- Email: [email protected]
Regular Article
Accident and Emergency Nursing (2001) 9, 101–108 © 2001 Harcourt Publishers Ltd doi: 10.1054/ aaen.2001.0209, available online at http://www.idealibrary.com on
Clinical Decisions using the National Triage Scale: how important is postgraduate education? Julie Considine RN, RM, BN, CertAcuteCareNsg (Emerg), GradDipNsg(Acute Care), MN, PhD candidate Clinical Nurse Educator, Emergency Department, Dandenong Hospital, David Street, Dandenong, 3175, Victoria, Australia Tel.: 03 9554 1000 pager 7374; Fax: 03 9554 8453 E-mail: jconsidine @alphalink.com.au Prof. Lerma Ung. Chair, Acute Care Nursing, Cabrini/Deakin Professorial Unit, Cabrini Hospital, 183 Wattletree Rd, Malvern, 3144 Victoria. Tel.: 9508 1905; Fax: 9508 1908 Prof. Shane Thomas BA (Hons); Dip Public Policy; Ph.D. Adjunct Professor, Faculty of Health and Behavioural Sciences, Deakin University Cabrini/Deakin Professorial Unit, Cabrini Hospital, 183 Wattletree Rd. Malvern, 3144, Victoria Correspondence to: Julie Considine Manuscript Accepted: 13 December 2000
J. Considine, L. Ung, S. Thomas Triage is the formal nursing assessment of all patients who present to an Emergency Department (ED). The National Triage Scale (NTS) is used in most Australian EDs. Triage decision making involves the allocation of every patient presenting to an ED to one of the five NTS categories. The NTS directly relates a triage category to illness or injury severity and need for emergency care. Triage nurses’ decisions not only have the potential to impact on the health outcomes of ED patients, they are also used, in part, to evaluate ED performance and allocate components of ED funding. This study was a correlational study that used survey methods. Triage decisions were classified as ‘expected triage’, ‘overtriage’ or ‘undertriage’ decisions. Participant’s qualifications were allocated to five categories: ‘nil’; ‘emergency nursing’; ‘critical care nursing’; ‘midwifery’; and ‘tertiary’ qualifications. There was no correlation between triage decisions and length of experience in emergency nursing or triage. ‘Expected triage’ decisions were more common when the predicted triage category was Category 3 (P < 0.001) and ‘overtriage’ decisions were less common when the predicted triage category was Category 2 (P < 0.0010). The frequency of ‘undertriage’ decisions decreased significantly when the predicted triage category was Category 3 (P < 0.001) or Category 4 (P < 0.001). There was no correlation between triage decisions and qualifications in the ‘nil’, ‘emergency nursing’ or ‘critical care nursing’ categories. A midwifery qualification demonstrated a positive correlation with ‘expected triage’ decisions (P = 0.048) and a negative correlation with ‘undertriage’ decisions (P = 0.012). There was also a positive correlation between a tertiary qualification and ‘expected triage’ decisions (P = 0.012). © 2001 Harcourt Publishers Ltd
Introduction Triage is the formal process of immediate assessment of all patients who present to an Emergency Department (ED) (Zwicke et al. 1982; Commonwealth Department of Health and Family Services and the Australasian College for Emergency Medicine 1997; Rowe 1992; Williams 1992). On the basis of this assessment, patients are classified into groups relating to illness or injury severity and need for medical and nursing care. This process of prioritization ensures immediate intervention and greatest resource allocation to
© 2001 Harcourt Publishers Ltd
those with life threatening illness or injury (George et al. 1993; Geraci and Geraci 1994; McDonald et al. 1995; Commonwealth Department of Health and Family Services and the Australasian College for Emergency Medicine 1997). In Australia, triage is predominantly a nursing role. The National Triage Scale (NTS) was formulated in 1993 by the Australasian College for Emergency Medicine with the aim of promoting a standardized approach to triage. The NTS is a five category scale and is used in most EDs throughout Australia. The five categories are shown in Table 1.
Accident and Emergency Nursing ( 2 0 0 1 )
9, 101–108
101
Clinical decisions using the National Triage Scale
Table 1 National Triage Scale for Australian emergency departments adapted from: Australasian college for emergency medicine 1993b, p10. Triage Code
Treatment Acuity*
National Triage Scale
Category 1
Immediate
Resuscitation
Category 2
Within 10 minutes
Emergency
Category 3
Half Hour
Urgent
Category 4
One Hour
Semi Urgent
Category 5
Two Hours
Non Urgent
* for medical care
All patients who present to an ED are allocated to one of the NTS categories, and it is a nurse who usually makes this decision. Triage decisions are made under conditions of uncertainty with limited or obscure information, minimal time and little margin for error (Edwards 1994; Brillman et al. 1996). Given the difficulties of clinical decision-making at triage, it may be argued triage nurses require greater clinical decision-making skills than nurses performing general emergency nursing duties (Purnell 1991). The triage nurse’s decision to allocate a triage category is based on a clinical assessment that determines illness or injury severity and the need for medical care (Australasian College for Emergency Medicine 1993b; Commonwealth Department of Health and Family Services and the Australasian College for Emergency Medicine 1997). At the present time, the NTS gives no formal consideration to the care provided by emergency nurses whilst the patient is waiting to see a doctor. Given the inherent difficulties of clinical decision making in the triage context, it is well documented by many authors that triage nurses require a certain degree of educational preparation. Descriptions of desirable, if not imperative, attributes of a triage nurse have included ‘highly competent’, ‘qualified’, ‘suitably experienced and trained’, and ‘appropriately skilled’ (Mallett and Woolwich 1990; Williams 1992; Australasian College for Emergency Medicine 1993; Commonwealth Department of Health and Family Services and the Australasian College for Emergency Medicine 1997, p5). Further definition of these qualities was not provided. It is also recognized that the
102
Accident and Emergency Nursing ( 2 0 0 1 )
9, 101–108
experiential and educational prerequisites of Victorian triage nurses are varied (Standen and Dilley 1997). Three months of emergency nursing experience is all that is required by some EDs before being assigned to triage, whilst other EDs restrict triage to nursing staff with a post graduate qualification in emergency nursing. It is important to note that the decisions made by triage nurses are a fundamental factor in the initiation of emergency care. Consequently, the decisions made by triage nurses have the potential to impact on the health outcomes of those patients seeking emergency care (Williams 1992; Commonwealth Department of Health and Family Services and the Australasian College for Emergency Medicine 1997). The uses for the NTS have extended beyond prioritising emergency care. The NTS is currently used to determine elements of ED funding, resource allocation and measurement of ED performance (Doherty 1996; Fitzgerald 1996; Jelinek and Little 1996; Standen and Dilley 1997). This clearly contradicts the National Triage Scale User Manual which states that “Triage requires a clinical decision, which has regard only to the patient’s individual need for care.” (Commonwealth Department of Health and Family Services and the Australasian College for Emergency Medicine 1997). Given that nurses usually make triage decisions, their decisions now have far reaching implications and therefore require critical analysis.
