays in stroke evaluation: findings from the North Carolina stroke care collaborative.stroke 2011;42: Iosif C, Papathanasiou M, Staboulis E, Gouliamos A. Social factors influencing hospital arrival time in acute ischemic stroke patients.neuroradiology 2012;54: Silvestrelli G, Parnetti L, Paciaroni M, et al. Early admission to stroke unit influences clinical outcome. Eur J Neurol 2006;13: IST-3 collaborative group, Sandercock P, Wardlaw J, et al. The benefits and harms of intravenous thrombolysis with recombinant tissue plasminogen activator within 6 h of acute ischaemic stroke (the third international stroke trial [IST-3]): a randomised controlled trial. Lancet 2012;379: Ebinger M, Winter B, Wendt M, et al. Effect of the use of ambulance-based thrombolysis on time to thrombolysis in acute ischemic stroke: a randomized clinical trial. Jama 2014;311: Lees KR, Bluhmki E, von Kummer R, et al. Time to treatment with intravenous alteplase and outcome in stroke: an updated pooled analysis of ECASS, ATLANTIS, NINDS, and EPITHET trials. Lancet 2010;375: Kothari R, Jauch E, Broderick J, et al. Acute stroke: delays to presentation and emergency department evaluation. Ann Emerg Med 1999;33: Lacy CR, Suh DC, Bueno M, Kostis JB. Delay in presentation and evaluation for acute stroke: Stroke Time Registry for Outcomes Knowledge and Epidemiology (S.T.R.O.K.E.). Stroke 2001;32: National Stroke Foundation. National Stroke Audit Acute Services Clinical Audit Report, Clinical Executive Summary Melbourne, Australia. 13. Hankey GJ, Blacker DJ. Is it a stroke? BMJ 2015;350:h Turc G, Maier B, Naggara O, et al. Clinical scales do not reliably identify acute ischemic stroke patients with largeartery occlusion. Stroke 2016;47: Kothari RU, Pancioli A, Liu T, Brott T, Broderick J. Cincinnati Prehospital Stroke Scale: reproducibility and validity. Ann Emerg Med 1999;33: Chenkin J, Gladstone DJ, Verbeek PR, et al. Predictive value of the Ontario prehospital stroke screening tool for the identification of patients with acute stroke. Prehosp Emerg Care 2009;13: Bray JE, Martin J, Cooper G, Barger B, Bernard S, Bladin C. An interventional study to improve paramedic diagnosis of stroke. Prehosp Emerg Care 2005;9: de la Ossa NP, Carrera D, Gorchs M, et al. Design and validation of a prehospital stroke scale to predict large arterial occlusion: The Rapid Arterial occlusion Evaluation Scale. Stroke 2014;45:

23 Williams: Accuracy of stroke identification by paramedics References (continued) 19. St John Ambulance Western Australia. Clinical practice guidelines for ambulance care in Western Australia Version 24. Belmont: St John Ambulance-Western Australia Bray JE, Coughlan K, Barger B, Bladin C. Paramedic diagnosis of stroke: examining long-term use of the Melbourne Ambulance Stroke Screen (MASS) in the field. Stroke 2010;41: Jaro MA. Probabilistic linkage of large public health data files. Stat Med 1995;14: Australasian College for Emergency Medicine. The Australasian Triage Scale. Emerg Med (Fremantle) 2002;14: Ambulance Victoria (2012). Clinical Practice Guidelines. Available at: Qualified-Paramedic-Training/Clinical-Practice-Guidelines. html [Accessed October 2012]. 24. Association of Amulance Chief Executives, Joint Royal Colleges Ambulance Liaison Committee, The University of Warwick. UK Ambulance Service Clinical Practice Guidelines Bridgwater, UK: Class Professional Publishing Ltd; Koster RW. Modern BLS, dispatch and AED concepts. Best Pract Res Clin Anaesthesiol 2013;27: Kidwell CS, Starkman S, Eckstein M, Weems K, Saver JL. Identifying stroke in the field. Prospective validation of the Los Angeles prehospital stroke screen (LAPSS). Stroke 2000;31: Department of Health Western Australia (2007). Information Management and Reporting. Emergency Department Data Collection Data Dictionary Version 1.0. Available at: www. health.wa.gov.au/healthdata/docs/eddc_dictionary.pdf [Accessed 2 October 2010]. 28. Williams TA, Finn J, Celenza A, Teng T-H, Jacobs IG. Paramedic Identification of Acute Pulmonary Edema in a Metropolitan Ambulance Service. Prehosp Emerg Care 2013;17: Commonwealth of Australia. The International Statistical Classification of Diseases and Realted Health Problems, Tenth Revision, Australian Modification (ICD-10-AM) 5th Ed. Sydney, NSW: National Centre for Classification in Health; Australian Institute of Health and Welfare. Stroke and its management in Australia: an update. Cardiovascular disease series no. 37. Cat. no. CVD 61. Canberra: AIHW, Fothergill RT, Williams J, Edwards MJ, Russell IT, Gompertz P. Does use of the recognition of stroke in the emergency room stroke assessment tool enhance stroke recognition by ambulance clinicians? Stroke 2013;44: Brandler ES, Sharma M, McCullough F, et al. Prehospital Stroke identification: factors associated with diagnostic accuracy. J Stroke Cerebrovasc Dis 2015;24: Frendl DM, Strauss DG, Underhill BK, Goldstein LB. Lack of impact of paramedic training and use of the cincinnati prehospital stroke scale on stroke patient identification and on-scene time. Stroke 2009;40: Ramanujam P, Guluma KZ, Castillo EM, et al. Accuracy of stroke recognition by emergency medical dispatchers and paramedics--san Diego experience. Prehosp Emerg Care 2008;12: Fernandes PM, Whiteley WN, Hart SR, Al-Shahi Salman R. Strokes: mimics and chameleons. Pract Neurol 2013;13: Merino JG, Luby M, Benson RT, et al. Predictors of acute stroke mimics in 8187 patients referred to a stroke service. J Stroke Cerebrovasc Dis 2013;22:e Bray JE, Martin J, Cooper G, Barger B, Bernard S, Bladin C. Paramedic identification of stroke: community validation of the Melbourne Ambulance Stroke Screen. Cerebrovasc Dis 2005;20: Bladin C. Stroke thrombolysis: per ardua, ad astra. Intern Med J 2014;44: Williams T, Finn J, Fatovich D, Perkins G, Summers Q, Jacobs I. Paramedic differentiation of asthma and COPD in the prehospital setting is difficult. Prehosp Emerg Care 2015;19:

24 Volume 14 Issue 2 Article 3 Evaluating the knowledge base and current training of paramedics in the southwest United States in assessing and managing toxic alcohol exposures Christopher Bare Dona Ana Community College, United States of America Oliver Grundmann University of Florida, United States of America 1

