- Email: [email protected]
Factors associated with triage assignment of emergency department patients ultimately diagnosed with acute myocardial infarction
G Model AUCC-277; No. of Pages 4
ARTICLE IN PRESS Australian Critical Care xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Australian Critical Care journal homepage: www.elsevier.com/locate/aucc
Research paper
Factors associated with triage assignment of emergency department patients ultimately diagnosed with acute myocardial infarction Kimberley Ryan BN, GradDip Nurs (Crit Care), BHSc (Nat) a,∗ , Jaimi Greenslade PhD a,b,c , Emily Dalton BSc, BNursing a , Kevin Chu MBBS, MSc, FACEM a,b , Anthony F.T. Brown MBChB, FRCP, FRCSEd, FACEM, FCEM a,b , Louise Cullen MBBS, FACEM a,b,c a b c
Royal Brisbane and Women’s Hospital, Department of Emergency Medicine, Brisbane 4006, Australia School of Medicine, The University of Queensland, St Lucia, 4067, Australia School of Public Health, Queensland University of Technology, Kelvin Grove, 4059, Australia
article information Article history: Received 19 October 2014 Received in revised form 27 April 2015 Accepted 4 May 2015 Available online xxx Keywords: Chest pain Emergency department Emergency nursing Myocardial infarction Triage
a b s t r a c t Background: The objective of this study was to explore factors associated with the triage category assigned by the triage nurse for patients ultimately diagnosed with acute myocardial infarction. Methods: This was a retrospective analysis of 12 months of data, on adult emergency department patients ultimately diagnosed with acute myocardial infarction. Data were obtained from hospital databases and included patient demographics, patient clinical characteristics and nurses’ experience. Results: Of the 153 patients, 20% (95% CI: 14–27%) were given a lower urgency triage category than recommended by international guidelines. Compared to patients who were triaged Australasian Triage Category 1 or 2, patients with an Australasian Triage Category 3–5 were older (mean age 76 versus 68 years), more likely to be female (63% versus 32%), more likely to present without chest pain (93% versus 35%) and less likely to have a cardiac history (3.3% versus 17.9%). A slightly higher proportion of patients Australasian Triage Category 3–5 were triaged by an experienced nurse (50%) compared to patients categorised Australasian Triage Category 1–2 (35.2%) but this finding did not reach statistical significance. Conclusions: One in five presentations was given a lower urgency triage category than recommended by international guidelines, potentially leading to delays in medical treatment. The absence of chest pain was the defining characteristic in this group of patients, along with other factors identified by previous research such as being of female sex and elderly. © 2015 Published by Elsevier Ltd. on behalf of Australian College of Critical Care Nurses Ltd.
1. Introduction Acute Myocardial Infarction (AMI) is a leading cause of premature death and disability for Australian men and women.1 Rapid assessment and treatment of patients with AMI is essential as mortality associated with AMI is directly linked to time taken to receive treatment.2,3 The initial clinical assessment for patients
∗ Corresponding author at: Royal Brisbane and Women’s Hospital, Department of Emergency Medicine, Butterfield Street, Herston, 4006 Queensland, Australia. Tel.: +61 7 36464629; fax: +61 7 3646 8732. E-mail address: [email protected] (K. Ryan).
who present to hospital emergency department (ED) occurs by a triage officer. In most Western countries, the triage officer is a registered nurse who has specialised in emergency nursing. The key role of the triage nurse is to accurately identify potential patients who may have an AMI as early as possible to expedite necessary immediate cardiac care.4 Triage is a system which allows the clinical urgency of a presenting problem to be categorised. Similar 5-level triage systems are used in Canada (the Canada Triage and Acuity Scale),5 Europe, the United Kingdom (the Manchester Triage scale) and Australia (the Australasian Triage Scale (ATS))6,7 (Web Appendix 1). In Australia, experienced ED nurses complete mandatory training from the Emergency Triage Education Kit (ETEK) and must be well
http://dx.doi.org/10.1016/j.aucc.2015.05.001 1036-7314/© 2015 Published by Elsevier Ltd. on behalf of Australian College of Critical Care Nurses Ltd.