Research aims The aim of this research was to identify any relationships between triage nurses’ decisions on triage category allocation and the type and length
© 2000 Harcourt Publishers Ltd
Clinical decisions using the National Triage Scale
of nursing experience and/or the level of educational preparation of the triage nurse.
Method A correlational design using survey methods was employed in this study. This design was selected as it allowed for the identification of relationships between variables whilst allowing answers to be sought from participants based on the same amount of information.
Setting The study was conducted in two Victorian EDs. One ED was in a large Melbourne metropolitan teaching hospital that provided a wide range of general and specialist adult services as well as being a major health care provider for the poor and disadvantaged. The other ED was in a Victorian rural base hospital that provides general adult and paediatric services and operates an accident retrieval service. This hospital is situated in a large country town that is surrounded by isolated farming properties.
Instrument A self-administered questionnaire was developed by the researcher, examined by nominated experts and piloted to assess clarity and ease of administration. Participants were asked to complete a section on demographic details that asked about each participant’s years of experience in emergency and triage nursing, and level of educational preparation in nursing. Participants were then required to allocate one of the NTS categories to each of ten patient scenarios. Only adult scenarios were addressed and specialty triage scenarios, for example, paediatrics and obstetrics were excluded as these were only seen routinely at one of the two participating EDs. Patient scenarios have, in the past, been considered a valid method of evaluating analytical, problem solving and decision making skills (Holzemer et al. 1981; Carroll and Johnson 1990). The use of patient scenarios enabled the researcher to elicit responses from all participants based on the same information. The variability in the patients’ condition that would occur in real
© 2001 Harcourt Publishers Ltd
life triage episodes was also minimised using this method. The researcher developed the patient scenarios based on real life triage episodes. There were two patient scenarios suited to each of the five NTS categories. A panel of peer nominated expert triage nurses completed the patient scenarios and their responses were compared with the response expected by the researcher. Complete (100%) agreement was reached between the researcher and the expert triage nurses. The agreed triage category for each scenario was referred to as the ‘predicted triage category’. As the patient scenarios used in this study have been published elsewhere, they will not be outlined in detail in this paper (Considine et al. 2000). Participants’ responses were classified into one of three triage decisions. If the participant allocated the same triage category as the predicted triage category, the decision was termed ‘expected triage’ decision. Allocation of a triage category of higher acuity than expected was termed ‘over triage’ decision. This decision indicates that waiting time for the patient would be less than anticipated and therefore would not be detrimental to the patient being triaged. This triage decision may, however, result in the inappropriate use of staff and resources and may be detrimental on a departmental basis. ‘Under triage’ decision was used to describe the allocation of a triage category of a lower acuity than expected, resulting in the patient waiting longer than anticipated and is potentially detrimental to patient safety (Considine et al. 2000; Hollis & Sprivulis 1996).
Sampling Process The population for the study consisted of all the Registered Nurses working in the two EDs selected. As this study focused on triage decision making, only those nurses who were able to be rostered to the triage role were eligible to participate in the study. The total number of registered nurses who were able to be rostered to triage, and therefore met the inclusion criteria for the study was 42. Given the small numbers of triage nurses, random sampling was not employed and convenience sampling was used. All 42 triage nurses were invited to participate in the study and 31 triage nurses responded. This gave a response rate of 74%.
Accident and Emergency Nursing ( 2 0 0 1 )
9, 101–108
103
Clinical decisions using the National Triage Scale
Data Collection
Table 2
Questionnaires were distributed to all nursing staff who were eligible to be rostered to the triage role in the EDs of the participating institutions. Consent was implied by the return of the questionnaire and confidentiality and anonymity were assured.
Participant’s Qualification Categories n
Nil qualification
f (%)
5
16.1
Emergency nursing qualification
11
35.4
Critical care nursing qualification
13
41.9
6
19.3
16
51.6
Midwifery qualification
Data Analysis The computer software SPSS ® Version 6.1 was used for data analysis. Frequency and descriptive statistics were obtained and Spearman’s correlation was used to identify the existence of correlations between triage decisions and the type and length of nursing experience and educational preparation. Chi-square was used to identify significant differences in triage decisions for each of the five NTS categories. Statistical significance was identified by a probability value of less than 5% (P < 0.05).
