- Email: [email protected]
Barriers to dispatcher-assisted cardiopulmonary resuscitation in Singapore
Resuscitation 105 (2016) 149–155
Contents lists available at ScienceDirect
Resuscitation journal homepage: www.elsevier.com/locate/resuscitation
Clinical paper
Barriers to dispatcher-assisted cardiopulmonary resuscitation in Singapore夽 Andrew Fu Wah Ho a,∗ , Zariel Jiaying Sim b , Nur Shahidah c , Ying Hao d , Yih Yng Ng e , Benjamin S.H. Leong f , Siti Zarinah g , Winston K.L. Teo g , Geraldine Shu Yi Goh g , Hamizah Jaafar g , Marcus E.H. Ong c,h a
Emergency Medicine Residency, Singapore Health Services, Singapore, Singapore Barts and the London School of Medicine and Dentistry, Greater London, United Kingdom c Department of Emergency Medicine, Singapore General Hospital, Singapore, Singapore d Division of Medicine, Singapore General Hospital, Singapore, Singapore e Medical Department, Singapore Civil Defence Force, Singapore, Singapore f Emergency Medicine Department, National University Hospital, Singapore, Singapore g Operations Department, Singapore Civil Defence Force, Singapore, Singapore h Health Services & Systems Research, Duke-NUS Medical School, Singapore, Singapore b
a r t i c l e
i n f o
Article history: Received 5 February 2016 Received in revised form 31 March 2016 Accepted 9 May 2016 Keywords: Cardiac arrest Dispatcher-assisted Bystander Cardiopulmonary resuscitation Telephone CPR
a b s t r a c t Background: Dispatcher-assisted cardiopulmonary resuscitation (DA-CPR) is effective in increasing bystander CPR in out-of-hospital cardiac arrests (OHCA). Singapore has recently implemented a DACPR program. We aimed to characterize barriers to commencement of chest compressions by callers in Singapore. Methods: We analyzed dispatch recordings of OHCA cases received by the ambulance call center between July 2012 and March 2015. Audio recordings of poor quality were excluded. Trained reviewers noted the sequential stages of the dispatcher’s recognition of CPR, delivering CPR instructions and caller performing CPR. Time taken to reach these milestones was noted. Barriers to chest compressions were identified. Results: A total of 4897 OHCA occurred during the study period, overall bystander CPR rate was 45.7%. 1885 dispatch recordings were reviewed with 1157 cases qualified for dispatcher CPR. In 1128 (97.5%) cases, the dispatcher correctly recognized the need for CPR. CPR instructions were delivered in 1056 (91.3%) cases. Of these, 1007 (87.0%) callers performed CPR to instruction. One or more barriers to chest compressions were identified in 430 (37.2%) cases. The commonest barrier identified was “could not move patient” (27%). Cases where barriers were identified were less likely to have the need for CPR recognized by the dispatcher (94.9% vs. 99.0%, p < 0.001), CPR instructions given (79.3% vs. 98.3%, p < 0.001) and CPR started (67.9% vs. 98.3%, p < 0.001), while the time taken to reach each of these stages were significantly longer (p < 0.001). Conclusion: Barriers were present in 37% of cases. They were associated with lower proportion of CPR started and longer delay to CPR. © 2016 Elsevier Ireland Ltd. All rights reserved.
Introduction Out-of-hospital cardiac arrest (OHCA) is a leading cause of death in Singapore, mirroring worldwide trends. In the Asia-Pacific, OHCA
夽 A Spanish translated version of the abstract of this article appears as Appendix in the final online version at http://dx.doi.org/10.1016/j.resuscitation.2016.05.006. ∗ Corresponding author at: c/o Department of Emergency Medicine, Singapore General Hospital, Outram Road, Singapore 169608, Singapore. E-mail address: [email protected] (A.F.W. Ho). http://dx.doi.org/10.1016/j.resuscitation.2016.05.006 0300-9572/© 2016 Elsevier Ireland Ltd. All rights reserved.
survival rates are generally low, ranging from 2 to 11%,1 with the survival to hospital discharge rate in Singapore found to be 11.0% between 2010 and 2012.2 Effective treatment of OHCA hinges on Cummins’ “Chain of Survival” concept which describes the rapid commencement and seamless provision of a set of rescuer actions.3 One crucial link in the chain is early commencement of cardiopulmonary resuscitation (CPR), which improves outcomes in OHCA.4 As OHCAs generally occur away from the immediate presence of healthcare providers, early CPR depends heavily on the recognition of cardiac arrest by bystanders as well as bystander CPR, which can more than double the patient’s chance of survival.5
150
A.F.W. Ho et al. / Resuscitation 105 (2016) 149–155
In recent years, dispatcher-assisted CPR (DA-CPR) has emerged as an effective intervention to increase bystander CPR and hence OHCA outcomes.6 DA-CPR functions on the notion that when a bystander calls emergency medical services (EMS), not only does the call provide dispatch service, it also creates the opportunity for the EMS to catalyze recognition of cardiac arrest and initiation of bystander CPR through systematic interrogation and “just-in-time” pre-arrival instructions.6 Indeed, DA-CPR can nearly double the proportion of OHCA patients who receive bystander CPR.7,8 In Singapore, bystander CPR rate was low and remained relatively constant over the years, being 19.7% from 2001 to 2004 and 22.4% from 2010 to 2012.2 This was despite a relatively high proportion of the population having been trained in CPR, with a figure of 31.4% found in a national population based survey.9 DA-CPR was thought to be useful in such a setting to increase bystander CPR rate. We hypothesized that DA-CPR would be associated with a high rate of CPR performance by overcoming some of the barriers encountered by emergency callers. As part of the Pan-Asian Resuscitation Outcome Study Phase II (PAROS-II), Singapore implemented a DA-CPR package in 2012,10 which included a DA-CPR protocol, dispatcher training, systematic quality improvement (QI) through review of all dispatch calls and a public education campaign around DA-CPR. This study aimed to provide a look into the barriers to DA-CPR in Singapore, in order to identify system-dependent or cultural-based barriers. Such data may inform the implementation of DA-CPR in the other developing EMS systems. Materials and methods Setting Singapore is a highly-urbanized island city-state situated at the southernmost tip of the Malaysian peninsula with a population of 5.3 million over a land area of 715.1 square kilometers.12 EMS is provided by the Singapore Civil Defence Force (SCDF), which operates a fleet of 46 ambulances and 15 first responders on motorcycles in a fire-based system activated by a centralized 9-9-5 dispatching system.12 SCDF utilizes computer aided dispatch protocols, global positioning satellite automatic vehicle location systems and road traffic monitoring systems. SCDF paramedics (equivalent to North American emergency medical technician-intermediate) are trained in basic life support, automated external defibrillator usage and specific interventions including adrenaline administration. Mean time taken from call to arrival at scene was 9.9 min from 2010 to 2012.2 EMS for a minority of emergency cases are provided by private operators that charge a fee whereby the caller can specify the hospital of conveyance. These are mainly used for inter-facility transfers and do not respond to ‘995’. The emergency ambulance services provided by calling 995 is free of charge to all users and supported through government taxes.12 Dispatcher-assisted cardiopulmonary resuscitation in Singapore Singapore adopted a DA-CPR protocol in 201113 ; however, due to the lack of systems for training and QI, CPR instructions were only given in less than 5% of cases. As part of the Pan-Asian Resuscitation Outcome Study Phase II (PAROS-II), Singapore was selected for implementation of a comprehensive DA-CPR package. This package consisted of (1) a standardized dispatch protocol, (2) a standardized training package for dispatchers, (3) a measurement tool to collect data on individual dispatcher and organization-level performance, (4) an integrated QI program and (5) a community education program.11 The program was implemented in July 2012