Please cite this article in press as: Ryan K, et al. Factors associated with triage assignment of emergency department patients ultimately diagnosed with acute myocardial infarction. Aust Crit Care (2015), http://dx.doi.org/10.1016/j.aucc.2015.05.001
G Model AUCC-277; No. of Pages 4
ARTICLE IN PRESS K. Ryan et al. / Australian Critical Care xxx (2015) xxx–xxx
2
versed with the ATS. The ATS ranges from 1 to 5 with the ranking correlating with the recommended maximum time a patient should wait for treatment. Patients categorised as ATS 1 require immediate treatment, while patients categorised as ATS 2, 3, 4 or 5 are expected to receive medical assessment and treatment within 10, 30, 60, and 120 min respectively.8 In line with the Australasian College of Emergency Medicine guidelines and the ETEK, a patient presenting with symptoms suggestive of Acute Coronary Syndrome (ACS) should be triaged as a Category 2.9,10 These symptoms may include acute chest, epigastric, neck, jaw, or arm pain; or discomfort or pressure without an apparent non cardiac source.11 According to the ETEK, ‘under-triage’ describes the process whereby the patient receives a triage code lower than their actual level of urgency.10 Although Australian data are lacking, international research suggests a significant number of patients with AMIs are under-triaged, with one study finding up to half of all AMI patients being assigned a lower priority triage category on presentation.5 The reasons for under-triage of AMI patients have not been adequately elucidated, though research suggests factors including age, sex, an absence of chest pain at triage, and history of diabetes mellitus or heart failure make AMI harder to recognise.12–14 Given that triage is a modifiable factor influencing delays to treatment, further understanding of the determinants of triage category in an Australian setting is an essential step in enhancing the triage process. 1.1. Purpose of the study This study explored the triage category assigned to patients with myocardial infarction in a large tertiary hospital ED in Queensland, Australia. The study aim was to identify the factors associated with the triage of patients presenting to ED with AMI. Predictors of interest included patient demographics, clinical characteristics and nursing triage experience. 2. Patients and methods 2.1. Study design and setting This was an analysis of retrospective collected data on adult patients presenting to the ED who were ultimately diagnosed with AMI. The data were collected from the Royal Brisbane and Women’s Hospital (RBWH) between 1 June 2009 and 31 May 2010. The RBWH is a 929 bed adult tertiary-referral teaching hospital; the RBWH ED has an annual attendance rate of 72,000 patients over the age of 14 years. This paper reports the findings of a research study that adhered to the National Statement on the Conduct of Human Research by the Australian National Health and Medical Research Council, and has been approved by the RBWH Human Research Ethics Committee on the 30th April 2010. 2.2. Case selection and data collection Patients with an AMI were identified through several sequential steps. The RBWH pathology department provided a list of all patients who presented to the ED and had a serum troponin I (TNI) performed as part of their emergency workup. This list was then refined to include only those patients with a TNI value of ≥0.06 mcg/L. The TNI assay used in at the RBWH during the study period was the Beckman Coulter AccuTnI assay and 0.06 was the clinical decision cut-off point. The next step was to undertake a review of the discharge diagnosis on the Emergency Department Information System (EDIS) to confirm a diagnosis of AMI for those patients. Patients with an EDIS diagnosis of ‘Chest Pain’ or ‘Acute Coronary Syndrome’ (ACS) were further examined to identify whether they had a diagnosis of AMI on the
Table 1 Baseline characteristics of the cohort (n = 153). Characteristic
n (%)
Mean ± SD age (years) Male sex English as primary language Employment status Employed Pensioner Unemployed Other/unknown Mode of arrival to emergency department Own transport Ambulance service
69.5 ± 14.4 94 (61.4) 144 (94.1) 42 (27.5) 82 (53.6) 12 (7.8) 17 (11.1) 39 (25.5) 114 (74.5)
index admission. This initial examination was undertaken using charts and electronic medical discharge summaries where the diagnosis was documented by a cardiologist or general physician. For the purposes of this study, AMI included diagnosis of AMI (NSTEMI or STEMI) on the index admission or urgent revascularisation on index admission including coronary angioplasty, coronary artery stenting and coronary artery bypass grafting. Exclusion criteria included pregnancy, age < 18 years and any patients transferred to the ED from another hospital. Once the population was identified, study data were obtained from a number of sources. Data on patient demographics (age and sex), presenting symptoms, cardiac history, ambulance use and triage category were collected from the EDIS database. Presenting symptoms were categorised as typical or atypical for AMI. Typical symptoms referred to the presence of any chest pain during the event (present or resolved on arrival), while atypical symptoms included dizziness, syncope, nausea or vomiting and dyspnoea with the absence of chest pain prior to arrival or during presentation.15–17 The name of the nurse who triaged the patient was provided from the ED admission system, and the years of nursing triage experience was then sought from the ED Nurse Educator’s records. Data was de-identified prior to analysis. 2.3. Data analysis Data were analysed using SPSS version 20. Baseline characteristics of the sample were reported. A triage category of 1 or 2 was categorised as high urgency while a triage category of 3–5 was deemed lower urgency. Standard descriptive statistics were used to report the characteristics of the correctly triaged and under-triaged patient groups. Chi-square tests (or Fisher’s exact tests where cell sizes were small) were performed to compare dichotomous data across triage categories. T-tests were performed to compare continuous characteristics across triage categories. There only were a small number of patients with an ATS 3–5 (n = 30) and so it was not deemed appropriate to perform multivariable analyses to identify the independent predictors of triage category. 3. Results There were a total of 153 patients identified with an index AMI for analysis. The sample included 94 (61.4%) males and the mean age was 69.5 years (SD = 14.1 years). Baseline characteristics of the cohort are provided in Table 1 and demonstrate that the majority of the patients were English-speaking pensioners who arrived via ambulance. One hundred and twenty three (80.4%, 95% CI: 73.2–86.4%) patients were provided an appropriate triage category (ATS 1–2).
Please cite this article in press as: Ryan K, et al. Factors associated with triage assignment of emergency department patients ultimately diagnosed with acute myocardial infarction. Aust Crit Care (2015), http://dx.doi.org/10.1016/j.aucc.2015.05.001
G Model AUCC-277; No. of Pages 4
ARTICLE IN PRESS K. Ryan et al. / Australian Critical Care xxx (2015) xxx–xxx
3
Table 2 Patient characteristics and nursing triage experience for patients categorised as ATS 1–2 (n = 123) and ATS 3–5 (n = 30).
Male Mean (SD) patient age, years Arrival by ambulance Presence of chest pain Cardiac history Nursing triage experience >2 years
Triage category 1–2 (n = 123)
Triage category 3–5 (n = 30)
p
83 (67.5%) 67.79 (13.59) 93 (75.6%) 79 (64.2%) 22 (17.9%) 43 (35.2%)
11 (36.7%) 76.33 (14.03) 21 (70%) 2 (6.7%) 1 (3.3%) 15 (50%)
<0.01 <0.01 0.53 <0.01 0.05 0.14
Data are number (%) unless otherwise indicated.
