ISSUES IN CLINICAL NURSING

Triage in emergency departments: national survey

Katarina E Go ransson
rebro University, O rebro, Sweden O

Anna Ehrenberg
rebro University, O rebro, Sweden O

Margareta Ehnfors
rebro University, O rebro, Sweden O

Submitted for publication: 11 May 2004 Accepted for publication: 1 February 2005

Correspondence: Katarina E Go ransson Department of Caring Sciences rebro University O SE-701 82 rebro O Sweden Telephone: 46 739 088249 E-mail: katarina.goransson@ivo.oru.se

Journal of Clinical Nursing 14, 10671074 Triage in emergency departments: national survey Aim. This paper reports a study the aim of which was to describe how triage-related work was organized and performed in Swedish emergency departments. Background. Hospitals in many developed countries use some kind of system to prioritize the patients attending emergency departments. Triage is a commonly used term to refer to the process of sorting and prioritizing patients for care. How the triage procedure is organized and which personnel perform this type of work vary considerably throughout the world. In Sweden, few studies have explored this important issue. Method. A national survey was conducted using telephone interviews, with nurse managers at each of the emergency departments. The sample represented 87% of emergency departments in Sweden. Results. The ndings clearly illustrate the organization of emergency department triage, focusing on personnel who perform triage, as well as the facilities, resources and procedures available for triage. However, the results indicate that work associated with such triage in Sweden is not organized in any consistent matter. In 81% of the emergency departments a clerk, Licensed Practical Nurse or Registered Nurse were assigned to assess patients not arriving by ambulance. There was also diversity in other areas, including requirements for staff to have particular qualications and clinical experience for being allocated to triage work, as well as facilities for triage personnel assessing and prioritizing patients. The use of triage scales and acuity ratings also lacked uniformity and disparities were observed in both the design and use of triage scales. A little less than half (46%) of the emergency departments did not use any kind of triage scale to document patient acuity ratings. Conclusion. In contrast to several other countries, this study shows that Swedish emergency departments do not adhere well to established standards and guidelines about triage in emergency care. Research on emergency department triage, especially in the areas of personnel performing triage, triage scales and standards and guidelines are recommended. Relevance to clinical practice. The diversity among several aspects of nursing triage (e.g. use of less qualied personnel performing triage, the use of different triage scales) presented in the study points to a safety risk for the patients. It also shows the
1067

RANSSON KE, EHRENBERG A & EHNFORS M (2005) GO

2005 Blackwell Publishing Ltd

KE Go ransson et al.

need of further education for the personnel in clinical practice as well as further research on triage in order to gain national consensus about this nursing task. Key words: accident and emergency department, acute care, emergency, nurse, survey, triage

Background
The concept of triage is used in three distinct situations: emergency departments (EDs), military situations and disaster conditions. Triage is dened as the process of sorting and prioritizing patients for care (Estrada 1981, Thompson & Dains 1982, Read et al. 1992, Handysides 1996, Manchester Triage Group 1997). Baron Dominique Jean-Larrey, surgeon to Napoleon, rst described the process of triage during the 18th century (Beveridge 2000). When performed in an ED, the aim of triage is to ensure that patients are treated in the order of their clinical urgency and that they receive treatment in a timely and appropriate manner (Australasian College for Emergency Medicine 2000). To prioritize a patient, a triage assessment has to be made. This assessment consists of the collection of information about a patients chief complaint and it can be combined with physical examinations. Tasks performed during triage can be divided into two groups: primary triage functions and nonprimary functions. Primary triage functions relate to an initial patient assessment, including physical assessments and reassessments, recording vital signs, assigning an acuity rating to the patient, providing emergency interventions and determining patient disposition. Non-primary triage functions, such as giving telephone advice, taking blood samples and approving or completing taxi vouchers, may be performed by the same personnel performing triage (Geraci & Geraci 1994, Gerdtz & Bucknall 2000). In the study reported here, primary triage functions are in focus. Several studies have shown that there are variations in how long an assessment takes and what kinds of observations and examinations are performed during the process (Purnell 1991, Ciof 1998, Gerdtz & Bucknall 2001). After an assessment has been performed, the patient is assigned an acuity rating (Australasian College for Emergency Medicine 2000), which indicates the length of time they can wait before being seen by a physician (Gerdtz & Bucknall 2000). Acuity ratings are normally based on a triage scale, but the design of these scales varies considerably between EDs. Specic countries (e.g. Australia) have nationally determined triage scales with ve ratings indicating waiting time (Manchester Triage Group 1997, Beveridge 1998, Australasian College for Emergency Medicine 2000). Canada uses a nationally accepted ve-level triage scale, which has been
1068