with a planned 6 month “run-in” period. By Jan 2013, all dispatchers were able to provide DA-CPR. This package consisted of11 1. A standardized dispatch protocol that guides dispatchers to systematically question callers to accurately and rapidly determine whether the patient is in cardiac arrest. When a cardiac arrest patient is identified, the protocol guides the dispatcher to give CPR instructions to assist bystanders if CPR is not already ongoing. 2. A training package consisting of 1 day intensive training course both call-takers and dispatch center managers. It comprises didactic teaching as well as practical exercises and scenarios for dispatchers as well as supervisors/medical directors. 3. A standardized measurement QI tool to collect data on individual dispatcher and organizational level performance through review of the dispatch audio recordings. A sample of the QI tool is shown in Appendix 1. 4. An integrated quality improvement program that includes cooperation and collaboration of pre-hospital and hospital stakeholders. The program provides feedback at the individual and organizational level. It involved review of dispatcher CPR audio recordings and use of the QI tool mentioned in step 3. 5. A community education program known as Dispatcher Assisted First REsponder (DARE) program was developed to update the public on DA-CPR, how to recognize cardiac arrest, the importance of early activation of EMS and staying on the line to perform DACPR. In SCDF, dispatchers are trained to adopt a streamlined, two-step question approach to identify the need for CPR. The caller is asked (1) “Is the victim conscious?” and (2) “Is the victim breathing normally?” If a negative response to both questions are received, the presumptive diagnosis is cardiac arrest and the dispatcher enters into a protocol to deliver CPR instructions.14 Instructions for chest compression-only CPR are given, unless there are specific indications for telephone instructions on ventilations, like drowning or pediatric cardiac arrest. The QI process involves having a team of four dispatch nurses review recordings of cardiac arrest calls every shift at the dispatch center. They randomly review calls either real-time or almost real-time for quality improvement, aiming for at least 20% of OHCA calls. At the end of each shift, dispatchers are given feedback on the specific DA-CPR cases they handled. Monthly, there is a sharing session where dispatchers, nurses and an emergency physician come together to discuss the cases from the preceding month. Study population We analyzed OHCA cases where a call was received by SCDF call center that subsequently received an EMS diagnosis of cardiac arrest on the ambulance run sheet between July 2012 and March 2015. We excluded cases where audio recordings were incomplete (corrupted data) or of insufficient audio quality for analysis. We also excluded cases where bystander CPR was already ongoing at the time the call was received (Fig. 1). Data collection and processing Audio recordings of the calls received by the SCDF dispatch center for OHCA cases were retrospectively reviewed by four trained reviewers, each with nursing backgrounds. The reviewers are briefed on a set of parameters to gather. The recordings were first examined for whether bystander CPR was already in progress when the call was received (Fig. 1). For
A.F.W. Ho et al. / Resuscitation 105 (2016) 149–155
151
Total OHCA cases (July 2012 – Mar 2015)
4897
OHCA with BCPR 2237 (45.7%)
OHCA cases without BCPR 2660 (54.3%)
Dispatch recordings reviewed (Jul 2012 – Mar 2015)
1885
Unable to match with ambulance records 549 (29.1%)
Recordings matched with ambulance records 1336 (70.9%)
Excluded incomplete recordings 57 (3.0%) Complete recordings 1279 (67.8%) Excluded BCPR in progress 122 (6.5%) BCPR not in progress
1157 with 3 unknown (a)* Dispatcher did not recognize need for CPR 29 (2.5%) with 2 unknowns (b)*
Dispatcher recognized need for CPR 1128 (97.5%)
CPR instructions not given 72 (6.2%) Callers refused to do CPR before instructions given 24 (2.1%) Callers claimed to know CPR and declined instruction 7 (0.6%)
Stage 1: Recognizing need for CPR
CPR instructions given 1056 (91.3%)
Stage 2: Instructing CPR
Chest compressions not performed by caller 49 (4.2%) Chest compressions performed to instruction 1007 (87.0%)
Stage 3: Executing chest compression
Others (c)* 41 (3.5%) *Breakdown of ‘Unknowns’ are shown in Table 1
Fig. 1. Population flow diagram.
cases where bystander CPR was not already in progress, the reviewers then determined whether the dispatcher recognized the need for CPR (use the two-step question approach). For cases where the dispatcher recognized the need for CPR, the reviewers then determined whether CPR instructions were given. Where CPR instructions were given, the reviewers then determined whether chest compressions were performed. The reviewers recorded the time taken (measured from the start of the call) for dispatchers to recognize cardiac arrest, for CPR instructions to be delivered and for chest compressions to be started by the caller. They also identified barriers that delayed or prevented successful compressions, and classified them into one or more of nine pre-determined barrier types (Table 2), with the provision to enter any unexpected barriers as free-text. These nine types were derived based on a prior literature review and a focused group discussion with EMS physicians, paramedics and dispatchers. Ethics approval The Centralised Institutional Review Board (ref no: 2013/604/C) and Domain Specific Review Board (ref no: 2013/00929) granted approval for this study with a waiver of patient consent.
Statistical analysis Data entry was performed using a spreadsheet application (Excel 2003, Microsoft Corp., Redmond, WA) and data analysis using SPSS version 22 (SPSS, Chicago, IL) and R 3.2.2. Missing data were excluded from analysis. Categorical and continuous data were presented as frequency with percentage and median with interquartile range (IQR), respectively. Association between categorical variables were assessed using chi-square test, while association between continuous variables were assessed using Wilcox rank sum test. Results Study population During the study period, there were a total of 4897 OHCA cases for which 2237 (45.7%) had bystander CPR done. A total of 1885 dispatch recordings were retrieved but only 1336 (70.9%) recordings were successfully matched with ambulance records at the time of study. Fig. 1 shows the study flow diagram. 57 cases were excluded from analysis due to incomplete recording or poor audio quality. A further 122 cases were excluded as bystander CPR was already
152
A.F.W. Ho et al. / Resuscitation 105 (2016) 149–155
Table 1 Table showing reasons for unknowns and “other” reasons in Fig. 1.
delivered CPR instructions. Of these, in 1007 cases, chest compressions were performed by the caller. This meant that 1007 (87.0%) out of the 1157 recordings reviewed received DA-CPR (Table 3).