Table 2 provides nurse and patient characteristics for patients who were categorised ATS 1–2 versus ATS 3–5. At a univariate level, females and older individuals were less often assigned a triage category 1–2. Females were older (mean = 76.56 years, SD = 14.10) than males (mean = 65.01 years, SD = 12.12). However, the finding that triage category 1–2 were younger was true both for males and females (p = 0.31) Patients with a cardiac history and patients with chest pain were more often assigned a triage category 1–2. A slightly higher proportion of under-triaged patients were triaged by an experienced nurse (50%) compared to correctly triaged patients (35.2%) but this finding did not reach statistical significance. 4. Discussion Though the ETEK was published to help support consistency in triage practice across Australia, the implementation of it as a mandatory learning tool was not initiated in the RBWH DEM until the end of 2010. Thus the collection of data in this study pre-dates the implementation of ETEK; rather the ATS guided triage at this time. The appropriate triage of AMI patients to an ATS Category 1 or 2 is a challenge, irrespective of the extent of clinical nursing experience. We found that 20% of patients with an AMI were given an ATS category of 3–5 and were therefore under-triaged. This important finding is an improvement on past estimates; prior research has found that up to 50% of AMI patients were under-triaged, when comparing to a similar tertiary hospital population in Canada with a similar triage scale5 (Web Appendix 1). We are unable to explain the lower rates of under-triaging at our institution; our local practice mandates that triage nurses complete standardised ETEK training after completing the Transition to Emergency Nursing Program over at least twelve months, and utilises a standardised protocol for assigning triage scores. Though our results show an improvement in triage of AMI patients, under-triage may still be a problem; the data in this study suggests one in five patients may experience unnecessary delays to medical treatment. Ninety three percent of the under-triaged group (n = 30) presented with atypical features. This finding is in line with previous research, which suggests that despite atypical presentations for patients with an AMI being common, such patients are often misdiagnosed and undertreated.13,18 This group of patients have signs and symptoms that are clinically ambiguous, and have often occurred on a background of multiple co-morbidities.4 The undertriage of this group is therefore understandable but remains of concern; patients who experience AMI without chest pain are significantly more likely to have an overall longer length of stay and are more likely to die in hospital when compared with AMI patients with chest pain.4,13 This study supports past research that found females and the elderly were more likely to be under-triaged.13,19 We too found that AMI patients presenting with an absence of chest pain tended to be older and female.18,19 Women presenting with AMI are known to not be as easy to identify as men.15 Of the under-triaged group,
63% were female with a mean age of 76 years compared with 68 years in the correctly triaged cohort. One third of the patients who were under-triaged were elderly patients presenting to the ED after sustaining a ‘fall’ of some kind, with an absence of chest pain. Similar data were also found by Grosmaitrea et al.15 Ours was an incidental finding which could help illuminate a potential underlying contributing factor in the fall. Older patients have a higher incidence of impaired cognition and memory, which could affect recall of symptoms when arriving at the emergency department for triage.15 Level of triage experience has been identified previously as a factor that may be associated with triage decisions.20 This may be a result of experienced nurses’ false reliance on their ability to differentiate ambiguous clinical signs and symptoms, along with a broader knowledge of differential diagnoses. Also, junior nursing staff may be cautiously more protocol-oriented and refer directly to the Australian Triage Criteria when allocating a triage category. While our study found that a slightly lower proportion of patients were assigned an ATS category 1 or 2 by experienced Registered Nurses (>2 years experience), the difference was non significant. It is unclear why our finding differed from previous research but it may be because it was underpowered to detect an effect. A larger study with greater power to provide robust estimates is required to understand whether triage experience is an important factor. 4.1. Limitations, implications for emergency nurses, conclusions On examining level of triage experience, we did not assess Triage nurses’ total years experience as a Registered Nurse, as triage experience was only determined by date of accreditation at this facility. Some of these nurses may have had prior Emergency experience from other hospitals. It is difficult to constructively critique a nurse’s triage allocation for an AMI patient as the end diagnosis of AMI is determined by a collection of clinical data gathered over the patient’s admission. The symptoms presented to the nurse at the time of triage form just one small part of the clinical picture that will ultimately inform the final diagnosis. Therefore, while the retrospective nature of the data is valuable in identifying possible clinical features that are common to the under-triaged group, this is unlikely to improve the actual triage process at this time. This study includes only a small number of participants and future research would need to validate findings in a bigger sample. However, it should be noted that the demographic characteristics of the sample we used are similar to AMI patients in other Australian ED cohorts21 and to high risk ED populations in the Asia Pacific region.22 This provides some evidence that the sample is representative of AMI patients in Australian settings. Finally, the conclusions in this study are based on simple descriptive statistics and so we could not provide data on independent predictors of under-triage. Future studies with larger numbers of patients would be required to enable multivariable analyses.
Please cite this article in press as: Ryan K, et al. Factors associated with triage assignment of emergency department patients ultimately diagnosed with acute myocardial infarction. Aust Crit Care (2015), http://dx.doi.org/10.1016/j.aucc.2015.05.001
G Model
ARTICLE IN PRESS
AUCC-277; No. of Pages 4
K. Ryan et al. / Australian Critical Care xxx (2015) xxx–xxx
4
4.2. Implications for emergency nurses
References
This study highlights a number of the patient characteristics that ultimately place them at risk for under-triage following AMI. In addition, the significance of the finding that nurses with greater than two years’ experience were more likely to under-triage a patient presenting with an AMI is unclear but warrants further investigation. The ATS protocol is in place to guide clinicians in triaging patients and it may serve triage nurses better with increased referral to and reflection on those guidelines. The implementation of ETEK guidelines may help to consistently guide nurses when triaging these patients. Further, these results may suggest the need for revised hospital protocols or enhanced clinical guidelines within the current triage system to better capture and treat those vulnerable patients who are presently under-served. In the RBWH DEM nursing in-services have been utilised to highlight these outcomes and raise awareness about being mindful when triaging patients with atypical features (age, sex, history of a mechanical fall); all patients over 70 years who present to our department now receive an ECG on arrival, regardless of what their presenting problem is. Further questioning at triage to potentially elucidate if there is an underlying cardiac cause has been suggested as a way to improve on the clinical assessment of these patients.