developed based on the Australasian version. Substantial work has taken place in Canada since 1995 when the Canadian Association of Emergency Physicians (CAEP) rst introduced the ve-level scale and this has resulted in the nationally accepted Canadian ED Triage and Acuity Scale (CTAS). In many countries (e.g. Australia and the UK), a Registered Nurse (RN) is assigned to perform triage in EDs (Purnell 1991, Manchester Triage Group 1997, Gerdtz & Bucknall 2001). In 2000, a survey covering 70 of 81 Swedish EDs found that half used RNs to perform triage (Palmquist & Lindell 2000). The qualications required for being rostered for triage have been found to vary (Purnell 1991, Gerdtz & Bucknall 2000, Palmquist & Lindell 2000) and several authors have suggested that performing triage requires special knowledge (Shields 1976, Purnell 1991, Rock & Pledge 1991, Australasian College for Emergency Medicine 2000). An inappropriate triage decision could result in delays for all patients in the ED, not to mention excess costs for the department (Wuerz et al. 1998, Fernandes et al. 1999). It might also mean that patients in need of urgent care would have to wait, which could have serious consequences. Many patients arriving at an ED feel that they are ill and need immediate care. Thus, having to wait for care can be experienced as highly frustrating and annoying. Performing triage and deciding on an acuity rating is an advanced and challenging task that is the foundation of a patients visit to the ED. Because the acuity rating largely determines how long the patient will wait before seeing a physician, correct ratings are of fundamental importance. The safety of patients who present at EDs demands that personnel have appropriate knowledge and experience and that adequate facilities and equipment to perform triage are accessible. In 1996, only 50% of Swedish EDs used RNs to perform triage (Palmquist & Lindell 2000). The absence of national standards and guidelines on triage, both regarding triaging personnel and the design and use of acuity rating scales, calls for further study in this area. Before data collection for the study reported here, information about standards and guidelines concerning triage personnel and triage scales in Sweden was gathered from three sources: the National Board of Health and Welfare, and several Swedish physician and RN associations. One publication (Socialstyrelsen 1994) on acuity rating scales

2005 Blackwell Publishing Ltd, Journal of Clinical Nursing, 14, 10671074

Issues in clinical nursing

Triage in emergency departments

observed that several EDs in Sweden used a three-level acuity rating system but that the denition of the need for care at the EDs differed. Further, the report suggested that the third acuity level should be divided into three subgroups, giving a ve-level acuity scale with the following time limits: Level 1, in which care occurs within 515 minutes; Level 2, in which care occurs within one to two hours; Level 3, in which care occurs within six hours; Level 4, in which care occurs within 24 hours; and Level 5, in which care occurs within three days. The report was goal-oriented and was thus intended to serve as a supportive document for changes in ED care. The lack of information on standards and guidelines for triage staff and acuity rating scales in Sweden, as well as the absence of published studies in this area, was the reason for the present study.

The study
Aim
The aim of this study was to describe how triage-related tasks were organized and performed in Swedish EDs.

Sample
All Swedish hospitals (n 79) with an ED for somatically ill and injured patients were identied through a national register rdsinformation 2001) and later contacted (HSI Ha lso & Sjukva by letter regarding their participation in the study. The nal sample consisted of 69 EDs, representing 87% of those in Sweden. Of the 10 EDs that did not participate, two did not approve of participation, two declined because of a shortage of staff, two claimed not to have time to participate and two felt it was not possible to nd any personnel to respond to the questions. In the remaining two non-participating EDs, one received further information about the study but chose not to participate and one claimed that their ED was more comparable with an out-patient clinic than an ED. Of the 10 nonparticipating EDs, ve were situated in the northern region and ve in the southern region of Sweden. One of the nonparticipating hospitals was a university hospital, three were county hospitals and six were local hospitals.