Unknown or others areas
Reasons
Unknown if CPR is in progress: Unknown (a)
2 Third party callers 1 Language barrier
Unknown if dispatch recognized the need for CPR: Unknown (b)
1 Third party caller 1 Unable to obtain further information due to emotional distress
Other reasons for CPR instructions not given (c)
9 No reason given 7 Call center called back but no one picked up 5 Ambulance arrived before instructions could be given 3 Caller left phone before instructions could be given 3 Body seemed cold or hard 3 Change in patient’s status 2 Caller hanged up 2 Third party caller 1 Language barrier 1 Terminally ill patient–daughter declined CPR
Barriers to chest compressions
n (%)
Couldn’t move patient Caller refused Hung up phone Patient’s status changed Caller left phone Overly distraught Caller not with patient Difficult access to patient Language barrier Others: Quality of instructions, Technical difficulties, Help arrived
118 (27) 64 (15) 47 (11) 31 (7) 30 (7) 23 (5) 20 (5) 16 (4) 15 (3) 117 (27)
Effect of encountering barriers during calls on reaching key stages In cases where one or more barriers were encountered, dispatchers were able to recognize the need for CPR less frequently, 94.9% vs. 99.0%, p < 0.001. For these same cases, CPR instructions were started less frequently, 79.3% vs. 98.3%, p < 0.001. For these same cases, chest compressions were eventually started less frequently, 67.9% vs. 98.3%, p < 0.001 (Table 3).
Table 2 Barriers to DACPR for 430 calls. Barriers to DACPR
One or more barriers to chest compressions eventually being started were identified in 430 (37.2%) cases. Table 2 shows the breakdown of the types of barriers identified. The most prevalent barrier identified was “could not move patient” which was present in 118 (27%) cases. This was followed by “Others”, 117 (27%) cases, “Caller refused”, 54 (15%) cases and “Hung up phone”, 47 (11%) cases. Free text entries under “Others” consisted of (1) unclear instructions, (2) technical difficulties and (3) help having arrived.
Effect of encountering barriers during call on time taken to reach key stages
ongoing at the time that the call was received. Consequently, 1157 cases qualified for analysis. Amongst these, in three cases it was unknown whether bystander CPR was already in progress. There were also two cases where it was unknown whether the dispatcher recognized the need for CPR. Table 1 shows the breakdown for the reasons that these cases were unknown. Delivery of CPR instructions and chest compressions In 1128 (97.5%) cases, the dispatcher recognized the need for CPR. Of these, in 1056 (91.3%) of cases the dispatcher successfully
In cases where one or more barriers were encountered, the time taken to reach key stages was significantly longer. The median time for the dispatcher to recognize the need for CPR was 60 s (interquartile range, IQR 43–93), compared to 68 s (IQR 48–114) in cases with barriers identified, p < 0.001. Median time for the dispatcher to begin CPR instructions was 131 s (IQR 100–175), compared to 159 s (IQR 110–219) in cases with barriers identified, p < 0.001. Median time to first chest compression was 200 s (IQR 160–254), compared to 240 s (IQR 180–330) in cases with barriers identified, p < 0.001 (Table 4). Discussion This study revealed that comprehensive implementation of DA-CPR in Singapore, including audio review of calls and giving feedback to dispatchers, resulted in relatively high proportions of OHCA cases being recognized by the dispatchers, high proportion
Table 3 Effect of encountering barriers during call on reaching key stages of dispatcher assisted cardiopulmonary resuscitation.
Recognized need for CPR? (%) CPR instructions started? (%) Chest compression started? (%)
Total (n = 1157)
No barriers (n = 727)
One or more barriers (n = 430)
P-value
1128 (97.5) 1056 (91.3) 1007 (87.0)
720 (99.0) 715 (98.3) 715 (98.3)
408 (94.9) 341 (79.3) 292 (67.9)
<0.001 <0.001 <0.001
Table 4 Comparison of time taken to reach key stages of dispatcher-assisted cardiopulmonary resuscitation between calls encountering barriers and calls that did not. Key stages
Time dispatcher recognizes need for CPR, secs (n = 1032) Time dispatcher begins instructions, secs (n = 711) Time to first compression, secs (n = 890)
All
No Barrier
Barrier
P-value
Median
Interquartile range
Median
Interquartile range
Median
Interquartile range
62 140 211
45–100 102–190 164–270
60 131 200
43–93 100–175 160–254
68 159 240
48–114 110–219 180–330
<0.001 <0.001 <0.001
A.F.W. Ho et al. / Resuscitation 105 (2016) 149–155
of cases where the dispatchers delivered CPR instructions and high proportion of callers commencing CPR based on these instructions. 87.0% of the cases analyzed in our study successfully received DA-CPR, a figure higher than that found in studies of other DACPR implementations. This may possibly be attributed to a rigorous QI process that was implemented in parallel with DA-CPR protocols. This involves having a dispatch nurse on shift at the dispatch center to perform review of recordings in real-time or almost realtime and feedback provided to dispatchers both immediately after the shift as well as during monthly sharing session. The monthly sharing sessions, attended by dispatchers, nurses as well as an emergency physician, allow for sharing of learning points, strategies for overcoming certain difficulties as well as refinement of the protocols. In a society where bystander CPR rate is low despite having a sizeable proportion of the population having been trained in CPR, DA-CPR is thought to be a cost-effective complement to public CPR efforts. The barriers to DA-CPR found in this study reveals specific barriers that impedes bystander CPR. It is of significance that amongst cases where one or more barriers were identified, CPR was still eventually commenced in 67.9%. While this proportion is lower than for those without barriers identified (and compressions also delayed), it suggests that DA-CPR may have encouraged the caller to do chest compressions through resolution of barriers. A review of 100 audio recordings of dispatch calls in Sweden found that agonal breathing patterns impeded the willingness of callers to follow CPR instructions.14 In another study, a review of 404 audio recordings in an American EMS system, bystander reporting signs of life was the most common barrier (51%) to the dispatchers not providing CPR instructions, while physical limitations of bystanders or patient habitus was the most common reason (20%) for callers to not follow CPR instructions when offered.15 We have similarly found that “couldn’t move patient” was a significant barrier in our population. Our study population did not detect bystanders reporting signs of life as a significant barrier, likely due to the low rate of cardiac arrests undetected by the dispatcher and also the design of our two-step question to interrogate for “abnormal breathing”. Also, in parallel to previous studies,16,17 we found that being emotionally distraught (5%) did not feature highly as a barrier to DA-CPR in our population. It is of interest that the barriers to DA-CPR appear to be quite different from that found for bystander CPR in general, such as fear of medicolegal consequences and concerns of disease transmission.18–20 These were not prominent in our study – the latter ostensibly because our DA-CPR protocol was chest-compression only. Also, when the recommendation to start CPR originates from the dispatcher who is seen as an authoritative figure, rather than the bystander, it possibly reduces the fear of medicolegal consequences or the fear of doing harm to the patient. DA-CPR consists of a chain of events that need to happen in quick succession to reduce delay to CPR. The finding that cases with no barriers identified took shorter time to reach key stages and eventual chest compression implies that addressing these barriers may improve the performance and survival advantage of DA-CPR. For the barriers that are potentially modifiable, such as “caller refused” and “hanged up phone”, improvement strategies could involve deliberately addressing concerns during public education efforts, as well as exploring specific techniques for dispatchers to handle such cases when the barrier is found to have arisen. Some barriers such as “couldn’t move patient” are not insurmountable but needs the development of dispatcher protocols through simulation and role-playing for such situations, to find ways to give unambiguous and succinct verbal instructions to safely move patients into a position appropriate for CPR.