1. Mathur S. Epidemic of coronary heart disease and its treatment in Australia, cardiovascular disease series. Canberra: AIHW; 2002. 2. De Luca G, Suryapranata H, Marino P. Reperfusion strategies in acute STelevation myocardial infarction: an overview of current status. Prog Cardiovasc Dis 2008;50:352–82. 3. Keeley E, Hillis L. Primary PCI for myocardial infarction with ST-segment elevation. N Engl J Med 2007;356(47-54). 4. Gillis N, Arslanian-Engoren C, Struble L. Acute coronary syndromes in older adults: a review of literature. J Emerg Nurs 2014;40(3):270–5. 5. Atzema C, Austin P, Tu J, Schull M. ED triage of patients with acute myocardial infarction: predictors of low acuity triage. Am J Emerg Med 2010;28:694–702. 6. Providencia R, Gomes PL, Barra B, Silva J, Seca L, Antunes A, et al. Importance of Manchester Triage in acute myocardial infarction: impact on prognosis. Emerg Med J 2011;28(212–216). 7. Santos A, Freitas P, Martins HMG. Manchester triage system version II and resource utilisation in emergency department. Emerg Med J 2014;31(2). 8. ACEM. Policy on the Australasian Triage Scale; 2013. https://www.acem.org. au/getattachment/693998d7-94be-4ca7-a0e7-3d74cc9b733f/Policy-on-theAustralasian-Triage-Scale.aspx 9. ACEM. Guidelines on the implementation of the Australasian Triage Scale; https://wwwacemorgau/getattachment/d19d5ad3-e1f4-4e4f-bf832008. 7e09cae27d76/G24-Implementation-of-the-Australasian-Triage-Scalaspx 10. DOHA. Emergency triage education kit. Canberra: Ageing DoHa; 2009. 11. Luepker R, Apple F, Christenson R. Case definitions for acute coronary heart disease in epidemiology and clinical research studies. Circulation 2003;108:2543–9. 12. Goel PKSS, Ashfaq F, Gupta PR, Saxena PC, Agarwal R, Kumar S, et al. A study of clinical presentation and delays in management of acute myocardial infarction in the community. Indian Heart J 2012;6403:295–301. 13. Kuhn LPK, Rolley JX, Worrall-Carter L. Effect of patient sex on triage for ischaemic heart disease and treatment onset times: a retrospective analysis of Australian emergency department data. Int Emerg Nurs 2014;22(2):88–93. 14. Kuhn L, Worrall-Carter L, Ward J, Page K. Factors associated with delayed treatment onset for acute myocardial infarction in Victorian emergency departments: a regression tree analysis. AENJ 2014;16:160–9. 15. Grosmaitrea P, Le Vavasserurb O, Yachouhc E, Courtiald Y, Jacobe X, Meyranf S, et al. Significance of atypical symptoms for the diagnosis and management of myocardial infarction in elderly patients admitted to emergency departments. Arch Cardiovasc Dis 2013;106:586–92. 16. Neill K. Review of atypical clinical manifestations of acute myocardial infarction. J Intensive Care Med 1987;2(25):25–32. 17. Ross DC, Cooperrider C, Homan MB. Acute coronary ischemia identified by EMS providers in a standing middle-aged male with atypical symptoms. Prehosp Emerg Care 2014;18:450–5. 18. Brieger D, Eagle KA, Goodman SG, Steg PG, Budaj A, White K, et al. Acute coronary syndromes without chest pain, an underdiagnosed and undertreated high-risk group. Chest 2004;126(2):461–9. 19. Canto J, Rogers WJ, Goldberg RJ, Peterson ED, Wenger NK, Vaccarino V, et al. Association of age and sex with myocardial infarction symptom presentation and in-hospital mortality. JAMA 2012;307(8):813–22. 20. Sammons S [nursing dissertations] Accuracy of emergency department nurse triage level designation and delay in care of patients with symptoms suggestive of acute myocardial infarction; 2012. 21. Macdonald SPJ, Nagree Y, Fatovich D, Flavell HL, Loutsky F. Comparison of two clinical scoring systems for emergency department risk stratification of suspected acute coronary syndrome. Emerg Med Australas 2011;23:717–25. 22. Than M, Cullen L, Reid CM, Lim SH, Aldous Sa, Ardagh MW, et al. A 2-h diagnostic protocol to assess patients with chest pain symptoms in the AsiaPacific region (ASPECT): a prospective observational validation study. Lancet 2011;377:1077–84.