25 Bare: Toxic alcohol online survey Research Evaluating the knowledge base and current training of paramedics in the southwest United States in assessing and managing toxic alcohol exposures Christopher Bare MS, BSN is Instructor 1 ; Oliver Grundmann PhD, MS is Clinical Associate Professor 2 Affiliations: 1 Dona Ana Community College, New Mexico State University, United States of America 2 Department of Medicinal Chemistry, College of Pharmacy, University of Florida, United States of America Abstract Introduction Toxic alcohol overdose poses a problem in patients who are suffering from the toxicological consequences of toxic alcohol exposure and front line emergency care providers, such as paramedics, need to be well versed in assessing patients, identifying specific toxicities, and implementing appropriate therapies. Objectives The aim of this research was to identify the current knowledge base of paramedics to recognise key clinical, pathophysiological and treatment features of toxic alcohol poisoning. Additionally, the emphasis in training and education of paramedics in terms of the level of importance assigned to toxic alcohol emergencies was evaluated. Methods The study was conducted as an anonymous online survey. Demographic data, timed choice and open-ended questions were collected to evaluate knowledge and identify gaps in toxic alcohol poisoning training by paramedics. The survey link was sent out to emergency medical services (EMS) organisations and individual paramedics in the southwest United States. Bivariate analysis via Pearson correlation coefficient (PCC) was used to compare variables. Results Eighty paramedics participated in the survey, with 60 (75%) participants completing all questions. Respondents were able to identify common sources of toxic alcohol exposure to varying degrees with 58% for methanol and 89% for ethylene glycol. Data indicated that a lack of knowledge of the underlying pathophysiology was related to missing education (PCC <0.05) in toxicology. Education appears to have been insufficient in regards to recognising and treating toxic alcohol exposure. A majority of respondents (68.6%) believed that assessment and treatment of toxic alcohol poisoning is an important component of their training. Conclusion Although symptom recognition for toxic alcohol poisoning is present in most EMS providers, pharmacological intervention and treatment approaches were often not known. The results indicate that there is a need for educators and curriculum builders to include additional coverage of topics of toxicological importance such as the anion gap and toxic alcohol assessment, pathophysiology and treatment. Keywords: Toxic alcohols, paramedics, education, anion gap Corresponding author: Christopher Bare, cbare@nmsu.edu 01

26 Bare: Toxic alcohol online survey Introduction Toxic alcohols are examples of hydroxylated hydrocarbons associated with specific toxicological consequences. The three major toxic alcohols are isopropyl alcohol, ethylene glycol and methanol (1). According to 2012 data involving poison control centre reports in the United States (US) there was a total of 16,458 cases of isopropanol ingestion and 8773 cases of actual toxicity with over 60 reports of complications including one death (2). Many of these cases were the result of ingesting common household products such as rubbing alcohol, hand sanitisers and various cleaning agents (3). Reports of methanol and ethylene glycol exposure were considerably less but not trivial in terms of mortality. According to the same data set there were 5869 cases of ethylene glycol ingestion and 1612 cases of methanol ingestion (3). In spite of these lower exposure numbers, there were expressively higher rates of mortality in methanol and ethylene glycol exposed patients. Of the methanol patients, there were over 200 patients that experienced severe complications and six fatalities (3). Of the ethylene glycol patients, over 200 patients experienced complications requiring extended hospitalisation with 23 deaths. Clearly, toxic alcohol overdoses pose a problem in patients who are suffering from the toxicological consequences of toxic alcohol exposure, and front line emergency care providers such as paramedics need to be well versed in assessing patients, identifying specific toxicities and implementing appropriate therapies. Sixteen percent of all paramedics in the US work in emergency rooms where they provide initial assessment and management of patients who are admitted with poisoning emergencies in addition to their pivotal role in the pre-hospital setting (4). Because paramedics are often in situations where they encounter patients that are experiencing toxicological emergencies, they are required to assess, identify and properly manage victims of toxic alcohol emergencies. A review of the current US National Emergency Medical Services Educational Standards, Paramedic Instructional Guidelines found that the term alcohols was present among a list of other specific agents of toxicological significance such as cyanide and carbon monoxide (5). A list of specific antidotal agents such as methylene blue and cyanide antidotes are also presented in this document but fomepizole and ethanol were not included in this list. Because the national educational guidelines appear to be vague regarding the amount of attention that should be focused on toxic alcohol exposure, it is reasonable to suspect that the amount of time spent discussing specific toxicological issues in paramedic educational programs may vary significantly. However, while cyanide poisonings and antidotes are specifically mentioned in this document, only about 202 cases of cyanide exposure were reported to poison control centres in 2012 (3). While some predict that cyanide toxicity may contribute to the morbidity and mortality of perhaps up to 10,000 deaths following smoke inhalation (6), there were still more reported cases of methanol, ethylene glycol and isopropyl alcohol exposures in 2012 making them at least as common as cyanide exposure even when accounting for smoke inhalation. The aim of this study was to use an anonymous survey that would help determine the knowledge base of paramedics with a focus on the ability to identify key clinical, pathophysiological and treatment features of toxic alcohol poisoning. Additionally, the survey attempted to assess the current training of paramedics in terms of the level of importance assigned to toxic alcohol emergencies and how much initial and recurrent training paramedics feel they received on toxic alcohols. Finally, this study aimed to identify how much time was spent covering toxicology in general among paramedics educated in the US. Methods The study protocol was developed and approved by the Institutional Review Board at the University of Florida (IRB 2014-U-1393). This study employed an anonymous online survey using the Qualtrics software program and server available to university faculty and students. The data collection included demographics, timed multiple choice questions and brief text answers of a targeted convenience sample readily available to the investigators. The survey link was sent out to various emergency medical service (EMS) organisations and individual paramedics for further distribution. Bivariate logistic regression analysis and Pearson correlation coefficient (PCC) were used to relate quiz questions with demographic and training variables. Differences were considered significant if p Inclusion criteria: To be included in the study, participants had to be at least 18 years of age and also an active paramedic in the US, which had to be acknowledged at the beginning of the survey. Due to the anonymity of the survey, a small risk exists that a person under the age of 18 participated. Additionally, there was a small risk that non-paramedic credentialed providers could take the survey. Exclusion criteria: Participants were excluded from the survey if they did not meet the inclusion criteria. Survey sections and variables Demographic variables: The initial assessment tool of the study gathered demographic information from each participant. This information included: gender, age range, number of years employed as a paramedic and highest level of education. 02