for face validity and usability and underwent several revisions before its nal use. We assumed that the word triage was not used in everyday language among the respondents, therefore, the concept of triage was referred to as assessing and prioritizing patients throughout the interview. Based on previous studies and the specic aim of this study, the instrument had four components: demographic data, personnel working in the ED, knowledge of legislation about triage and use of decision support and acuity rating scales. The demographic section included ve items. The section addressing staff categories performing triage, types of triage function performed and what facilities were available when performing triage contained 19 items. Items in this section dealt with what hours triage personnel were on duty, qualications required to perform triage and the role of triage personnel in refusing patient access to the ED. The section focusing on knowledge of standards, guidelines and legislation about triage and the use of decision support consisted of seven items: knowledge of triage guidelines, reasons underlying organization of triage work, access to decision support during triage work and knowledge of triage. The last section covered the use of acuity rating scales, documentation of ratings and the reasons for using the current scale. The total instrument contained 36 items. The majority of questions were closed-ended in combination with the possibility to make additional comments.

Data collection
The medical directors at each of the 79 identied EDs were contacted by letter to obtain approval to participate in the study. Following approval, nurse managers at each of the participating EDs (n 69) were contacted by mail to receive permission and arrange time for a telephone interview. Telephone interviews were selected to increase the number of interviews possible, facilitate data gathering, strengthen validity and increase the response rate. In some cases, the nurse manager appointed another RN or other staff member to respond to the questions. The telephone interviews took place when the respondents were on duty in the ED. Overall, 61 charge nurses or their equivalents, six staff nurses and two people belonging to other personnel categories were interviewed. Data collection took place during the autumn of 2002, with each interview lasting between 20 and 35 minutes. One researcher (KG) conducted all the interviews.

Interview guide
One of the authors (KG) developed a structured interview guide. The content validity of the instrument was informed by earlier triage studies (Purnell 1991, Geraci & Geraci 1994, Palmquist & Lindell 2000). The instrument was piloted twice

Ethical considerations
Before approval of the study (Dnr: CF 18-2003) by the Ethics rebro University, all hospitals gave their Committee at O
1069

2005 Blackwell Publishing Ltd, Journal of Clinical Nursing, 14, 10671074

KE Go ransson et al.

formal approval. Moreover, before conducting the interviews informed consent was obtained from each of the participating nurse managers, nurses and other participants.

Data analysis
Descriptive statistics was used to examine the data. Data analysis was performed using SPSS Version 11.0.

an RN or a Licensed Practical Nurse (LPN) performing this work. A designated RN triaged patients in seven (10%) EDs. In the remaining 62 (90%), all RNs on duty could perform triage with incoming patients. In ve of the seven EDs that used designated RNs, the designated triage nurse was on duty 24 hours a day; in the other two, the designated triage nurse was on duty during day and evening hours. Non-ambulance presenting patients The process through the ED of non-ambulance arriving patients was determined at the reception desk in 56 (81%) of the EDs. One of three staff (a clerk, LPN or RN) made the decision as to whether the patient would have a seat in the waiting room or be attended to at once in the treatment area. All patients (i.e. both waiting room patients and those sent to the ED treatment area) were then given an acuity rating by the nurse in the treatment area. In seven of the remaining 13 (19%) EDs patients went directly from the reception desk to the nurse at the treatment area for triage. In ve EDs, a designated RN performing triage had the authority to decide if the patient was to be sent to the waiting room or to the treatment area; in the remaining EDs, patients were taken either directly to the treatment nurse or to a designated RN for triage. The percentages of personnel performing the rst assessment and prioritization on non-ambulance presenting patients are shown in Table 2. Designated RNs were responsible for triaging non-ambulance arriving patients at 24 EDs (i.e. one-third of the sample). In 14 (58%) of these, a designated RN was scheduled 24 hours a day for triaging. In one (4%) ED, nurses were designated during evenings only, in eight (33%) during both day and evening and in one (4%) during evening and night shifts. Table 3 illustrates the type of hospital and the use of designated RNs performing triage for ambulance and non-ambulance presenting patients. Of the university hospitals, three of six (50%) used designated RNs. Three of four (75%) of the regional hospitals also used designated RNs, while 12 (57%) of the
Table 2 Personnel performing the rst assessment of non-ambulance arriving patients in Swedish EDs (n 69) Either RN or not RN n (%) 11 (16) 13 (19) 20 (29)

Results
The 69 participating hospitals were representative of the Swedish context by hospital type and geographic region (Table 1). The concept of triage was known to 54 (78%) of the 69 respondents. A more common expression to describe the process of assessing and prioritizing patients was prioritizing the patients, where staff members performing this work were often called prioritizing nurses or a prioritizing team. In this study, prioritizing patients was treated as equivalent to triaging patients.