5. Conclusion Estimates of under-triage of AMI patients in this cohort are lower than past research yet one fifth of patients are still undertriaged. This may lead to delays in medical assessment and treatment and potentially compromise patient outcomes including mortality. Elderly and female patients are still being under-triaged despite prior research highlighting this issue. The absence of chest pain in AMI is a significant factor in under-triage. Clinical signs and symptoms can be complex to discern against a background of comorbidities, especially in older patients. This study may assist more experienced triage nurses to be mindful of the increased potential for these sub-groups to present with an AMI. Authors’ contributions All authors contributed to all stages of the study, and all authors have approved this final version. Acknowledgment The Royal Brisbane and Women’s Foundation (2010-kr) had reviewed and granted $15,000 towards this research project as a Nurse Initiative Research Grant in February 2010. Web Appendix 1. Triage scale and maximum recommended wait time in minutes Scale
Category/level 1
Category/level 2
Category/level 3
Category/level 4
Category/level 5
Australasian Triage Scale Canadian Triage and Acuity Scale Manchester Triage Scale
0 0 0
10 15 10
30 30 60
60 60 120
120 120 240
Please cite this article in press as: Ryan K, et al. Factors associated with triage assignment of emergency department patients ultimately diagnosed with acute myocardial infarction. Aust Crit Care (2015), http://dx.doi.org/10.1016/j.aucc.2015.05.001
ARTICLE IN PRESS Australian Critical Care xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Australian Critical Care journal homepage: www.elsevier.com/locate/aucc
Research paper
Factors associated with triage assignment of emergency department patients ultimately diagnosed with acute myocardial infarction Kimberley Ryan BN, GradDip Nurs (Crit Care), BHSc (Nat) a,∗ , Jaimi Greenslade PhD a,b,c , Emily Dalton BSc, BNursing a , Kevin Chu MBBS, MSc, FACEM a,b , Anthony F.T. Brown MBChB, FRCP, FRCSEd, FACEM, FCEM a,b , Louise Cullen MBBS, FACEM a,b,c a b c
Royal Brisbane and Women’s Hospital, Department of Emergency Medicine, Brisbane 4006, Australia School of Medicine, The University of Queensland, St Lucia, 4067, Australia School of Public Health, Queensland University of Technology, Kelvin Grove, 4059, Australia
article information Article history: Received 19 October 2014 Received in revised form 27 April 2015 Accepted 4 May 2015 Available online xxx Keywords: Chest pain Emergency department Emergency nursing Myocardial infarction Triage
a b s t r a c t Background: The objective of this study was to explore factors associated with the triage category assigned by the triage nurse for patients ultimately diagnosed with acute myocardial infarction. Methods: This was a retrospective analysis of 12 months of data, on adult emergency department patients ultimately diagnosed with acute myocardial infarction. Data were obtained from hospital databases and included patient demographics, patient clinical characteristics and nurses’ experience. Results: Of the 153 patients, 20% (95% CI: 14–27%) were given a lower urgency triage category than recommended by international guidelines. Compared to patients who were triaged Australasian Triage Category 1 or 2, patients with an Australasian Triage Category 3–5 were older (mean age 76 versus 68 years), more likely to be female (63% versus 32%), more likely to present without chest pain (93% versus 35%) and less likely to have a cardiac history (3.3% versus 17.9%). A slightly higher proportion of patients Australasian Triage Category 3–5 were triaged by an experienced nurse (50%) compared to patients categorised Australasian Triage Category 1–2 (35.2%) but this finding did not reach statistical significance. Conclusions: One in five presentations was given a lower urgency triage category than recommended by international guidelines, potentially leading to delays in medical treatment. The absence of chest pain was the defining characteristic in this group of patients, along with other factors identified by previous research such as being of female sex and elderly. © 2015 Published by Elsevier Ltd. on behalf of Australian College of Critical Care Nurses Ltd.
1. Introduction Acute Myocardial Infarction (AMI) is a leading cause of premature death and disability for Australian men and women.1 Rapid assessment and treatment of patients with AMI is essential as mortality associated with AMI is directly linked to time taken to receive treatment.2,3 The initial clinical assessment for patients
∗ Corresponding author at: Royal Brisbane and Women’s Hospital, Department of Emergency Medicine, Butterfield Street, Herston, 4006 Queensland, Australia. Tel.: +61 7 36464629; fax: +61 7 3646 8732. E-mail address: [email protected] (K. Ryan).
who present to hospital emergency department (ED) occurs by a triage officer. In most Western countries, the triage officer is a registered nurse who has specialised in emergency nursing. The key role of the triage nurse is to accurately identify potential patients who may have an AMI as early as possible to expedite necessary immediate cardiac care.4 Triage is a system which allows the clinical urgency of a presenting problem to be categorised. Similar 5-level triage systems are used in Canada (the Canada Triage and Acuity Scale),5 Europe, the United Kingdom (the Manchester Triage scale) and Australia (the Australasian Triage Scale (ATS))6,7 (Web Appendix 1). In Australia, experienced ED nurses complete mandatory training from the Emergency Triage Education Kit (ETEK) and must be well
http://dx.doi.org/10.1016/j.aucc.2015.05.001 1036-7314/© 2015 Published by Elsevier Ltd. on behalf of Australian College of Critical Care Nurses Ltd.
Please cite this article in press as: Ryan K, et al. Factors associated with triage assignment of emergency department patients ultimately diagnosed with acute myocardial infarction. Aust Crit Care (2015), http://dx.doi.org/10.1016/j.aucc.2015.05.001
G Model AUCC-277; No. of Pages 4
ARTICLE IN PRESS K. Ryan et al. / Australian Critical Care xxx (2015) xxx–xxx
2