Results Characteristics of participants When participants’ years of experience in emergency nursing was examined, 29.1% of participants had two or less years experience and 32.2% of participants had ten or more years of emergency nursing experience. Analysis of participants’ years of triage experience revealed that 32.3% of participants had two or less years of triage experience, almost half of the participants (48.4%) reported three to six years triage experience and 19.3% had ten or more years of triage experience. As the educational preparation of participants was varied, five major categories were established to classify participants’ qualifications. These categories were: i) Nil qualification ii) Emergency nursing qualification iii) Critical care nursing qualification iv) Midwifery qualification v) Tertiary qualification Nil qualification was used to describe participants with no post registration and no tertiary qualifications in nursing. Emergency
104
Accident and Emergency Nursing ( 2 0 0 1 )
9, 101–108
Tertiary qualification
31
100
nursing qualification refers to participants with a hospital certificate, graduate certificate or graduate diploma in emergency nursing or who had successfully completed the Trauma Nursing Core Course. Critical care nursing qualification was used to encompass participants with a hospital certificate, graduate certificate or graduate diploma in intensive care, coronary care or critical care nursing. Midwifery qualification was used to describe participants who were Registered Midwives and tertiary qualification was used to describe participants with a Bachelor’s Degree or Graduate Diploma in Nursing. There were no participants who held a Master’s Degree. Approximately half (51.7%) the participants had qualifications in two (2) or more categories.
Triage decisions and experience Each of the 31 participants completed the same ten patient scenarios giving a total of 310 triage episodes. Of these, 58% were ‘expected triage’ decisions, 21% were ‘overtriage’ decisions and 21% were ‘undertriage’ decisions. No correlation was identified between triage decisions and length of experience in either emergency nursing or triage.
Triage decisions and National Triage Scale categories Participant’s triage decisions were then examined in relation to the five NTS categories. The frequency of ‘expected triage’ decisions was significantly higher when the predicted triage category was Category 3 (χ2 = 20.903,
© 2001 Harcourt Publishers Ltd
Clinical decisions using the National Triage Scale
Table 3
Participants’ triage decisions for each National Triage Scale Category Undertriage decision f(%)
Expected triage decision f (%)
Overtriage decision f (%)
NTS Category 1
40.3
59.7
0.0
NTS Category 2
38.7
58.1
3.2*
NTS Category 3
21.0*
79.0*
0.0
NTS Category 4
4.8*
54.8
40.3
NTS Category 5
0.0
40.3
59.7
*P < 0.05
df = 1, P < 0.001). There was a significant reduction in the frequency of ‘overtriage’ decisions when the predicted triage category was Category 2 (χ2 = 54.258, df = 1, P < 0.001). The frequency of ‘undertriage’ decisions decreased significantly when the predicted triage category was Category 3 (χ2 = 20.903, df = 1, P < 0.001) or Category 4 (χ2 = 50.581, df = 1, P < 0.001). It should be pointed out that an ‘undertriage’ decision is not possible for a predicted triage category of Category 5 and an ‘overtriage’ decision is not possible for a predicted triage category of Category 1. This reduces the potential margin for error in triage decision-making within these categories.
Triage decisions and educational preparation of triage nurses No correlation was found between triage decisions and qualifications in the nil, emergency nursing and critical care nursing categories. There was a positive correlation between a midwifery qualification and ‘expected triage’ decisions (rs = 0.3580, P = 0.048) and a negative correlation between a midwifery qualification and ‘undertriage’ decisions (rs = – .4437, P = 0.012). These findings suggest that triage nurses with a midwifery qualification had a higher incidence of ‘expected triage’ decisions and were less inclined to ‘undertriage’. There was a positive correlation between a qualification in the tertiary qualification category and ‘expected triage’ decisions (rs = 0.3859, P = 0.012).
Discussion The findings of this study challenge the assumption that better clinical decision-making is
© 2001 Harcourt Publishers Ltd
Table 4 Correlation of expected triage decisions with participant’s qualifications Expected Triage Decisions rs
P
Nil Qualification
–0.110
.556
Emergency Nursing Qualification
–0.207
.236
Critical Care Qualification
–0.120
.521
Midwifery Qualification
0.358
.048*
Tertiary Qualification
0.386
.032*
* P < 0.05 Table 5 Correlation of undertriage triage decisions with participant’s qualifications Undertriage Decisions rs
P
Nil Qualification
0.271
.140
Emergency Nursing Qualification
0.124
.508
Critical Care Qualification
–0.0120
.521
Midwifery Qualification
–0.435
.015*
Tertiary Qualification
–0.207
.264
* P < 0.05
an unequivocal outcome of advanced experience and specialty educational preparation. The absence of identifiable correlations between
Accident and Emergency Nursing ( 2 0 0 1 )
9, 101–108
105
Clinical decisions using the National Triage Scale
experience in emergency nursing and triage nursing disputes the notion that experienced nurses make more accurate decisions than their less experienced counterparts. Jelinek and Little (1996) concluded that triage responses were unaffected by triage nurse experience. Likewise, a study of clinical decision-making by critical care nurses by Sims and Fought (1989) also reported no relationship between accuracy of decisionmaking and length of experience. Questions should be raised as to the value of chronological measurement of experience and broader criteria such as the quality of nursing experience should be examined. The reason for significant differences in triage decisions related to specific triage categories is unclear. Almost three-quarters (79.0%) of participants arrived at an ‘expected triage’ decision when the predicted triage category was Category 3. There is no available literature to assist in expiation of this finding. It may be postulated that Category 3 is perceived as a ‘safe choice’ for triage nurses as it is the ‘middle’ category in the NTS. Selection of Category 3 neither implies that the patient’s presenting problem is of an emergent or semi-urgent nature. This may be a preferable option to triage nurses who are unsure of the severity of a patient’s presenting problem. Only 3.2% of participants arrived at an ‘overtriage’ decision when the predicted triage category was Category 2. Again, there is no available literature to explain this finding. As ‘overtriage’ decisions are associated with inappropriate use of ED resources and personnel, it is encouraging that the only significant finding for overtriage decisions was the decrease in overtriage decisions for Category 2. It may be postulated that this reduction in overtriage decisions is related to perceived consequence. An ‘overtriage’ decision for a Category 2 patient would result in the allocation of the NTS Category 1. As medical care is required immediately for Category 1 patients, the decision to allocate Category 1 often results in many staff becoming aware of the patients’ presence in the ED. As the ability of an ED to see Category 1 patients immediately is important when assessing ED performance and allocating ED funding, allocation of Category 1 may also result in increased scrutiny at both departmental and organizational levels.