Triage facilities
The availability of triage facilities differed at the various EDs. Of 24 EDs using designated RNs to perform triage, 16 used rooms specically intended for this purpose. The remaining eight EDs had no specic equipped space for triage and used any available area in the ED. Patients presented at the EDs either by ambulance or by other means (i.e. non-ambulance). Non-ambulance patients arrived at the reception area by walking or by wheelchair. The ways in which patients presented to the hospital inuenced which personnel performed triage, assessment and prioritization. To clarify mode of presentation, the terms ambulance and non-ambulance arrival were used.

Triage personnel
Patients presenting by ambulance Patients arriving by ambulance were triaged by an RN in 68 (99%) EDs. In the single remaining ED, there could be either
Table 1 Distribution of the different types of hospital included in the study (n 69) Type of hospital Primary health care n (%) 1 (1)

RN n (%) Daytime Evening Night time 37 (54) 42 (61) 36 (52)

Not RN n (%) 20 (30) 14 (20) 13 (19)

Total n (%) 68* (100) 69 (100) 69 (100)

University n (%) 6 (9)

Regional n (%) 4 (6)

County n (%) 21 (30)

Local n (%) 37 (54)

Total n (%) 69 (100)

*n 68. One ED only operated during evening and night time hours. RN Registered Nurse.

1070

2005 Blackwell Publishing Ltd, Journal of Clinical Nursing, 14, 10671074

Issues in clinical nursing Table 3 Type of hospital in the study and the use of designated RNs for triage (n 24) Type of hospital University Regional County Local General Total n (%) n (%) n (%) n (%) n (%) n (%) Designated 3 (13) RN 3 (13) 12 (50) 5 (20) 1 (4) 24 (100)

Triage in emergency departments

within the personnel categories, depending on different rules about delegating at the EDs. Type of examinations performed during triage varied from merely taking the patients chief complaints to perform a complete check of vital signs, drawing blood samples and ordering an X-ray examination.

Triage scales and acuity rating

Of the participating EDs, 37 (54%) used some kind of triage scale to communicate and document the acuity rating of the patients. Eighteen of these used a triage scale with three levels, 15 used one with four levels and four used ve levels. There were also differences in time intervals connected with each level. The difference occurred both within scales using the same number of levels and between scales with a different number of levels. The acuity rating for level one, however, was equivalent in all triage scales, indicating that the patient was to be assessed by a physician immediately (Table 4). The most common method, which was used by 32 EDs, was to represent the acuity rating through numeric scales. Two EDs used colour-coding to indicate rating, one used text, one used text and colour-coding and one used a numeric scale in combination with text. The remaining 32 (46%) did not use any kind of triage scale. Of these 32 that did not use a triage scale, 18 referred to their working tradition as a reason for not using a scale, whereas the remaining 14 reported other motives. A common argument for not using triage scales was that the ED had so few patient admissions that staff were able to prioritize without the support of standardized methods. Furthermore, some of the EDs stated that staff organized the patients records in a special order in an effort to know which patient the physician was to see next.

county hospitals and ve (13%) of the local hospitals used designated RNs. The only primary healthcare hospital also used designated RNs. The requirements for educational background and clinical experience of these RNs were found to vary widely by EDs. The requirements for educational skills ranged from basic nursing education (21 or 875%) to specic in-service training for triage (2 or 83%), e.g. education about acuity ratings, basics on how to prioritize patients. The RNs clinical ED experience demanded for being scheduled for triage ranged from three (125%) departments reporting that they did not use newly employed RNs for triage to three (125%) that required that the RNs had several years of experience.