versed with the ATS. The ATS ranges from 1 to 5 with the ranking correlating with the recommended maximum time a patient should wait for treatment. Patients categorised as ATS 1 require immediate treatment, while patients categorised as ATS 2, 3, 4 or 5 are expected to receive medical assessment and treatment within 10, 30, 60, and 120 min respectively.8 In line with the Australasian College of Emergency Medicine guidelines and the ETEK, a patient presenting with symptoms suggestive of Acute Coronary Syndrome (ACS) should be triaged as a Category 2.9,10 These symptoms may include acute chest, epigastric, neck, jaw, or arm pain; or discomfort or pressure without an apparent non cardiac source.11 According to the ETEK, ‘under-triage’ describes the process whereby the patient receives a triage code lower than their actual level of urgency.10 Although Australian data are lacking, international research suggests a significant number of patients with AMIs are under-triaged, with one study finding up to half of all AMI patients being assigned a lower priority triage category on presentation.5 The reasons for under-triage of AMI patients have not been adequately elucidated, though research suggests factors including age, sex, an absence of chest pain at triage, and history of diabetes mellitus or heart failure make AMI harder to recognise.12–14 Given that triage is a modifiable factor influencing delays to treatment, further understanding of the determinants of triage category in an Australian setting is an essential step in enhancing the triage process. 1.1. Purpose of the study This study explored the triage category assigned to patients with myocardial infarction in a large tertiary hospital ED in Queensland, Australia. The study aim was to identify the factors associated with the triage of patients presenting to ED with AMI. Predictors of interest included patient demographics, clinical characteristics and nursing triage experience. 2. Patients and methods 2.1. Study design and setting This was an analysis of retrospective collected data on adult patients presenting to the ED who were ultimately diagnosed with AMI. The data were collected from the Royal Brisbane and Women’s Hospital (RBWH) between 1 June 2009 and 31 May 2010. The RBWH is a 929 bed adult tertiary-referral teaching hospital; the RBWH ED has an annual attendance rate of 72,000 patients over the age of 14 years. This paper reports the findings of a research study that adhered to the National Statement on the Conduct of Human Research by the Australian National Health and Medical Research Council, and has been approved by the RBWH Human Research Ethics Committee on the 30th April 2010. 2.2. Case selection and data collection Patients with an AMI were identified through several sequential steps. The RBWH pathology department provided a list of all patients who presented to the ED and had a serum troponin I (TNI) performed as part of their emergency workup. This list was then refined to include only those patients with a TNI value of ≥0.06 mcg/L. The TNI assay used in at the RBWH during the study period was the Beckman Coulter AccuTnI assay and 0.06 was the clinical decision cut-off point. The next step was to undertake a review of the discharge diagnosis on the Emergency Department Information System (EDIS) to confirm a diagnosis of AMI for those patients. Patients with an EDIS diagnosis of ‘Chest Pain’ or ‘Acute Coronary Syndrome’ (ACS) were further examined to identify whether they had a diagnosis of AMI on the
Table 1 Baseline characteristics of the cohort (n = 153). Characteristic
n (%)
Mean ± SD age (years) Male sex English as primary language Employment status Employed Pensioner Unemployed Other/unknown Mode of arrival to emergency department Own transport Ambulance service
69.5 ± 14.4 94 (61.4) 144 (94.1) 42 (27.5) 82 (53.6) 12 (7.8) 17 (11.1) 39 (25.5) 114 (74.5)
index admission. This initial examination was undertaken using charts and electronic medical discharge summaries where the diagnosis was documented by a cardiologist or general physician. For the purposes of this study, AMI included diagnosis of AMI (NSTEMI or STEMI) on the index admission or urgent revascularisation on index admission including coronary angioplasty, coronary artery stenting and coronary artery bypass grafting. Exclusion criteria included pregnancy, age < 18 years and any patients transferred to the ED from another hospital. Once the population was identified, study data were obtained from a number of sources. Data on patient demographics (age and sex), presenting symptoms, cardiac history, ambulance use and triage category were collected from the EDIS database. Presenting symptoms were categorised as typical or atypical for AMI. Typical symptoms referred to the presence of any chest pain during the event (present or resolved on arrival), while atypical symptoms included dizziness, syncope, nausea or vomiting and dyspnoea with the absence of chest pain prior to arrival or during presentation.15–17 The name of the nurse who triaged the patient was provided from the ED admission system, and the years of nursing triage experience was then sought from the ED Nurse Educator’s records. Data was de-identified prior to analysis. 2.3. Data analysis Data were analysed using SPSS version 20. Baseline characteristics of the sample were reported. A triage category of 1 or 2 was categorised as high urgency while a triage category of 3–5 was deemed lower urgency. Standard descriptive statistics were used to report the characteristics of the correctly triaged and under-triaged patient groups. Chi-square tests (or Fisher’s exact tests where cell sizes were small) were performed to compare dichotomous data across triage categories. T-tests were performed to compare continuous characteristics across triage categories. There only were a small number of patients with an ATS 3–5 (n = 30) and so it was not deemed appropriate to perform multivariable analyses to identify the independent predictors of triage category. 3. Results There were a total of 153 patients identified with an index AMI for analysis. The sample included 94 (61.4%) males and the mean age was 69.5 years (SD = 14.1 years). Baseline characteristics of the cohort are provided in Table 1 and demonstrate that the majority of the patients were English-speaking pensioners who arrived via ambulance. One hundred and twenty three (80.4%, 95% CI: 73.2–86.4%) patients were provided an appropriate triage category (ATS 1–2).
Please cite this article in press as: Ryan K, et al. Factors associated with triage assignment of emergency department patients ultimately diagnosed with acute myocardial infarction. Aust Crit Care (2015), http://dx.doi.org/10.1016/j.aucc.2015.05.001
G Model AUCC-277; No. of Pages 4
ARTICLE IN PRESS K. Ryan et al. / Australian Critical Care xxx (2015) xxx–xxx
3
Table 2 Patient characteristics and nursing triage experience for patients categorised as ATS 1–2 (n = 123) and ATS 3–5 (n = 30).
Male Mean (SD) patient age, years Arrival by ambulance Presence of chest pain Cardiac history Nursing triage experience >2 years
Triage category 1–2 (n = 123)
Triage category 3–5 (n = 30)
p
83 (67.5%) 67.79 (13.59) 93 (75.6%) 79 (64.2%) 22 (17.9%) 43 (35.2%)
11 (36.7%) 76.33 (14.03) 21 (70%) 2 (6.7%) 1 (3.3%) 15 (50%)
<0.01 <0.01 0.53 <0.01 0.05 0.14
Data are number (%) unless otherwise indicated.