106
Accident and Emergency Nursing ( 2 0 0 1 )
9, 101–108
The decrease in ‘undertriage’ decisions when the predicted triage category was Category 3 or 4 was also an unexpected finding. Given that ‘undertriage’ decisions have the potential to be detrimental to patient safety, it is reassuring that the significant findings in terms of ‘undertriage’ decisions all relate to decreases in the frequency of ‘undertriage’ decisions. One possible explanation is that triage nurses who are unsure of the actual or potential severity of a patients’ presenting problem may ‘err on the side of caution’ and make an overtriage decision. The lack of identifiable correlations between triage decisions and further educational preparation raises questions as to the adequacy of the educational preparation of triage nurses. Participants with no post registration and no tertiary qualifications in nursing have an understandable explanation for deficits in clinical decision making. It is of concern, however, that triage nurses with specialist educational preparation in emergency nursing and critical care nursing also appear to have deficiencies in these areas. This result bears similarity to a study by Sims and Fought (1989) who reported that certification in critical care nursing made no significant difference in decision making when compared with nurses who were not certified. The relationship between triage decisions and a midwifery qualification was a surprising finding and reason for this is unclear. One possible explanation may be the anecdotal similarities shared by nursing practice and clinical decision-making in the areas of triage and midwifery, in particular, delivery suite. In both practice settings there is an absence of routine and patients may present to either area at any time and with varying degrees of acuity. Both areas of practice involve autonomous decision making by nurses based on patient assessment and the setting of priorities related to patient acuity. It may be postulated that the assessment and clinical decision making skills of midwives may be applied to any practice setting, however more research in this area is warranted. The positive correlation between triage decisions and tertiary qualifications may be explained by the higher degree of educational preparation in this group of participants. This notion is supported by the literature. It is stated by del Bueno (1983) that educational preparation is positively correlated with
© 2001 Harcourt Publishers Ltd
Clinical decisions using the National Triage Scale
acceptable decision-making and that nurses with higher educational preparation had a higher incidence of correct decisions. Pardue (1987) and Holl (1994) also state that critical thinking ability and decision-making were better in nurses who had a higher standard of education. If formal education within a nursing specialty does not show any relationship with improved decision making in that specialty, questions should be raised as to how nurses should be educated for specialty practice, specifically the quality of clinical and formal educational preparation of triage nurses. It may be hypothesized that education in areas such as critical thinking, clinical decision making and problem solving are a more adequate preparation for clinical practice than simply imparting specific knowledge about specialist nursing, for example emergency nursing or triage.
Limitations This study was limited by its sample size of 31 participants. This limitation occurred because of the relatively small number of both emergency nurses and triage nurses. Scenario based assessment of triage decisions may be seen by some as a limitation, although the validity of the use of scenarios has been acknowledged in the literature (Holzemer et al 1981; Carroll and Johnson 1990). With regard to the statistically significant findings concerning participants with a qualification in the midwifery category, it should be noted that there were only six (6) participants (19.3%) in this group. To overcome this limitation, replication of the study with a larger sample size and a larger proportion of participants with a midwifery qualification is required.
Recommendations There is a need for a review of the educational preparation of triage nurses with specific content, length and scope of nursing specialty education. Clinical decision-making at both undergraduate and postgraduate levels should be examined, as should the differences in the decision-making skills of beginning and specialist practitioners. There is a need for further research into the effect of educational preparation on decisions made by triage nurses and, more specifically, their application of the National Triage scale.