Triage assessment
The kind of examination the patients would receive was dependent on their presentation at the ED. For those presenting by ambulance, examinations for triage were being performed in all 69 (100%) EDs, regardless of the use of designated RNs. The most common approach was to examine the patient in any room available, perform various examinations and tests and then conclude by assigning an acuity rating. Examinations performed on non-ambulance arriving patients varied among the EDs. In the 45 (65%) not using designated RNs for triage variation depended on whether an RN, LPN or clerk assessed the patients and the availability of facilities for triage. However, there were also differences
Table 4 Different triage scales used in Swedish EDs (n 37)

Knowledge of triage guidelines and legislation

Fifty-three (77%) of the 69 respondents were not aware of standards and guidelines, legislation or other directing

Time frame for treatment in minutes 3 levels n 18 Level 1 Level 2 Level 3 Level 4 Level 5 Immediately 15120 minutes (No time limit) 180 minutes (No time limit) 4 levels n 15 Immediately 1560 minutes (No time limit) 60180 minutes, 24 hours (No time limit) 120 minutes, 12 hours (No time limit) 5 levels n4 Immediately 3060 minutes 60120 minutes, 6 hours 120 minutes, 24 hours (No time limit) 3 days (No time limit)

2005 Blackwell Publishing Ltd, Journal of Clinical Nursing, 14, 10671074

1071

KE Go ransson et al.

documents for ED triage on a national level. Of the remaining 16, seven (10%) claimed that they knew or thought they knew of such national standards and guidelines, four (6%) reported that they knew of regional standards and guidelines and ve (7%) stated that they had local standards and guidelines. The EDs reported different rationales for using designated RNs for triage, with nine stating other reasons as the main reason. Another ve EDs had worked with local staff teams to organize the work. Four had read about the use of triage nurses and another four responded that tradition was the primary reason for their organization. Two had been stimulated by other EDs to use designated RNs. Of those not using designated RNs for triage, the majority (60%) reasoned that they had never used them before and saw no need to change the routine of the organization.

Discussion
Considerable diversity was identied throughout Sweden in the way in which staff assessed and prioritized patients. The greatest likelihood of being triaged by a designated RN was if the patient arrived by ambulance at a primary healthcare hospital or a regional hospital. The least likelihood of being triaged by a designated RN was if the patient arrived without an ambulance at a local hospital. One explanation of why so few local hospitals used designated RNs for triage could be that the staff are few and therefore it is not possible to have an RN especially scheduled for triage. Another reason could be that, by tradition, the waiting times at the smaller EDs are not very long and therefore there is little or no need to prioritize patients. Further, this survey showed noticeable heterogeneity in the use of triage scales and acuity ratings in Sweden. Almost half of the EDs 46% did not use any kind of triage scale and the scales that were used had different designs and content. The scales in use ranged from three to ve acuity levels with a wide variety of time spans within the levels. This diversity suggests that patient safety during triage is largely dependent on the ED at which patients present and the way (ambulance vs. non-ambulance) in which they arrive. Another effect of this diversity in scales could be that patients do not receive the same care all over the country. A possible effect of the different triage scale designs in use is that, when RNs change work places, they will need to adjust to another triage scale, which could involve some risk to patient safety. It could also have a negative effect on the possibility of performing research and comparing different EDs on acuity ratings. The low number of EDs using any kind of triage scale and the fact that there was a large diversity in design among the scales used are not consistent with ndings from surveys in other
1072

countries, where many use a scale that has been tested for both reliability and validity (Manchester Triage Group 1997, Fernandes et al. 1999, Australasian College for Emergency Medicine 2000). A reliable and safe scale is a fundamental tool when triaging patients at EDs. The diversity in the use and design of triage scales and acuity ratings in Swedish EDs makes it extremely difcult to evaluate the effectiveness and safety of such scales. Between 52 and 61% of Swedish EDs used an RN sometime during the day in triaging non-ambulance arriving patients. In the remaining EDs, it could be either an RN or less qualied personnel who performed the rst assessment of non-ambulance arriving patients. This result is consistent with that reported by Palmquist and Lindell (2000), although 56% of the EDs in their study claimed to be changing their way of triaging patients. In other developed countries (e.g. the USA and Australia), RNs are more often assigned to triage patients. The nding that nearly half of the EDs in our study used either an RN, with or without specic training for triage, or less qualied personnel for the rst assessment is, from a safety perspective, alarming. In this study, a designated RN was scheduled for triaging patients 24 hours a day in 20% of the EDs while in 35% a designated RN was scheduled for triaging patients sometime during the day. Among those using designated RNs, about two-thirds were organized such that it was possible for the nurse to perform different examinations and checks during triage. In contrast to other countries use of, and recommendations for the use of, RNs specially trained for triaging patients, this study shows that Swedish EDs did not adhere well to established standards and guidelines for emergency care triage (Purnell 1991, Manchester Triage Group 1997, Australasian College for Emergency Medicine 2000). Our ndings show that the vast majority of respondents knew that triage referred to the process of assessing and prioritizing patients. However, it was not common to use the word triage to describe this process. Our preunderstanding and belief that the word triage was not used in everyday language in EDs were justied. It was a positive nding that a large number of respondents were familiar with the concept because this will facilitate future communication in Sweden about triage. The use of designated RNs for triage, and the absence of a nationally accepted ve-level triage scale, differs from some other countries. These countries (e.g. Australia, Canada and the UK) have performed research in the area for several years, and consequently, seem to have well organized systems for triaging patients in EDs. To varying extents, they all use designated RNs. There are different views on the use of such RNs, but several studies have concluded that RNs are suitable