153
This study suffered from several limitations. While efforts were made to standardize reviewers’ interpretation of barrier types through standardized data forms and briefings, inter-observer variability is difficult to eliminate. Another limitation lies in the inability to collect data on the demographics of the patient and the caller (e.g. age, ethnicity, gender), which may have revealed certain characteristics of the two which might affect the completion of DA-CPR. A third limitation is that the pre-determined categories of barriers may have led reviewers to neglect unexpected barrier types. This was reduced by allowing free-text entries for “other” barriers. A fourth limitation is the possible incompleteness of case capture. Two types of cases may not be featured in our data. (1) In cases where the caller called the police hotline instead of the ambulance hotline, the police transmits a text ambulance request to the dispatcher center. When the description in the request was not suspicious of cardiac arrest, the dispatchers might not have called back to provide DA-CPR. (2) In rare cases where DA-CPR was deemed to be futile by the dispatcher, such as falls from high-rise building with badly mangled bodies, or have potential legal implications, such as hanging, DA-CPR was not offered and the recording was not analyzed. These cases would have had successful DA-CPR and when not contributing to the denominator, possibly inflates the DA-CPR rate. A fifth limitation is the lack of blinding of audio recordings as the reviewers operate in the same facility as the dispatchers. The strength in this study lies in its ability to use reviews of audio recordings to capture qualitative data from an observer’s perspective. This is also to our knowledge the largest study that has sought to review audio recordings of DA-CPR dispatch calls. It also offers the first glimpse into the implementation of a comprehensive DA-CPR package in Singapore.
Conclusion A comprehensive implementation of DA-CPR including audio review resulted in dispatchers being able to recognize the need for CPR in most cases of OHCA. In most cases, the dispatchers provided CPR instructions and most of the callers followed the instructions to commence CPR. Common barriers to DA-CPR included “couldn’t move patient” and “caller refused”. Cases where one or more barriers were detected saw lower proportion of chest compressions eventually being performed, and for cases that did, took longer time to reach every key stage.
Conflict of interest statement A/Prof Ong has licensing agreement and patent filing (Application no: 13/047,348) with ZOLL Medical Corporation for a study titled ‘Method of predicting acute cardiopulmonary events and survivability of a patient’. No further conflict of interests for other authors.
Acknowledgements This study was supported by grants from National Medical Research Council, Clinician Scientist Award, Singapore (NMRC/ CSA/024/2010 and NMRC/CSA/0049/2013), Ministry of Health, Health Services Research Grant, Singapore (HSRG/0021/2012) and Duke-NUS Khoo Research Student Award (Duke-NUS-KRSA/ 2013/0001).
154
Appendix 1.
A.F.W. Ho et al. / Resuscitation 105 (2016) 149–155
A.F.W. Ho et al. / Resuscitation 105 (2016) 149–155
References 1. Shin S, Do Ong MEH, Tanaka H, et al. Comparison of emergency medical services systems across Pan-Asian countries: a Web-based survey. Prehosp Emerg Care 2012;16:477–96. 2. Lai H, Choong CV, Fook-Chong S, et al. Interventional strategies associated with improvements in survival for out-of-hospital cardiac arrests in Singapore over 10 years. Resuscitation 2015;89:155–61. 3. Cummins R, Ornato JP, Thies WH, Pepe PE. Improving survival from sudden cardiac arrest: the chain of survival concept. Circulation 1991;83:1832–47. 4. Cummins RO, Eisenberg MS, Hallstrom AP, Litwin PE. Survival of out-of-hospital cardiac arrest with early initiation of cardiopulmonary resuscitation. Am J Emerg Med 1985;3:114–9. 5. Rea TD, Eisenberg MS, Culley LL, Becker L. Dispatcher-assisted cardiopulmonary resuscitation and survival in cardiac arrest. Circulation 2001;104:2513–6. 6. Lerner EB, Rea TD, Bobrow BJ, et al. Emergency medical service dispatch cardiopulmonary resuscitation prearrival instructions to improve survival from out-of-hospital cardiac arrest: a scientific statement from the American Heart Association. Circulation 2012;125:648–55. 7. Rea TD. Temporal trends in sudden cardiac arrest: a 25-year emergency medical services perspective. Circulation 2003;107:2780–5. 8. Vaillancourt C, Verma A, Trickett J, et al. Evaluating the effectiveness of dispatch-assisted cardiopulmonary resuscitation instructions. Acad Emerg Med 2007;14:877–83. 9. Ong MEH, Quah JLJ, Ho AFW, et al. National population based survey on the prevalence of first aid, cardiopulmonary resuscitation and automated external defibrillator skills in Singapore. Resuscitation 2013;84:1633–6. 10. Ong ME, Cho J, Ma MHM, et al. Comparison of emergency medical services systems in the pan-Asian resuscitation outcomes study countries: report from a literature review and survey. Emerg Med Australas 2013;25:55–63.
155
11. Ong MEH, Shin S, Do Tanaka H, et al. Rationale, methodology, and implementation of a dispatcher-assisted cardiopulmonary resuscitation trial in the Asia-Pacific (Pan-Asian resuscitation outcomes study phase 2). Prehospital Emerg Care 2015;19:87–95. 12. Statistics Singapore – Department of Statistics Singapore. http://www.singstat. gov.sg/statistics/latest data.html#14 [accessed 12.15]. 13. Ho AFW, Chew D, Wong TH, et al. Prehospital trauma care in Singapore. Prehosp Emerg Care 2015;19:409–15. 14. Harjanto S, Na MX, Hao Y, et al. A before–after interventional trial of dispatcher-assisted cardio-pulmonary resuscitation for out-of-hospital cardiac arrests in Singapore. Resuscitation 2016;102:85–93, http://dx.doi.org/10.1016/ j.resuscitation.2016.02.014. 15. Bång A, Herlitz J, Martinell S. Interaction between emergency medical dispatcher and caller in suspected out-of-hospital cardiac arrest calls with focus on agonal breathing. A review of 100 tape recordings of true cardiac arrest cases. Resuscitation 2003;56:25–34. 16. Hauff SR, Rea TD, Culley LL, Kerry F, Becker L, Eisenberg MS. Factors impeding dispatcher-assisted telephone cardiopulmonary resuscitation. Ann Emerg Med 2003;42:731–7. 17. Bång A, Biber B, Isaksson L, Lindqvist J, Herlitz J. Evaluation of dispatcher-assisted cardiopulmonary resuscitation. Eur J Emerg Med 1999;6:175–83. 18. Meron G, Frantz O, Sterz F, Müllner M, Kaff A, Laggner AN. Analysing calls by lay persons reporting cardiac arrest. Resuscitation 1996;32:23–6. 19. Rowe BH, Shuster M, Zambon S, et al. Preparation, attitudes and behaviour in nonhospital cardiac emergencies: evaluating a community’s readiness to act. Can J Cardiol 1998;14:371–7. 20. Locke CJ, Berg RA, Sanders AB, et al. Bystander cardiopulmonary resuscitation. Concerns about mouth-to-mouth contact. Arch Intern Med 1995;155: 938–43.