Table 2 provides nurse and patient characteristics for patients who were categorised ATS 1–2 versus ATS 3–5. At a univariate level, females and older individuals were less often assigned a triage category 1–2. Females were older (mean = 76.56 years, SD = 14.10) than males (mean = 65.01 years, SD = 12.12). However, the finding that triage category 1–2 were younger was true both for males and females (p = 0.31) Patients with a cardiac history and patients with chest pain were more often assigned a triage category 1–2. A slightly higher proportion of under-triaged patients were triaged by an experienced nurse (50%) compared to correctly triaged patients (35.2%) but this finding did not reach statistical significance. 4. Discussion Though the ETEK was published to help support consistency in triage practice across Australia, the implementation of it as a mandatory learning tool was not initiated in the RBWH DEM until the end of 2010. Thus the collection of data in this study pre-dates the implementation of ETEK; rather the ATS guided triage at this time. The appropriate triage of AMI patients to an ATS Category 1 or 2 is a challenge, irrespective of the extent of clinical nursing experience. We found that 20% of patients with an AMI were given an ATS category of 3–5 and were therefore under-triaged. This important finding is an improvement on past estimates; prior research has found that up to 50% of AMI patients were under-triaged, when comparing to a similar tertiary hospital population in Canada with a similar triage scale5 (Web Appendix 1). We are unable to explain the lower rates of under-triaging at our institution; our local practice mandates that triage nurses complete standardised ETEK training after completing the Transition to Emergency Nursing Program over at least twelve months, and utilises a standardised protocol for assigning triage scores. Though our results show an improvement in triage of AMI patients, under-triage may still be a problem; the data in this study suggests one in five patients may experience unnecessary delays to medical treatment. Ninety three percent of the under-triaged group (n = 30) presented with atypical features. This finding is in line with previous research, which suggests that despite atypical presentations for patients with an AMI being common, such patients are often misdiagnosed and undertreated.13,18 This group of patients have signs and symptoms that are clinically ambiguous, and have often occurred on a background of multiple co-morbidities.4 The undertriage of this group is therefore understandable but remains of concern; patients who experience AMI without chest pain are significantly more likely to have an overall longer length of stay and are more likely to die in hospital when compared with AMI patients with chest pain.4,13 This study supports past research that found females and the elderly were more likely to be under-triaged.13,19 We too found that AMI patients presenting with an absence of chest pain tended to be older and female.18,19 Women presenting with AMI are known to not be as easy to identify as men.15 Of the under-triaged group,
63% were female with a mean age of 76 years compared with 68 years in the correctly triaged cohort. One third of the patients who were under-triaged were elderly patients presenting to the ED after sustaining a ‘fall’ of some kind, with an absence of chest pain. Similar data were also found by Grosmaitrea et al.15 Ours was an incidental finding which could help illuminate a potential underlying contributing factor in the fall. Older patients have a higher incidence of impaired cognition and memory, which could affect recall of symptoms when arriving at the emergency department for triage.15 Level of triage experience has been identified previously as a factor that may be associated with triage decisions.20 This may be a result of experienced nurses’ false reliance on their ability to differentiate ambiguous clinical signs and symptoms, along with a broader knowledge of differential diagnoses. Also, junior nursing staff may be cautiously more protocol-oriented and refer directly to the Australian Triage Criteria when allocating a triage category. While our study found that a slightly lower proportion of patients were assigned an ATS category 1 or 2 by experienced Registered Nurses (>2 years experience), the difference was non significant. It is unclear why our finding differed from previous research but it may be because it was underpowered to detect an effect. A larger study with greater power to provide robust estimates is required to understand whether triage experience is an important factor. 4.1. Limitations, implications for emergency nurses, conclusions On examining level of triage experience, we did not assess Triage nurses’ total years experience as a Registered Nurse, as triage experience was only determined by date of accreditation at this facility. Some of these nurses may have had prior Emergency experience from other hospitals. It is difficult to constructively critique a nurse’s triage allocation for an AMI patient as the end diagnosis of AMI is determined by a collection of clinical data gathered over the patient’s admission. The symptoms presented to the nurse at the time of triage form just one small part of the clinical picture that will ultimately inform the final diagnosis. Therefore, while the retrospective nature of the data is valuable in identifying possible clinical features that are common to the under-triaged group, this is unlikely to improve the actual triage process at this time. This study includes only a small number of participants and future research would need to validate findings in a bigger sample. However, it should be noted that the demographic characteristics of the sample we used are similar to AMI patients in other Australian ED cohorts21 and to high risk ED populations in the Asia Pacific region.22 This provides some evidence that the sample is representative of AMI patients in Australian settings. Finally, the conclusions in this study are based on simple descriptive statistics and so we could not provide data on independent predictors of under-triage. Future studies with larger numbers of patients would be required to enable multivariable analyses.
Please cite this article in press as: Ryan K, et al. Factors associated with triage assignment of emergency department patients ultimately diagnosed with acute myocardial infarction. Aust Crit Care (2015), http://dx.doi.org/10.1016/j.aucc.2015.05.001
G Model
ARTICLE IN PRESS
AUCC-277; No. of Pages 4
K. Ryan et al. / Australian Critical Care xxx (2015) xxx–xxx
4
4.2. Implications for emergency nurses
References
This study highlights a number of the patient characteristics that ultimately place them at risk for under-triage following AMI. In addition, the significance of the finding that nurses with greater than two years’ experience were more likely to under-triage a patient presenting with an AMI is unclear but warrants further investigation. The ATS protocol is in place to guide clinicians in triaging patients and it may serve triage nurses better with increased referral to and reflection on those guidelines. The implementation of ETEK guidelines may help to consistently guide nurses when triaging these patients. Further, these results may suggest the need for revised hospital protocols or enhanced clinical guidelines within the current triage system to better capture and treat those vulnerable patients who are presently under-served. In the RBWH DEM nursing in-services have been utilised to highlight these outcomes and raise awareness about being mindful when triaging patients with atypical features (age, sex, history of a mechanical fall); all patients over 70 years who present to our department now receive an ECG on arrival, regardless of what their presenting problem is. Further questioning at triage to potentially elucidate if there is an underlying cardiac cause has been suggested as a way to improve on the clinical assessment of these patients.