2005 Blackwell Publishing Ltd, Journal of Clinical Nursing, 14, 10671074

Issues in clinical nursing

Triage in emergency departments

for performing triage. However, there is a difference in these countries concerning the triage scales. Although the CTAS has its foundations in the Australasian National Triage Scale (NTS), they have different time levels for each category. In conclusion, although there exist disparate opinions about triage scales and the use of RNs for this work, it is evident that several countries use the scales and RNs to a great extent. There may be several reasons why the Swedish EDs are not as well organized as in some other countries. In Sweden, the use of ED physicians is a relatively new phenomenon, and to date such physicians are used in only a few hospitals. Physicians working in EDs are normally scheduled on an oncall basis at the ED and therefore seldom have a greater commitment to the department. Consequently, it is very much up to the personnel ordinarily stafng the ED, i.e. RNs and LPNs, to initiate and carry out changes. In Sweden, our understanding is that RNs and LPNs have little power and therefore their possibility of inuencing care is minimal. By tradition and law in Sweden, every individual should be treated equally (SFS 1982:763). As late as 1997, an addition was made to healthcare regulations on the need to prioritize patients (SFS 1997:142), and in 1998 another regulation was added that a medical examination of attending patients health status should be made as soon as possible (SFS 1998:1659). It is not clear whether these additions to the law have had any impact on clinical practice on ED care. In our study, only 10% of the EDs reported that they knew of some standards and guidelines or legislation on a national level about qualications for personnel performing triage. Therefore, it can be assumed that knowledge is lacking about the law on prioritization of patients. However, the need for prioritizing patients at the ED is growing in Sweden because the numbers of patients seeking health care are increasing. This increase is the result of several EDs throughout the country being closed for political reasons, and the fact that Swedes are following international trends and seeking care at EDs to a greater extent than before. This will have considerable impact on all types of ED. Those that are still operating will have increasing numbers of patients and therefore longer waiting times, which calls for well-trained triage nurses and reliable triage scales. The lack of knowledge about standards and guidelines or other governing documents on ED triage may partly be a result of the absence of operational and research attention given to this issue in Sweden. One way of changing knowledge levels on triage is through research, as well as to ensure the dissemination of research ndings to EDs and decision-makers. Further, the decision-making skills of RNs in performing triage need to be further explored. In the years to come, new and improved educational programmes in triage will be needed.

One limitation of the present study is that not all EDs in Sweden participated. Those that did participate, however, we consider representative of the entire country, both geographically and according to hospital type. Thus, there is no reason to believe that this absence of respondents signicantly affected the results. Another limitation is that the instrument used was designed expressly for this study and therefore might have inadvertently affected the results. The fact that some respondents were nursing staff or from other personnel categories and not charge nurses might have inuenced their knowledge of standards and guidelines about use of triage personnel, scales and acuity ratings, as well as the reasons for the current organization of triage work at their departments.

Conclusion
This study clearly shows that work associated with ED triage in Sweden is not organized uniformly or systematically. It identies the need for further research on ED triage, especially in the areas of personnel performing triage, triage scales and standards and guidelines. Further knowledge about triaging is required in order to develop personnel, educational and organizational aspects of this complex work.

Relevance to clinical practice

The diversity among several aspects of nursing triage (e.g. use of less qualied personnel performing triage, the use of different triage scales) presented in the study points to a safety risk for the patients. It also shows the need of further education for the personnel in clinical practice as well as further research on triage in order to gain national consensus about this nursing task.