27 Bare: Toxic alcohol online survey Toxic alcohol assessment and management variables: Following the demographic assessment, participants were asked 10 questions about the assessment and management of toxic alcohol overdose. The questions focused on identifying key features of toxic alcohol overdose, key pathophysiological processes and pharmacological treatment modalities for three significant toxic alcohols; methanol, isopropyl alcohol and ethylene glycol. Participants were given 30 seconds to answer each question. The time limit was designed to prevent participants from looking answers up as opposed to truthfully answering the question using innate knowledge. The quiz questions are listed in Table 1 with their respective answer choices (Table 1). Post assessment and training variables: The final section of the survey contained questions that attempted to ascertain what the participant thought about the importance of incorporating toxic alcohol education into initial and recurrent paramedic education and how much time the participant spent on toxic alcohols and general toxicology during initial and recurrent training. Results The total response to the online survey was relatively low, with 80 responses even after multiple reminders were sent to various EMS providers and individuals in the field. Of these 80, three-quarters (n=60) completed the full survey, with participants mainly dropping out during the timed multiplechoice responses that evaluated their actual knowledge about toxic alcohols. The demographics (Table 2) indicate a mainly male population response (90% male), which is fairly representative of the overall EMS field in the US. The main age range for this population was years (62.9% of all respondents) with equal tails to the younger and older age ranges. A majority of respondents (65.7%) had worked in the EMS field between 6 20 years (Table 2), which indicates experienced professionals likely to have been involved with toxic alcohol exposures in their careers and practical settings. No respondent answered all 10 quiz questions correctly. Only one respondent was able to identify nine correct answers while a majority of respondents (54%) answered between 4 6 questions correctly. The distribution of incorrect responses was not specific to a toxic alcohol. One main question of this survey was to seek an association between prior education and knowledge of specific toxic alcohol symptoms and treatments. About half (55%) of respondents had either a non-college degree or an Associate degree in the field of EMS (Figure 1). A comparison of percentage correct responses to education showed that overall there were no significant differences in response rates. However, one question that indicated significant differences was question 2 ( Which of the following findings would you expect to find among patients with significant toxic alcohol overdose? ) with lower correct response rates from participants with either an Associate degree or Master degree (Figure 1). This may indicate that there is inconsistent knowledge in initial diagnosis of toxic alcohol exposure. This was further confirmed by the association between percentage correct response rates and training in anion gap, and how it relates to clinical symptoms (Figure 2). Interestingly, the differences here between groups indicates a wide variability in knowledge even if the anion gap was discussed. For question 3 ( Which of the following products would be most likely to contain methanol? ) a higher correct response rate was observed by those who did not learn about the anion gap as for those who learned about it or do not remember. The opposite was the case for correct responses to question 8 ( What is the pathophysiological basis of methanol toxicity? ), which also asked about methanol toxicity. For this question a majority of correct responses came from participants who learned about the anion gap during their training compared to those who did not or could not remember (Figure 2). All other correlations between age, length of working in the field and prior exposure to toxicology did not indicate significant differences with the timed multiple choice toxic alcohol questions. About two-thirds of respondents (68.6%) either agreed or strongly agreed that knowledge of toxic alcohol symptoms and treatment approaches was important knowledge to them (Table 3). All bivariate analysis responses associating quiz questions with demographic variables and corresponding significance levels are listed in Table 4. Discussion Toxic alcohol exposure has a significant impact on the healthcare system and the individual patient. As first responders, EMS providers and paramedics are in a position to diagnose and treat initial symptoms when confronted with them. The results of this small sample size survey indicate that there is a need for further education on toxic alcohol exposures that can translate into life-saving measures and potentially reduce the burden on the healthcare system by decreasing morbidity and mortality in the general population. Knowledge of toxic alcohol symptoms, pathophysiology, and treatment approaches by EMS providers appears to be variable and inconsistently taught during formal and recertification training. Although small in scale, the results of this survey may serve the larger EMS community, especially educational providers and educators at colleges and universities, to dedicate more hours to toxic alcohol training. Initial symptom recognition may provide pre-hospital treatment approaches that prevent progression of organ damage. 03