Contents lists available at ScienceDirect
Resuscitation journal homepage: www.elsevier.com/locate/resuscitation
Clinical paper
Barriers to dispatcher-assisted cardiopulmonary resuscitation in Singapore夽 Andrew Fu Wah Ho a,∗ , Zariel Jiaying Sim b , Nur Shahidah c , Ying Hao d , Yih Yng Ng e , Benjamin S.H. Leong f , Siti Zarinah g , Winston K.L. Teo g , Geraldine Shu Yi Goh g , Hamizah Jaafar g , Marcus E.H. Ong c,h a
Emergency Medicine Residency, Singapore Health Services, Singapore, Singapore Barts and the London School of Medicine and Dentistry, Greater London, United Kingdom c Department of Emergency Medicine, Singapore General Hospital, Singapore, Singapore d Division of Medicine, Singapore General Hospital, Singapore, Singapore e Medical Department, Singapore Civil Defence Force, Singapore, Singapore f Emergency Medicine Department, National University Hospital, Singapore, Singapore g Operations Department, Singapore Civil Defence Force, Singapore, Singapore h Health Services & Systems Research, Duke-NUS Medical School, Singapore, Singapore b
a r t i c l e
i n f o
Article history: Received 5 February 2016 Received in revised form 31 March 2016 Accepted 9 May 2016 Keywords: Cardiac arrest Dispatcher-assisted Bystander Cardiopulmonary resuscitation Telephone CPR
a b s t r a c t Background: Dispatcher-assisted cardiopulmonary resuscitation (DA-CPR) is effective in increasing bystander CPR in out-of-hospital cardiac arrests (OHCA). Singapore has recently implemented a DACPR program. We aimed to characterize barriers to commencement of chest compressions by callers in Singapore. Methods: We analyzed dispatch recordings of OHCA cases received by the ambulance call center between July 2012 and March 2015. Audio recordings of poor quality were excluded. Trained reviewers noted the sequential stages of the dispatcher’s recognition of CPR, delivering CPR instructions and caller performing CPR. Time taken to reach these milestones was noted. Barriers to chest compressions were identified. Results: A total of 4897 OHCA occurred during the study period, overall bystander CPR rate was 45.7%. 1885 dispatch recordings were reviewed with 1157 cases qualified for dispatcher CPR. In 1128 (97.5%) cases, the dispatcher correctly recognized the need for CPR. CPR instructions were delivered in 1056 (91.3%) cases. Of these, 1007 (87.0%) callers performed CPR to instruction. One or more barriers to chest compressions were identified in 430 (37.2%) cases. The commonest barrier identified was “could not move patient” (27%). Cases where barriers were identified were less likely to have the need for CPR recognized by the dispatcher (94.9% vs. 99.0%, p < 0.001), CPR instructions given (79.3% vs. 98.3%, p < 0.001) and CPR started (67.9% vs. 98.3%, p < 0.001), while the time taken to reach each of these stages were significantly longer (p < 0.001). Conclusion: Barriers were present in 37% of cases. They were associated with lower proportion of CPR started and longer delay to CPR. © 2016 Elsevier Ireland Ltd. All rights reserved.
Introduction Out-of-hospital cardiac arrest (OHCA) is a leading cause of death in Singapore, mirroring worldwide trends. In the Asia-Pacific, OHCA
夽 A Spanish translated version of the abstract of this article appears as Appendix in the final online version at http://dx.doi.org/10.1016/j.resuscitation.2016.05.006. ∗ Corresponding author at: c/o Department of Emergency Medicine, Singapore General Hospital, Outram Road, Singapore 169608, Singapore. E-mail address: [email protected] (A.F.W. Ho). http://dx.doi.org/10.1016/j.resuscitation.2016.05.006 0300-9572/© 2016 Elsevier Ireland Ltd. All rights reserved.
survival rates are generally low, ranging from 2 to 11%,1 with the survival to hospital discharge rate in Singapore found to be 11.0% between 2010 and 2012.2 Effective treatment of OHCA hinges on Cummins’ “Chain of Survival” concept which describes the rapid commencement and seamless provision of a set of rescuer actions.3 One crucial link in the chain is early commencement of cardiopulmonary resuscitation (CPR), which improves outcomes in OHCA.4 As OHCAs generally occur away from the immediate presence of healthcare providers, early CPR depends heavily on the recognition of cardiac arrest by bystanders as well as bystander CPR, which can more than double the patient’s chance of survival.5
150
A.F.W. Ho et al. / Resuscitation 105 (2016) 149–155
In recent years, dispatcher-assisted CPR (DA-CPR) has emerged as an effective intervention to increase bystander CPR and hence OHCA outcomes.6 DA-CPR functions on the notion that when a bystander calls emergency medical services (EMS), not only does the call provide dispatch service, it also creates the opportunity for the EMS to catalyze recognition of cardiac arrest and initiation of bystander CPR through systematic interrogation and “just-in-time” pre-arrival instructions.6 Indeed, DA-CPR can nearly double the proportion of OHCA patients who receive bystander CPR.7,8 In Singapore, bystander CPR rate was low and remained relatively constant over the years, being 19.7% from 2001 to 2004 and 22.4% from 2010 to 2012.2 This was despite a relatively high proportion of the population having been trained in CPR, with a figure of 31.4% found in a national population based survey.9 DA-CPR was thought to be useful in such a setting to increase bystander CPR rate. We hypothesized that DA-CPR would be associated with a high rate of CPR performance by overcoming some of the barriers encountered by emergency callers. As part of the Pan-Asian Resuscitation Outcome Study Phase II (PAROS-II), Singapore implemented a DA-CPR package in 2012,10 which included a DA-CPR protocol, dispatcher training, systematic quality improvement (QI) through review of all dispatch calls and a public education campaign around DA-CPR. This study aimed to provide a look into the barriers to DA-CPR in Singapore, in order to identify system-dependent or cultural-based barriers. Such data may inform the implementation of DA-CPR in the other developing EMS systems. Materials and methods Setting Singapore is a highly-urbanized island city-state situated at the southernmost tip of the Malaysian peninsula with a population of 5.3 million over a land area of 715.1 square kilometers.12 EMS is provided by the Singapore Civil Defence Force (SCDF), which operates a fleet of 46 ambulances and 15 first responders on motorcycles in a fire-based system activated by a centralized 9-9-5 dispatching system.12 SCDF utilizes computer aided dispatch protocols, global positioning satellite automatic vehicle location systems and road traffic monitoring systems. SCDF paramedics (equivalent to North American emergency medical technician-intermediate) are trained in basic life support, automated external defibrillator usage and specific interventions including adrenaline administration. Mean time taken from call to arrival at scene was 9.9 min from 2010 to 2012.2 EMS for a minority of emergency cases are provided by private operators that charge a fee whereby the caller can specify the hospital of conveyance. These are mainly used for inter-facility transfers and do not respond to ‘995’. The emergency ambulance services provided by calling 995 is free of charge to all users and supported through government taxes.12 Dispatcher-assisted cardiopulmonary resuscitation in Singapore Singapore adopted a DA-CPR protocol in 201113 ; however, due to the lack of systems for training and QI, CPR instructions were only given in less than 5% of cases. As part of the Pan-Asian Resuscitation Outcome Study Phase II (PAROS-II), Singapore was selected for implementation of a comprehensive DA-CPR package. This package consisted of (1) a standardized dispatch protocol, (2) a standardized training package for dispatchers, (3) a measurement tool to collect data on individual dispatcher and organization-level performance, (4) an integrated QI program and (5) a community education program.11 The program was implemented in July 2012