Contributions
Study design: KG, AE, ME; data collection and analysis: KG, AE, ME; manuscript preparation: KG, AE, ME.

References
Australasian College for Emergency Medicine (ACEM) (2000) Guidelines for the Implementation of the Australasian Triage Scale in Emergency Departments. Available at: http://www.acem.org.au/ open/documents/triageguide.htm (accessed 15 December 2003). Beveridge R (1998) The Canadian triage and acuity scale: A new and critical element in health care reform. The Journal of Emergency Medicine 16, 507511. Beveridge R (2000) The Science of Triage. Available at http:// www.saem.org/download/kelly.pdf (accessed 12 December 2003).

2005 Blackwell Publishing Ltd, Journal of Clinical Nursing, 14, 10671074

1073

KE Go ransson et al. Cioffi J (1998) Decision making by emergency nurses in triage assessments. Accident and Emergency Nursing 6, 184191. Estrada E (1981) Triage systems. Nursing Clinics of North America 16, 1324. Fernandes CM, Wuerz R, Clark S & Djurdev O (1999) How Reliable is Emergency Department Triage? Annals of Emergency Medicine 34, 141147. Geraci EB & Geraci TA (1994) An observational study of the emergency triage nursing role in a managed care facility. Journal of Emergency Nursing 20, 189194. Gerdtz MF & Bucknall T (2000) Australian triage nurses decisionmaking and scope of practice. Australian Journal of Advanced Nursing 18, 2433. Gerdtz MF & Bucknall T (2001) Triage nurses clinical decision making: an observational study of urgency assessment. Journal of Advanced Nursing 35, 550561. rdsinformation (2001) Fo HSI Ha rteckning o ver Sjukh lso & Sjukva usanla ggningar, Register of Hospitals (in Swedish). HIS Fo rlag AB, Ma rsta. Handysides G (1996) Triage in Emergency Practice. Mosby, St. Louis. Manchester Triage Group (1997) In Emergency Triage (MacwayJones K ed.). BMJ Publishing Group, London. Palmquist I & Lindell G (2000) Emergency Departments in Sweden rd i Norden 20, 2831. today and in the future. Va Purnell LDT (1991) A survey of emergency department triage in 185 hospitals: Physical facilities, fast-track systems, patient-classification systems, waiting times, and qualification, training, and skills of triage personnel. Journal of Emergency Nursing 17, 402407. Read S, George S, Westlake L, Williams B, Glasgow J & Potter T (1992) Piloting an evaluation of triage. International Journal of Nursing Studies 29, 275288. Rock D & Pledge M (1991) Prioritise of care for the walking wounded. Triage in accident and emergency. Professional Nurse 6, 463465. rdslag, The Health Care Act (in SweSFS (1982). Ha lso- och sjukva dish). Svensk fo rfattningssamling 1982:763, Liber: Allma nna Fo rlaget, Stockholm. rdslag, Addition to The Health SFS (1997) Tilla gg till Ha lso- och sjukva Care Act (in Swedish). Svensk fo rfattningssamling 1997:142, Liber: Allma nna Fo rlaget, Stockholm. rdslag, Additon to The Health SFS (1998) Tilla gg till Ha lso- och sjukva Care Act (in Swedish). Svensk fo rfattningssamling 1998:1659, Liber: Allma nna Fo rlaget, Stockholm. Shields JE (1976) Making Triage Work: The experiences of an Urban Emergency Department. Journal of Emergency Nursing 2, 3741. Socialstyrelsen (1994) Akut omha ndertagande Ett underlag fo r kompetensutveckling vid omha ndertagande av akut sjuka och skadade, Emergency Care A Base for Competence Development of Emergency Care of the Acute Ill and Injured (in Swedish). http:// sos.se/mars/pub/005/pur006.htm/ Thompson JD & Dains JE (1982) Comprehensive Triage: A Manual for Developing and Implementing a Nursing Care System. Reston Publishing Company, Inc., Virginia. Wuerz R, Fernandes CM & Alarcon J (1998) Inconsistency of Emergency Department Triage. Annals of Emergency Medicine 32, 431435.

1074

2005 Blackwell Publishing Ltd, Journal of Clinical Nursing, 14, 10671074