28 Bare: Toxic alcohol online survey Table 1. Online timed multiple-choice questions to evaluate current knowledge of toxic alcohol exposure by survey participants Question # Question Answer choices (correct answer is in bold) 1 Which of the following electrocardiogram Hypocalcaemia with prolonged QT interval (ECG) findings would be con- Hyperkalaemia with the development of atrial dysrhythmias cerning with a severe ethylene glycol Hypernatremia with significant AV blocks overdose? Hypomagnesemia with the development of atrial fibrillation and flutter 2 Which of the following findings would you expect to find among patients with significant toxic alcohol overdose? 3 Which of the following products would be most likely to contain methanol? 4 Which of the following products would be most likely to contain ethylene glycol? 5 Which of the following findings is unique in the setting of isopropyl alcohol overdose? 6 Which complaint would be most suggestive of methanol toxicity? 7 What is the pathophysiological basis of ethylene glycol toxicity? 8 What is the pathophysiological basis of methanol toxicity? 9 What is the front line pharmacological therapy for treating methanol and ethylene glycol toxicity? 10 If supportive care and front line pharmacological therapies fail to treat a severe toxic alcohol overdose, which of the following is considered definitive treatment? Elevated anion gap metabolic acidosis Hyperchloraemic metabolic acidosis Partially compensated respiratory alkalosis Uncompensated respiratory alkalosis Adulterated home brewed ethanol containing spirits Brake fluids Household insect killing sprays Household plant fertilisers Radiator fluid Brake fluid Household insect killing sprays Household toilet bowl cleaners Ketosis without the presence of acidosis on the arterial blood gas Ketoacidosis with partial compensation on the arterial blood gas Marked uremic syndrome with alkalosis on the arterial blood gas Marked leukocytosis with a left shift and uncompensated respiratory acidosis on the arterial blood gas Visual changes Flank pain Generalised myalgia Peripheral neuropathy Oxalic acid formation and nephrotoxicity Acetone formation and ketosis Acetaldehyde formation and inebriation Formic acid formation and optic nerve toxicity Formic acid formation and optic nerve toxicity Acetone formation and ketosis Oxalic acid formation and nephrotoxicity Acetaldehyde formation and inebriation Fomepizole Hypertonic saline solution Lipid emulsion therapy Metronidazole Haemodialysis Activated charcoal lavage Potassium therapy Sodium bicarbonate therapy 04