© 2001 Harcourt Publishers Ltd
Critique of the clinical decision making abilities of those responsible for educating triage nurses is of paramount importance and questions regarding the clinical credibility and recency of practice of those teaching nursing specialty courses should be raised. There is a need for clear definition about the difference in expected performance of generalist and specialist nurses, and in more specific terms, the decision-making skills of these groups. Consideration should be given to the clear definition of specialty practice education in emergency nursing as opposed to generic critical care courses. In nursing, the importance of chronological experience alone as valuable criteria for advancement should be questioned and other ways of assessing the quality and value of clinical experience need to be explored. The use of clinical fellowship and mentors in clinical practice for developing specialty practitioners needs to be explored as a means of ensuring that clinical experience has value in improving practice. References Australasian College for Emergency Medicine 1993a Policy document: triage Australasian College for Emergency Medicine 1993b Position paper: a National Triage Scale for Australian Emergency Departments Brillman J, Doezema D, Tandberg D, Sklar D, Davis K, Simms S, Skipper B 1996 Triage: Limitations in predicting the need for emergency care and hospital admissions. Annals of Emergency Medicine 27(4): 493–500 Carroll J, Johnson E 1990 Decision research: a field guide. Sage Publications: Newbury Park Commonwealth Department of Health and Family Services and the Australasian College for Emergency Medicine 1997 The Australian National Triage Scale: a user manual 1997 Considine J, Ung L, Thomas S 2000 Triage nurses’ decisions using the National Triage Scale for Australian emergency departments. Accident & Emergency Nursing 8 (4): 201–209 del Bueno D 1983 Doing the right thing: nurses’ ability to make clinical decisions. Nurse Educator 8 (3): 7–11 Doherty S 1996 Application of the National Triage Scale is not uniform Australian Emergency Nursing Journal 1 (1): 26 Edwards B 1994 Telephone triage: how experienced nurses reach decisions. Journal of Advanced Nursing 19: 717–724 Fitzgerald G 1996 The National Triage Scale. Emergency Medicine 8: 205–206 George S, Read S, Williams B 1993 Nurse triage. British Medical Journal 306 (Jan): 208
Accident and Emergency Nursing ( 2 0 0 1 )
9, 101–108
107
Clinical decisions using the National Triage Scale
Geraci E, and Geraci T, 1994 An observational study of the emergency triage nursing role in a managed care facility. Journal of Emergency Nursing 20 (3): 194–198 Holl R 1994 Characteristics of the registered nurse and professional beliefs and decision making. Critical Care Quarterly 17 (3): 60–66 Hollis P, Sprivulis P 1996 Reliability of the National Triage Scale with changes in emergency department activity levels. Emergency Medicine 8: 231–234 Holzemer W, Schleutermann J, Farrand L, Miller G 1981 A validation study: simulations as a measure of nurse practitioners’ problem-solving skills. Nursing Research 30 (3): 139–144 Jelinek G, Little M 1996 Inter-rater reliability of the National Triage Scale over 11,500 simulated occasions of triage. Emergency Medicine 8: 205–206 Mallett J, Woolwich C 1990 Triage in accident and emergency departments. Journal of Advanced Nursing 15: 1443–1451 McDonald L, Butterworth T, Yates D 1995 Triage: a literature review 1985–1993. Accident and Emergency Nursing 3: 201–207
108
Accident and Emergency Nursing ( 2 0 0 1 )
9, 101–108
Pardue S 1987 Decision-making skills and critical thinking ability among associate degree, diploma, baccalaureate and masters prepared nurses. Journal of Nursing Education 26 (9): 354–361 Purnell L, 1991 A Survey of emergency department triage in 185 hospitals. Physical facilities, fast track systems, patient classification systems, waiting times and qualifications, training and skills of triage personnel. Journal of Emergency Nursing 17 (6): 402–407 Rowe J 1992 Triage assessment tool. Journal of Emergency Nursing 18 (6): 540–544 Sims K, Fought S 1989 Clinical decision making in critical care. Critical Care Nursing Quarterly 12 (3): 79–84 Standen P, Dilley S 1997 A review of triage nursing practice and experience in Victorian public hospitals. Emergency Medicine 9: 301–305 Williams G. 1992 Sorting out triage. Nursing Times 88 (30): 34–36 Zwicke D, Bozien W, Wagner E 1982. Triage nurse decisions: A prospective study. Journal of Emergency Nursing 8 (3): 132–138
© 2001 Harcourt Publishers Ltd
Clinical Decisions using the National Triage Scale: how important is postgraduate education? Julie Considine RN, RM, BN, CertAcuteCareNsg (Emerg), GradDipNsg(Acute Care), MN, PhD candidate Clinical Nurse Educator, Emergency Department, Dandenong Hospital, David Street, Dandenong, 3175, Victoria, Australia Tel.: 03 9554 1000 pager 7374; Fax: 03 9554 8453 E-mail: jconsidine @alphalink.com.au Prof. Lerma Ung. Chair, Acute Care Nursing, Cabrini/Deakin Professorial Unit, Cabrini Hospital, 183 Wattletree Rd, Malvern, 3144 Victoria. Tel.: 9508 1905; Fax: 9508 1908 Prof. Shane Thomas BA (Hons); Dip Public Policy; Ph.D. Adjunct Professor, Faculty of Health and Behavioural Sciences, Deakin University Cabrini/Deakin Professorial Unit, Cabrini Hospital, 183 Wattletree Rd. Malvern, 3144, Victoria Correspondence to: Julie Considine Manuscript Accepted: 13 December 2000
J. Considine, L. Ung, S. Thomas Triage is the formal nursing assessment of all patients who present to an Emergency Department (ED). The National Triage Scale (NTS) is used in most Australian EDs. Triage decision making involves the allocation of every patient presenting to an ED to one of the five NTS categories. The NTS directly relates a triage category to illness or injury severity and need for emergency care. Triage nurses’ decisions not only have the potential to impact on the health outcomes of ED patients, they are also used, in part, to evaluate ED performance and allocate components of ED funding. This study was a correlational study that used survey methods. Triage decisions were classified as ‘expected triage’, ‘overtriage’ or ‘undertriage’ decisions. Participant’s qualifications were allocated to five categories: ‘nil’; ‘emergency nursing’; ‘critical care nursing’; ‘midwifery’; and ‘tertiary’ qualifications. There was no correlation between triage decisions and length of experience in emergency nursing or triage. ‘Expected triage’ decisions were more common when the predicted triage category was Category 3 (P < 0.001) and ‘overtriage’ decisions were less common when the predicted triage category was Category 2 (P < 0.0010). The frequency of ‘undertriage’ decisions decreased significantly when the predicted triage category was Category 3 (P < 0.001) or Category 4 (P < 0.001). There was no correlation between triage decisions and qualifications in the ‘nil’, ‘emergency nursing’ or ‘critical care nursing’ categories. A midwifery qualification demonstrated a positive correlation with ‘expected triage’ decisions (P = 0.048) and a negative correlation with ‘undertriage’ decisions (P = 0.012). There was also a positive correlation between a tertiary qualification and ‘expected triage’ decisions (P = 0.012). © 2001 Harcourt Publishers Ltd
Introduction Triage is the formal process of immediate assessment of all patients who present to an Emergency Department (ED) (Zwicke et al. 1982; Commonwealth Department of Health and Family Services and the Australasian College for Emergency Medicine 1997; Rowe 1992; Williams 1992). On the basis of this assessment, patients are classified into groups relating to illness or injury severity and need for medical and nursing care. This process of prioritization ensures immediate intervention and greatest resource allocation to
© 2001 Harcourt Publishers Ltd
those with life threatening illness or injury (George et al. 1993; Geraci and Geraci 1994; McDonald et al. 1995; Commonwealth Department of Health and Family Services and the Australasian College for Emergency Medicine 1997). In Australia, triage is predominantly a nursing role. The National Triage Scale (NTS) was formulated in 1993 by the Australasian College for Emergency Medicine with the aim of promoting a standardized approach to triage. The NTS is a five category scale and is used in most EDs throughout Australia. The five categories are shown in Table 1.