with a planned 6 month “run-in” period. By Jan 2013, all dispatchers were able to provide DA-CPR. This package consisted of11 1. A standardized dispatch protocol that guides dispatchers to systematically question callers to accurately and rapidly determine whether the patient is in cardiac arrest. When a cardiac arrest patient is identified, the protocol guides the dispatcher to give CPR instructions to assist bystanders if CPR is not already ongoing. 2. A training package consisting of 1 day intensive training course both call-takers and dispatch center managers. It comprises didactic teaching as well as practical exercises and scenarios for dispatchers as well as supervisors/medical directors. 3. A standardized measurement QI tool to collect data on individual dispatcher and organizational level performance through review of the dispatch audio recordings. A sample of the QI tool is shown in Appendix 1. 4. An integrated quality improvement program that includes cooperation and collaboration of pre-hospital and hospital stakeholders. The program provides feedback at the individual and organizational level. It involved review of dispatcher CPR audio recordings and use of the QI tool mentioned in step 3. 5. A community education program known as Dispatcher Assisted First REsponder (DARE) program was developed to update the public on DA-CPR, how to recognize cardiac arrest, the importance of early activation of EMS and staying on the line to perform DACPR. In SCDF, dispatchers are trained to adopt a streamlined, two-step question approach to identify the need for CPR. The caller is asked (1) “Is the victim conscious?” and (2) “Is the victim breathing normally?” If a negative response to both questions are received, the presumptive diagnosis is cardiac arrest and the dispatcher enters into a protocol to deliver CPR instructions.14 Instructions for chest compression-only CPR are given, unless there are specific indications for telephone instructions on ventilations, like drowning or pediatric cardiac arrest. The QI process involves having a team of four dispatch nurses review recordings of cardiac arrest calls every shift at the dispatch center. They randomly review calls either real-time or almost real-time for quality improvement, aiming for at least 20% of OHCA calls. At the end of each shift, dispatchers are given feedback on the specific DA-CPR cases they handled. Monthly, there is a sharing session where dispatchers, nurses and an emergency physician come together to discuss the cases from the preceding month. Study population We analyzed OHCA cases where a call was received by SCDF call center that subsequently received an EMS diagnosis of cardiac arrest on the ambulance run sheet between July 2012 and March 2015. We excluded cases where audio recordings were incomplete (corrupted data) or of insufficient audio quality for analysis. We also excluded cases where bystander CPR was already ongoing at the time the call was received (Fig. 1). Data collection and processing Audio recordings of the calls received by the SCDF dispatch center for OHCA cases were retrospectively reviewed by four trained reviewers, each with nursing backgrounds. The reviewers are briefed on a set of parameters to gather. The recordings were first examined for whether bystander CPR was already in progress when the call was received (Fig. 1). For
A.F.W. Ho et al. / Resuscitation 105 (2016) 149–155
151
Total OHCA cases (July 2012 – Mar 2015)
4897
OHCA with BCPR 2237 (45.7%)
OHCA cases without BCPR 2660 (54.3%)
Dispatch recordings reviewed (Jul 2012 – Mar 2015)
1885
Unable to match with ambulance records 549 (29.1%)
Recordings matched with ambulance records 1336 (70.9%)
Excluded incomplete recordings 57 (3.0%) Complete recordings 1279 (67.8%) Excluded BCPR in progress 122 (6.5%) BCPR not in progress
1157 with 3 unknown (a)* Dispatcher did not recognize need for CPR 29 (2.5%) with 2 unknowns (b)*
Dispatcher recognized need for CPR 1128 (97.5%)
CPR instructions not given 72 (6.2%) Callers refused to do CPR before instructions given 24 (2.1%) Callers claimed to know CPR and declined instruction 7 (0.6%)
Stage 1: Recognizing need for CPR
CPR instructions given 1056 (91.3%)
Stage 2: Instructing CPR
Chest compressions not performed by caller 49 (4.2%) Chest compressions performed to instruction 1007 (87.0%)
Stage 3: Executing chest compression
Others (c)* 41 (3.5%) *Breakdown of ‘Unknowns’ are shown in Table 1
Fig. 1. Population flow diagram.
cases where bystander CPR was not already in progress, the reviewers then determined whether the dispatcher recognized the need for CPR (use the two-step question approach). For cases where the dispatcher recognized the need for CPR, the reviewers then determined whether CPR instructions were given. Where CPR instructions were given, the reviewers then determined whether chest compressions were performed. The reviewers recorded the time taken (measured from the start of the call) for dispatchers to recognize cardiac arrest, for CPR instructions to be delivered and for chest compressions to be started by the caller. They also identified barriers that delayed or prevented successful compressions, and classified them into one or more of nine pre-determined barrier types (Table 2), with the provision to enter any unexpected barriers as free-text. These nine types were derived based on a prior literature review and a focused group discussion with EMS physicians, paramedics and dispatchers. Ethics approval The Centralised Institutional Review Board (ref no: 2013/604/C) and Domain Specific Review Board (ref no: 2013/00929) granted approval for this study with a waiver of patient consent.
Statistical analysis Data entry was performed using a spreadsheet application (Excel 2003, Microsoft Corp., Redmond, WA) and data analysis using SPSS version 22 (SPSS, Chicago, IL) and R 3.2.2. Missing data were excluded from analysis. Categorical and continuous data were presented as frequency with percentage and median with interquartile range (IQR), respectively. Association between categorical variables were assessed using chi-square test, while association between continuous variables were assessed using Wilcox rank sum test. Results Study population During the study period, there were a total of 4897 OHCA cases for which 2237 (45.7%) had bystander CPR done. A total of 1885 dispatch recordings were retrieved but only 1336 (70.9%) recordings were successfully matched with ambulance records at the time of study. Fig. 1 shows the study flow diagram. 57 cases were excluded from analysis due to incomplete recording or poor audio quality. A further 122 cases were excluded as bystander CPR was already
152
A.F.W. Ho et al. / Resuscitation 105 (2016) 149–155
Table 1 Table showing reasons for unknowns and “other” reasons in Fig. 1.
delivered CPR instructions. Of these, in 1007 cases, chest compressions were performed by the caller. This meant that 1007 (87.0%) out of the 1157 recordings reviewed received DA-CPR (Table 3).