1. Mathur S. Epidemic of coronary heart disease and its treatment in Australia, cardiovascular disease series. Canberra: AIHW; 2002. 2. De Luca G, Suryapranata H, Marino P. Reperfusion strategies in acute STelevation myocardial infarction: an overview of current status. Prog Cardiovasc Dis 2008;50:352–82. 3. Keeley E, Hillis L. Primary PCI for myocardial infarction with ST-segment elevation. N Engl J Med 2007;356(47-54). 4. Gillis N, Arslanian-Engoren C, Struble L. Acute coronary syndromes in older adults: a review of literature. J Emerg Nurs 2014;40(3):270–5. 5. Atzema C, Austin P, Tu J, Schull M. ED triage of patients with acute myocardial infarction: predictors of low acuity triage. Am J Emerg Med 2010;28:694–702. 6. Providencia R, Gomes PL, Barra B, Silva J, Seca L, Antunes A, et al. Importance of Manchester Triage in acute myocardial infarction: impact on prognosis. Emerg Med J 2011;28(212–216). 7. Santos A, Freitas P, Martins HMG. Manchester triage system version II and resource utilisation in emergency department. Emerg Med J 2014;31(2). 8. ACEM. Policy on the Australasian Triage Scale; 2013. https://www.acem.org. au/getattachment/693998d7-94be-4ca7-a0e7-3d74cc9b733f/Policy-on-theAustralasian-Triage-Scale.aspx 9. ACEM. Guidelines on the implementation of the Australasian Triage Scale; https://wwwacemorgau/getattachment/d19d5ad3-e1f4-4e4f-bf832008. 7e09cae27d76/G24-Implementation-of-the-Australasian-Triage-Scalaspx 10. DOHA. Emergency triage education kit. Canberra: Ageing DoHa; 2009. 11. Luepker R, Apple F, Christenson R. Case definitions for acute coronary heart disease in epidemiology and clinical research studies. Circulation 2003;108:2543–9. 12. Goel PKSS, Ashfaq F, Gupta PR, Saxena PC, Agarwal R, Kumar S, et al. A study of clinical presentation and delays in management of acute myocardial infarction in the community. Indian Heart J 2012;6403:295–301. 13. Kuhn LPK, Rolley JX, Worrall-Carter L. Effect of patient sex on triage for ischaemic heart disease and treatment onset times: a retrospective analysis of Australian emergency department data. Int Emerg Nurs 2014;22(2):88–93. 14. Kuhn L, Worrall-Carter L, Ward J, Page K. Factors associated with delayed treatment onset for acute myocardial infarction in Victorian emergency departments: a regression tree analysis. AENJ 2014;16:160–9. 15. Grosmaitrea P, Le Vavasserurb O, Yachouhc E, Courtiald Y, Jacobe X, Meyranf S, et al. Significance of atypical symptoms for the diagnosis and management of myocardial infarction in elderly patients admitted to emergency departments. Arch Cardiovasc Dis 2013;106:586–92. 16. Neill K. Review of atypical clinical manifestations of acute myocardial infarction. J Intensive Care Med 1987;2(25):25–32. 17. Ross DC, Cooperrider C, Homan MB. Acute coronary ischemia identified by EMS providers in a standing middle-aged male with atypical symptoms. Prehosp Emerg Care 2014;18:450–5. 18. Brieger D, Eagle KA, Goodman SG, Steg PG, Budaj A, White K, et al. Acute coronary syndromes without chest pain, an underdiagnosed and undertreated high-risk group. Chest 2004;126(2):461–9. 19. Canto J, Rogers WJ, Goldberg RJ, Peterson ED, Wenger NK, Vaccarino V, et al. Association of age and sex with myocardial infarction symptom presentation and in-hospital mortality. JAMA 2012;307(8):813–22. 20. Sammons S [nursing dissertations] Accuracy of emergency department nurse triage level designation and delay in care of patients with symptoms suggestive of acute myocardial infarction; 2012. 21. Macdonald SPJ, Nagree Y, Fatovich D, Flavell HL, Loutsky F. Comparison of two clinical scoring systems for emergency department risk stratification of suspected acute coronary syndrome. Emerg Med Australas 2011;23:717–25. 22. Than M, Cullen L, Reid CM, Lim SH, Aldous Sa, Ardagh MW, et al. A 2-h diagnostic protocol to assess patients with chest pain symptoms in the AsiaPacific region (ASPECT): a prospective observational validation study. Lancet 2011;377:1077–84.
5. Conclusion Estimates of under-triage of AMI patients in this cohort are lower than past research yet one fifth of patients are still undertriaged. This may lead to delays in medical assessment and treatment and potentially compromise patient outcomes including mortality. Elderly and female patients are still being under-triaged despite prior research highlighting this issue. The absence of chest pain in AMI is a significant factor in under-triage. Clinical signs and symptoms can be complex to discern against a background of comorbidities, especially in older patients. This study may assist more experienced triage nurses to be mindful of the increased potential for these sub-groups to present with an AMI. Authors’ contributions All authors contributed to all stages of the study, and all authors have approved this final version. Acknowledgment The Royal Brisbane and Women’s Foundation (2010-kr) had reviewed and granted $15,000 towards this research project as a Nurse Initiative Research Grant in February 2010. Web Appendix 1. Triage scale and maximum recommended wait time in minutes Scale
Category/level 1
Category/level 2
Category/level 3
Category/level 4
Category/level 5
Australasian Triage Scale Canadian Triage and Acuity Scale Manchester Triage Scale
0 0 0
10 15 10
30 30 60
60 60 120
120 120 240
Please cite this article in press as: Ryan K, et al. Factors associated with triage assignment of emergency department patients ultimately diagnosed with acute myocardial infarction. Aust Crit Care (2015), http://dx.doi.org/10.1016/j.aucc.2015.05.001