Accident and Emergency Nursing ( 2 0 0 1 )
9, 101–108
101
Clinical decisions using the National Triage Scale
Table 1 National Triage Scale for Australian emergency departments adapted from: Australasian college for emergency medicine 1993b, p10. Triage Code
Treatment Acuity*
National Triage Scale
Category 1
Immediate
Resuscitation
Category 2
Within 10 minutes
Emergency
Category 3
Half Hour
Urgent
Category 4
One Hour
Semi Urgent
Category 5
Two Hours
Non Urgent
* for medical care
All patients who present to an ED are allocated to one of the NTS categories, and it is a nurse who usually makes this decision. Triage decisions are made under conditions of uncertainty with limited or obscure information, minimal time and little margin for error (Edwards 1994; Brillman et al. 1996). Given the difficulties of clinical decision-making at triage, it may be argued triage nurses require greater clinical decision-making skills than nurses performing general emergency nursing duties (Purnell 1991). The triage nurse’s decision to allocate a triage category is based on a clinical assessment that determines illness or injury severity and the need for medical care (Australasian College for Emergency Medicine 1993b; Commonwealth Department of Health and Family Services and the Australasian College for Emergency Medicine 1997). At the present time, the NTS gives no formal consideration to the care provided by emergency nurses whilst the patient is waiting to see a doctor. Given the inherent difficulties of clinical decision making in the triage context, it is well documented by many authors that triage nurses require a certain degree of educational preparation. Descriptions of desirable, if not imperative, attributes of a triage nurse have included ‘highly competent’, ‘qualified’, ‘suitably experienced and trained’, and ‘appropriately skilled’ (Mallett and Woolwich 1990; Williams 1992; Australasian College for Emergency Medicine 1993; Commonwealth Department of Health and Family Services and the Australasian College for Emergency Medicine 1997, p5). Further definition of these qualities was not provided. It is also recognized that the
102
Accident and Emergency Nursing ( 2 0 0 1 )
9, 101–108
experiential and educational prerequisites of Victorian triage nurses are varied (Standen and Dilley 1997). Three months of emergency nursing experience is all that is required by some EDs before being assigned to triage, whilst other EDs restrict triage to nursing staff with a post graduate qualification in emergency nursing. It is important to note that the decisions made by triage nurses are a fundamental factor in the initiation of emergency care. Consequently, the decisions made by triage nurses have the potential to impact on the health outcomes of those patients seeking emergency care (Williams 1992; Commonwealth Department of Health and Family Services and the Australasian College for Emergency Medicine 1997). The uses for the NTS have extended beyond prioritising emergency care. The NTS is currently used to determine elements of ED funding, resource allocation and measurement of ED performance (Doherty 1996; Fitzgerald 1996; Jelinek and Little 1996; Standen and Dilley 1997). This clearly contradicts the National Triage Scale User Manual which states that “Triage requires a clinical decision, which has regard only to the patient’s individual need for care.” (Commonwealth Department of Health and Family Services and the Australasian College for Emergency Medicine 1997). Given that nurses usually make triage decisions, their decisions now have far reaching implications and therefore require critical analysis.
Research aims The aim of this research was to identify any relationships between triage nurses’ decisions on triage category allocation and the type and length
© 2000 Harcourt Publishers Ltd
Clinical decisions using the National Triage Scale
of nursing experience and/or the level of educational preparation of the triage nurse.
Method A correlational design using survey methods was employed in this study. This design was selected as it allowed for the identification of relationships between variables whilst allowing answers to be sought from participants based on the same amount of information.
Setting The study was conducted in two Victorian EDs. One ED was in a large Melbourne metropolitan teaching hospital that provided a wide range of general and specialist adult services as well as being a major health care provider for the poor and disadvantaged. The other ED was in a Victorian rural base hospital that provides general adult and paediatric services and operates an accident retrieval service. This hospital is situated in a large country town that is surrounded by isolated farming properties.