Unknown or others areas
Reasons
Unknown if CPR is in progress: Unknown (a)
2 Third party callers 1 Language barrier
Unknown if dispatch recognized the need for CPR: Unknown (b)
1 Third party caller 1 Unable to obtain further information due to emotional distress
Other reasons for CPR instructions not given (c)
9 No reason given 7 Call center called back but no one picked up 5 Ambulance arrived before instructions could be given 3 Caller left phone before instructions could be given 3 Body seemed cold or hard 3 Change in patient’s status 2 Caller hanged up 2 Third party caller 1 Language barrier 1 Terminally ill patient–daughter declined CPR
Barriers to chest compressions
n (%)
Couldn’t move patient Caller refused Hung up phone Patient’s status changed Caller left phone Overly distraught Caller not with patient Difficult access to patient Language barrier Others: Quality of instructions, Technical difficulties, Help arrived
118 (27) 64 (15) 47 (11) 31 (7) 30 (7) 23 (5) 20 (5) 16 (4) 15 (3) 117 (27)
Effect of encountering barriers during calls on reaching key stages In cases where one or more barriers were encountered, dispatchers were able to recognize the need for CPR less frequently, 94.9% vs. 99.0%, p < 0.001. For these same cases, CPR instructions were started less frequently, 79.3% vs. 98.3%, p < 0.001. For these same cases, chest compressions were eventually started less frequently, 67.9% vs. 98.3%, p < 0.001 (Table 3).
Table 2 Barriers to DACPR for 430 calls. Barriers to DACPR
One or more barriers to chest compressions eventually being started were identified in 430 (37.2%) cases. Table 2 shows the breakdown of the types of barriers identified. The most prevalent barrier identified was “could not move patient” which was present in 118 (27%) cases. This was followed by “Others”, 117 (27%) cases, “Caller refused”, 54 (15%) cases and “Hung up phone”, 47 (11%) cases. Free text entries under “Others” consisted of (1) unclear instructions, (2) technical difficulties and (3) help having arrived.
Effect of encountering barriers during call on time taken to reach key stages
ongoing at the time that the call was received. Consequently, 1157 cases qualified for analysis. Amongst these, in three cases it was unknown whether bystander CPR was already in progress. There were also two cases where it was unknown whether the dispatcher recognized the need for CPR. Table 1 shows the breakdown for the reasons that these cases were unknown. Delivery of CPR instructions and chest compressions In 1128 (97.5%) cases, the dispatcher recognized the need for CPR. Of these, in 1056 (91.3%) of cases the dispatcher successfully
In cases where one or more barriers were encountered, the time taken to reach key stages was significantly longer. The median time for the dispatcher to recognize the need for CPR was 60 s (interquartile range, IQR 43–93), compared to 68 s (IQR 48–114) in cases with barriers identified, p < 0.001. Median time for the dispatcher to begin CPR instructions was 131 s (IQR 100–175), compared to 159 s (IQR 110–219) in cases with barriers identified, p < 0.001. Median time to first chest compression was 200 s (IQR 160–254), compared to 240 s (IQR 180–330) in cases with barriers identified, p < 0.001 (Table 4). Discussion This study revealed that comprehensive implementation of DA-CPR in Singapore, including audio review of calls and giving feedback to dispatchers, resulted in relatively high proportions of OHCA cases being recognized by the dispatchers, high proportion
Table 3 Effect of encountering barriers during call on reaching key stages of dispatcher assisted cardiopulmonary resuscitation.
Recognized need for CPR? (%) CPR instructions started? (%) Chest compression started? (%)
Total (n = 1157)
No barriers (n = 727)
One or more barriers (n = 430)
P-value
1128 (97.5) 1056 (91.3) 1007 (87.0)
720 (99.0) 715 (98.3) 715 (98.3)
408 (94.9) 341 (79.3) 292 (67.9)
<0.001 <0.001 <0.001
Table 4 Comparison of time taken to reach key stages of dispatcher-assisted cardiopulmonary resuscitation between calls encountering barriers and calls that did not. Key stages
Time dispatcher recognizes need for CPR, secs (n = 1032) Time dispatcher begins instructions, secs (n = 711) Time to first compression, secs (n = 890)
All
No Barrier
Barrier
P-value
Median
Interquartile range
Median
Interquartile range
Median
Interquartile range
62 140 211
45–100 102–190 164–270
60 131 200
43–93 100–175 160–254
68 159 240
48–114 110–219 180–330
<0.001 <0.001 <0.001
A.F.W. Ho et al. / Resuscitation 105 (2016) 149–155
of cases where the dispatchers delivered CPR instructions and high proportion of callers commencing CPR based on these instructions. 87.0% of the cases analyzed in our study successfully received DA-CPR, a figure higher than that found in studies of other DACPR implementations. This may possibly be attributed to a rigorous QI process that was implemented in parallel with DA-CPR protocols. This involves having a dispatch nurse on shift at the dispatch center to perform review of recordings in real-time or almost realtime and feedback provided to dispatchers both immediately after the shift as well as during monthly sharing session. The monthly sharing sessions, attended by dispatchers, nurses as well as an emergency physician, allow for sharing of learning points, strategies for overcoming certain difficulties as well as refinement of the protocols. In a society where bystander CPR rate is low despite having a sizeable proportion of the population having been trained in CPR, DA-CPR is thought to be a cost-effective complement to public CPR efforts. The barriers to DA-CPR found in this study reveals specific barriers that impedes bystander CPR. It is of significance that amongst cases where one or more barriers were identified, CPR was still eventually commenced in 67.9%. While this proportion is lower than for those without barriers identified (and compressions also delayed), it suggests that DA-CPR may have encouraged the caller to do chest compressions through resolution of barriers. A review of 100 audio recordings of dispatch calls in Sweden found that agonal breathing patterns impeded the willingness of callers to follow CPR instructions.14 In another study, a review of 404 audio recordings in an American EMS system, bystander reporting signs of life was the most common barrier (51%) to the dispatchers not providing CPR instructions, while physical limitations of bystanders or patient habitus was the most common reason (20%) for callers to not follow CPR instructions when offered.15 We have similarly found that “couldn’t move patient” was a significant barrier in our population. Our study population did not detect bystanders reporting signs of life as a significant barrier, likely due to the low rate of cardiac arrests undetected by the dispatcher and also the design of our two-step question to interrogate for “abnormal breathing”. Also, in parallel to previous studies,16,17 we found that being emotionally distraught (5%) did not feature highly as a barrier to DA-CPR in our population. It is of interest that the barriers to DA-CPR appear to be quite different from that found for bystander CPR in general, such as fear of medicolegal consequences and concerns of disease transmission.18–20 These were not prominent in our study – the latter ostensibly because our DA-CPR protocol was chest-compression only. Also, when the recommendation to start CPR originates from the dispatcher who is seen as an authoritative figure, rather than the bystander, it possibly reduces the fear of medicolegal consequences or the fear of doing harm to the patient. DA-CPR consists of a chain of events that need to happen in quick succession to reduce delay to CPR. The finding that cases with no barriers identified took shorter time to reach key stages and eventual chest compression implies that addressing these barriers may improve the performance and survival advantage of DA-CPR. For the barriers that are potentially modifiable, such as “caller refused” and “hanged up phone”, improvement strategies could involve deliberately addressing concerns during public education efforts, as well as exploring specific techniques for dispatchers to handle such cases when the barrier is found to have arisen. Some barriers such as “couldn’t move patient” are not insurmountable but needs the development of dispatcher protocols through simulation and role-playing for such situations, to find ways to give unambiguous and succinct verbal instructions to safely move patients into a position appropriate for CPR.