29 Bare: Toxic alcohol online survey Table 2. Demographic data of survey participants Question Frequency Percent How long have you worked in the EMS field? 0-5 years years years years or more years What is your age in years? What is your gender? Female 7 10 Male Especially low correct response rates were observed for question 9 ( What is the front line pharmacological therapy for treating methanol and ethylene glycol toxicity? ) which is troublesome given the significant exposures to methanol and ethylene glycol toxicity a combined 7481 cases which is close to the roughly 10,000 cases reported for smoke inhalation exposures in An even more significant finding of the survey is that two-thirds of respondents agree that further education and training on toxic alcohol exposure should be part of either their initial or recertification training (Table 3). These findings clearly indicate that there is a need for education that can directly translate into clinical practice and better health outcomes. Limitations One obvious limitation of the findings is the small sample size. This limits interpretation of the results to the larger EMS community. This targeted convenience sample may not have been heterogeneous enough to reach the larger EMS community because recruitment was limited to personal contacts and reliance on word of mouth. Another limitation was the relatively brief time period for collection of data due to time restrictions imposed by the institutional review board. The attrition rate over the course of the survey was highest in the section evaluating paramedics knowledge of toxic alcohols pointing to a potential skewed sample representing the best informed paramedics. The survey focused on cognitive rather than applied skills and may therefore not account for treatment approaches learnt in the field other than current best practice guidelines tested in the knowledge section. Further limitations are a lack of stratification by location (urban versus rural) or employment (private versus public). This survey also does not account for alcohol as a co-ingestant with other toxicants such as opioids or stimulants, which is commonly occurring and may change the course of treatment. Conclusion These survey results provide a base for further discussion on the educational needs and current knowledge on toxic alcohol exposures by EMS providers and paramedics. Although the study sample size was small, the findings clearly indicate that providing further education in this area will improve provider knowledge, diagnosis and early treatment intervention that is linked to better health outcomes for patients. Further research is required to extend the findings of this study to the larger EMS community be it on the national or international level. We hope that our findings will stimulate further discussion and research to improve best practices on integrating knowledge on toxic alcohol symptoms, diagnosis and treatment, and consideration by educational providers and institutions to revise their curriculum for initial EMS training and recertification alike. Conflict of interest The authors declare they have no competing interests. Each author of this paper has completed the ICMJE conflict of interest statement. 05

30 Bare: Toxic alcohol online survey Table 3. Survey questions related to toxic alcohol training evaluated after completion of the timed multiple-choice questions Question Frequency Percent How much time did you spent discussing toxic alcohol poisoning during your paramedic training? 0 hours Less than one hour but more than 0 hours hours hours Do not remember 7 10 Not answered How much time did you spent discussing toxicology in general during your paramedic training? 0 hours Less than one hour but more than 0 hours hours hours hours hours Not answered Did you discuss toxic alcohol poisoning as part of your recurrent or refresher training for relicensure? Yes No Do not remember Not answered Do you believe that understanding how to assess and treat toxic alcohol poisoning is an important part of being a paramedic? Strongly agree Agree Slightly agree Disagree Strongly disagree Not answered

31 Bare: Toxic alcohol online survey Table 4. Principle component comparison and bivariate analysis (binned by correct vs. incorrect responses) for each quiz question. Bivariate analysis by education, work experience, and age for N=70 responses. Bivariate analysis by training and toxic alcohol refresher for N=60 responses Quizquestion Education Work experience Age N correct response (%) Toxic alcohol training General toxicology training Anion gap training Toxic alcohol refresher N correct response (%) (37) (38) * (41) (43) (50) * (57) (80) * 53 (88) (21) (25) (49) (57) (26) (30) (33) * (38) (14) (17) (66) (77) * = associations that reached statistical significance 07

32 Bare: Toxic alcohol online survey Figure 1. Distribution of survey participants by education (pie chart insert). Percent correct responses to each timed multiple-choice question by education (bar chart) 08

33 Bare: Toxic alcohol online survey Figure 2. Distribution of survey participants by training and knowledge regarding the anion gap (pie chart insert). Percent correct responses to each timed multiple-choice question by training or knowledge about the anion gap (bar chart) 09