Instrument A self-administered questionnaire was developed by the researcher, examined by nominated experts and piloted to assess clarity and ease of administration. Participants were asked to complete a section on demographic details that asked about each participant’s years of experience in emergency and triage nursing, and level of educational preparation in nursing. Participants were then required to allocate one of the NTS categories to each of ten patient scenarios. Only adult scenarios were addressed and specialty triage scenarios, for example, paediatrics and obstetrics were excluded as these were only seen routinely at one of the two participating EDs. Patient scenarios have, in the past, been considered a valid method of evaluating analytical, problem solving and decision making skills (Holzemer et al. 1981; Carroll and Johnson 1990). The use of patient scenarios enabled the researcher to elicit responses from all participants based on the same information. The variability in the patients’ condition that would occur in real
© 2001 Harcourt Publishers Ltd
life triage episodes was also minimised using this method. The researcher developed the patient scenarios based on real life triage episodes. There were two patient scenarios suited to each of the five NTS categories. A panel of peer nominated expert triage nurses completed the patient scenarios and their responses were compared with the response expected by the researcher. Complete (100%) agreement was reached between the researcher and the expert triage nurses. The agreed triage category for each scenario was referred to as the ‘predicted triage category’. As the patient scenarios used in this study have been published elsewhere, they will not be outlined in detail in this paper (Considine et al. 2000). Participants’ responses were classified into one of three triage decisions. If the participant allocated the same triage category as the predicted triage category, the decision was termed ‘expected triage’ decision. Allocation of a triage category of higher acuity than expected was termed ‘over triage’ decision. This decision indicates that waiting time for the patient would be less than anticipated and therefore would not be detrimental to the patient being triaged. This triage decision may, however, result in the inappropriate use of staff and resources and may be detrimental on a departmental basis. ‘Under triage’ decision was used to describe the allocation of a triage category of a lower acuity than expected, resulting in the patient waiting longer than anticipated and is potentially detrimental to patient safety (Considine et al. 2000; Hollis & Sprivulis 1996).
Sampling Process The population for the study consisted of all the Registered Nurses working in the two EDs selected. As this study focused on triage decision making, only those nurses who were able to be rostered to the triage role were eligible to participate in the study. The total number of registered nurses who were able to be rostered to triage, and therefore met the inclusion criteria for the study was 42. Given the small numbers of triage nurses, random sampling was not employed and convenience sampling was used. All 42 triage nurses were invited to participate in the study and 31 triage nurses responded. This gave a response rate of 74%.
Accident and Emergency Nursing ( 2 0 0 1 )
9, 101–108
103
Clinical decisions using the National Triage Scale
Data Collection
Table 2
Questionnaires were distributed to all nursing staff who were eligible to be rostered to the triage role in the EDs of the participating institutions. Consent was implied by the return of the questionnaire and confidentiality and anonymity were assured.
Participant’s Qualification Categories n
Nil qualification
f (%)
5
16.1
Emergency nursing qualification
11
35.4
Critical care nursing qualification
13
41.9
6
19.3
16
51.6
Midwifery qualification
Data Analysis The computer software SPSS ® Version 6.1 was used for data analysis. Frequency and descriptive statistics were obtained and Spearman’s correlation was used to identify the existence of correlations between triage decisions and the type and length of nursing experience and educational preparation. Chi-square was used to identify significant differences in triage decisions for each of the five NTS categories. Statistical significance was identified by a probability value of less than 5% (P < 0.05).
Results Characteristics of participants When participants’ years of experience in emergency nursing was examined, 29.1% of participants had two or less years experience and 32.2% of participants had ten or more years of emergency nursing experience. Analysis of participants’ years of triage experience revealed that 32.3% of participants had two or less years of triage experience, almost half of the participants (48.4%) reported three to six years triage experience and 19.3% had ten or more years of triage experience. As the educational preparation of participants was varied, five major categories were established to classify participants’ qualifications. These categories were: i) Nil qualification ii) Emergency nursing qualification iii) Critical care nursing qualification iv) Midwifery qualification v) Tertiary qualification Nil qualification was used to describe participants with no post registration and no tertiary qualifications in nursing. Emergency
104
Accident and Emergency Nursing ( 2 0 0 1 )
9, 101–108
Tertiary qualification
31
100
nursing qualification refers to participants with a hospital certificate, graduate certificate or graduate diploma in emergency nursing or who had successfully completed the Trauma Nursing Core Course. Critical care nursing qualification was used to encompass participants with a hospital certificate, graduate certificate or graduate diploma in intensive care, coronary care or critical care nursing. Midwifery qualification was used to describe participants who were Registered Midwives and tertiary qualification was used to describe participants with a Bachelor’s Degree or Graduate Diploma in Nursing. There were no participants who held a Master’s Degree. Approximately half (51.7%) the participants had qualifications in two (2) or more categories.
Triage decisions and experience Each of the 31 participants completed the same ten patient scenarios giving a total of 310 triage episodes. Of these, 58% were ‘expected triage’ decisions, 21% were ‘overtriage’ decisions and 21% were ‘undertriage’ decisions. No correlation was identified between triage decisions and length of experience in either emergency nursing or triage.