153
This study suffered from several limitations. While efforts were made to standardize reviewers’ interpretation of barrier types through standardized data forms and briefings, inter-observer variability is difficult to eliminate. Another limitation lies in the inability to collect data on the demographics of the patient and the caller (e.g. age, ethnicity, gender), which may have revealed certain characteristics of the two which might affect the completion of DA-CPR. A third limitation is that the pre-determined categories of barriers may have led reviewers to neglect unexpected barrier types. This was reduced by allowing free-text entries for “other” barriers. A fourth limitation is the possible incompleteness of case capture. Two types of cases may not be featured in our data. (1) In cases where the caller called the police hotline instead of the ambulance hotline, the police transmits a text ambulance request to the dispatcher center. When the description in the request was not suspicious of cardiac arrest, the dispatchers might not have called back to provide DA-CPR. (2) In rare cases where DA-CPR was deemed to be futile by the dispatcher, such as falls from high-rise building with badly mangled bodies, or have potential legal implications, such as hanging, DA-CPR was not offered and the recording was not analyzed. These cases would have had successful DA-CPR and when not contributing to the denominator, possibly inflates the DA-CPR rate. A fifth limitation is the lack of blinding of audio recordings as the reviewers operate in the same facility as the dispatchers. The strength in this study lies in its ability to use reviews of audio recordings to capture qualitative data from an observer’s perspective. This is also to our knowledge the largest study that has sought to review audio recordings of DA-CPR dispatch calls. It also offers the first glimpse into the implementation of a comprehensive DA-CPR package in Singapore.
Conclusion A comprehensive implementation of DA-CPR including audio review resulted in dispatchers being able to recognize the need for CPR in most cases of OHCA. In most cases, the dispatchers provided CPR instructions and most of the callers followed the instructions to commence CPR. Common barriers to DA-CPR included “couldn’t move patient” and “caller refused”. Cases where one or more barriers were detected saw lower proportion of chest compressions eventually being performed, and for cases that did, took longer time to reach every key stage.
Conflict of interest statement A/Prof Ong has licensing agreement and patent filing (Application no: 13/047,348) with ZOLL Medical Corporation for a study titled ‘Method of predicting acute cardiopulmonary events and survivability of a patient’. No further conflict of interests for other authors.
Acknowledgements This study was supported by grants from National Medical Research Council, Clinician Scientist Award, Singapore (NMRC/ CSA/024/2010 and NMRC/CSA/0049/2013), Ministry of Health, Health Services Research Grant, Singapore (HSRG/0021/2012) and Duke-NUS Khoo Research Student Award (Duke-NUS-KRSA/ 2013/0001).
154
Appendix 1.
A.F.W. Ho et al. / Resuscitation 105 (2016) 149–155
A.F.W. Ho et al. / Resuscitation 105 (2016) 149–155
References 1. Shin S, Do Ong MEH, Tanaka H, et al. Comparison of emergency medical services systems across Pan-Asian countries: a Web-based survey. Prehosp Emerg Care 2012;16:477–96. 2. Lai H, Choong CV, Fook-Chong S, et al. Interventional strategies associated with improvements in survival for out-of-hospital cardiac arrests in Singapore over 10 years. Resuscitation 2015;89:155–61. 3. Cummins R, Ornato JP, Thies WH, Pepe PE. Improving survival from sudden cardiac arrest: the chain of survival concept. Circulation 1991;83:1832–47. 4. Cummins RO, Eisenberg MS, Hallstrom AP, Litwin PE. Survival of out-of-hospital cardiac arrest with early initiation of cardiopulmonary resuscitation. Am J Emerg Med 1985;3:114–9. 5. Rea TD, Eisenberg MS, Culley LL, Becker L. Dispatcher-assisted cardiopulmonary resuscitation and survival in cardiac arrest. Circulation 2001;104:2513–6. 6. Lerner EB, Rea TD, Bobrow BJ, et al. Emergency medical service dispatch cardiopulmonary resuscitation prearrival instructions to improve survival from out-of-hospital cardiac arrest: a scientific statement from the American Heart Association. Circulation 2012;125:648–55. 7. Rea TD. Temporal trends in sudden cardiac arrest: a 25-year emergency medical services perspective. Circulation 2003;107:2780–5. 8. Vaillancourt C, Verma A, Trickett J, et al. Evaluating the effectiveness of dispatch-assisted cardiopulmonary resuscitation instructions. Acad Emerg Med 2007;14:877–83. 9. Ong MEH, Quah JLJ, Ho AFW, et al. National population based survey on the prevalence of first aid, cardiopulmonary resuscitation and automated external defibrillator skills in Singapore. Resuscitation 2013;84:1633–6. 10. Ong ME, Cho J, Ma MHM, et al. Comparison of emergency medical services systems in the pan-Asian resuscitation outcomes study countries: report from a literature review and survey. Emerg Med Australas 2013;25:55–63.
155
11. Ong MEH, Shin S, Do Tanaka H, et al. Rationale, methodology, and implementation of a dispatcher-assisted cardiopulmonary resuscitation trial in the Asia-Pacific (Pan-Asian resuscitation outcomes study phase 2). Prehospital Emerg Care 2015;19:87–95. 12. Statistics Singapore – Department of Statistics Singapore. http://www.singstat. gov.sg/statistics/latest data.html#14 [accessed 12.15]. 13. Ho AFW, Chew D, Wong TH, et al. Prehospital trauma care in Singapore. Prehosp Emerg Care 2015;19:409–15. 14. Harjanto S, Na MX, Hao Y, et al. A before–after interventional trial of dispatcher-assisted cardio-pulmonary resuscitation for out-of-hospital cardiac arrests in Singapore. Resuscitation 2016;102:85–93, http://dx.doi.org/10.1016/ j.resuscitation.2016.02.014. 15. Bång A, Herlitz J, Martinell S. Interaction between emergency medical dispatcher and caller in suspected out-of-hospital cardiac arrest calls with focus on agonal breathing. A review of 100 tape recordings of true cardiac arrest cases. Resuscitation 2003;56:25–34. 16. Hauff SR, Rea TD, Culley LL, Kerry F, Becker L, Eisenberg MS. Factors impeding dispatcher-assisted telephone cardiopulmonary resuscitation. Ann Emerg Med 2003;42:731–7. 17. Bång A, Biber B, Isaksson L, Lindqvist J, Herlitz J. Evaluation of dispatcher-assisted cardiopulmonary resuscitation. Eur J Emerg Med 1999;6:175–83. 18. Meron G, Frantz O, Sterz F, Müllner M, Kaff A, Laggner AN. Analysing calls by lay persons reporting cardiac arrest. Resuscitation 1996;32:23–6. 19. Rowe BH, Shuster M, Zambon S, et al. Preparation, attitudes and behaviour in nonhospital cardiac emergencies: evaluating a community’s readiness to act. Can J Cardiol 1998;14:371–7. 20. Locke CJ, Berg RA, Sanders AB, et al. Bystander cardiopulmonary resuscitation. Concerns about mouth-to-mouth contact. Arch Intern Med 1995;155: 938–43.