Triage decisions and National Triage Scale categories Participant’s triage decisions were then examined in relation to the five NTS categories. The frequency of ‘expected triage’ decisions was significantly higher when the predicted triage category was Category 3 (χ2 = 20.903,
© 2001 Harcourt Publishers Ltd
Clinical decisions using the National Triage Scale
Table 3
Participants’ triage decisions for each National Triage Scale Category Undertriage decision f(%)
Expected triage decision f (%)
Overtriage decision f (%)
NTS Category 1
40.3
59.7
0.0
NTS Category 2
38.7
58.1
3.2*
NTS Category 3
21.0*
79.0*
0.0
NTS Category 4
4.8*
54.8
40.3
NTS Category 5
0.0
40.3
59.7
*P < 0.05
df = 1, P < 0.001). There was a significant reduction in the frequency of ‘overtriage’ decisions when the predicted triage category was Category 2 (χ2 = 54.258, df = 1, P < 0.001). The frequency of ‘undertriage’ decisions decreased significantly when the predicted triage category was Category 3 (χ2 = 20.903, df = 1, P < 0.001) or Category 4 (χ2 = 50.581, df = 1, P < 0.001). It should be pointed out that an ‘undertriage’ decision is not possible for a predicted triage category of Category 5 and an ‘overtriage’ decision is not possible for a predicted triage category of Category 1. This reduces the potential margin for error in triage decision-making within these categories.
Triage decisions and educational preparation of triage nurses No correlation was found between triage decisions and qualifications in the nil, emergency nursing and critical care nursing categories. There was a positive correlation between a midwifery qualification and ‘expected triage’ decisions (rs = 0.3580, P = 0.048) and a negative correlation between a midwifery qualification and ‘undertriage’ decisions (rs = – .4437, P = 0.012). These findings suggest that triage nurses with a midwifery qualification had a higher incidence of ‘expected triage’ decisions and were less inclined to ‘undertriage’. There was a positive correlation between a qualification in the tertiary qualification category and ‘expected triage’ decisions (rs = 0.3859, P = 0.012).
Discussion The findings of this study challenge the assumption that better clinical decision-making is
© 2001 Harcourt Publishers Ltd
Table 4 Correlation of expected triage decisions with participant’s qualifications Expected Triage Decisions rs
P
Nil Qualification
–0.110
.556
Emergency Nursing Qualification
–0.207
.236
Critical Care Qualification
–0.120
.521
Midwifery Qualification
0.358
.048*
Tertiary Qualification
0.386
.032*
* P < 0.05 Table 5 Correlation of undertriage triage decisions with participant’s qualifications Undertriage Decisions rs
P
Nil Qualification
0.271
.140
Emergency Nursing Qualification
0.124
.508
Critical Care Qualification
–0.0120
.521
Midwifery Qualification
–0.435
.015*
Tertiary Qualification
–0.207
.264
* P < 0.05
an unequivocal outcome of advanced experience and specialty educational preparation. The absence of identifiable correlations between
Accident and Emergency Nursing ( 2 0 0 1 )
9, 101–108
105
Clinical decisions using the National Triage Scale
experience in emergency nursing and triage nursing disputes the notion that experienced nurses make more accurate decisions than their less experienced counterparts. Jelinek and Little (1996) concluded that triage responses were unaffected by triage nurse experience. Likewise, a study of clinical decision-making by critical care nurses by Sims and Fought (1989) also reported no relationship between accuracy of decisionmaking and length of experience. Questions should be raised as to the value of chronological measurement of experience and broader criteria such as the quality of nursing experience should be examined. The reason for significant differences in triage decisions related to specific triage categories is unclear. Almost three-quarters (79.0%) of participants arrived at an ‘expected triage’ decision when the predicted triage category was Category 3. There is no available literature to assist in expiation of this finding. It may be postulated that Category 3 is perceived as a ‘safe choice’ for triage nurses as it is the ‘middle’ category in the NTS. Selection of Category 3 neither implies that the patient’s presenting problem is of an emergent or semi-urgent nature. This may be a preferable option to triage nurses who are unsure of the severity of a patient’s presenting problem. Only 3.2% of participants arrived at an ‘overtriage’ decision when the predicted triage category was Category 2. Again, there is no available literature to explain this finding. As ‘overtriage’ decisions are associated with inappropriate use of ED resources and personnel, it is encouraging that the only significant finding for overtriage decisions was the decrease in overtriage decisions for Category 2. It may be postulated that this reduction in overtriage decisions is related to perceived consequence. An ‘overtriage’ decision for a Category 2 patient would result in the allocation of the NTS Category 1. As medical care is required immediately for Category 1 patients, the decision to allocate Category 1 often results in many staff becoming aware of the patients’ presence in the ED. As the ability of an ED to see Category 1 patients immediately is important when assessing ED performance and allocating ED funding, allocation of Category 1 may also result in increased scrutiny at both departmental and organizational levels.
106
Accident and Emergency Nursing ( 2 0 0 1 )
9, 101–108
The decrease in ‘undertriage’ decisions when the predicted triage category was Category 3 or 4 was also an unexpected finding. Given that ‘undertriage’ decisions have the potential to be detrimental to patient safety, it is reassuring that the significant findings in terms of ‘undertriage’ decisions all relate to decreases in the frequency of ‘undertriage’ decisions. One possible explanation is that triage nurses who are unsure of the actual or potential severity of a patients’ presenting problem may ‘err on the side of caution’ and make an overtriage decision. The lack of identifiable correlations between triage decisions and further educational preparation raises questions as to the adequacy of the educational preparation of triage nurses. Participants with no post registration and no tertiary qualifications in nursing have an understandable explanation for deficits in clinical decision making. It is of concern, however, that triage nurses with specialist educational preparation in emergency nursing and critical care nursing also appear to have deficiencies in these areas. This result bears similarity to a study by Sims and Fought (1989) who reported that certification in critical care nursing made no significant difference in decision making when compared with nurses who were not certified. The relationship between triage decisions and a midwifery qualification was a surprising finding and reason for this is unclear. One possible explanation may be the anecdotal similarities shared by nursing practice and clinical decision-making in the areas of triage and midwifery, in particular, delivery suite. In both practice settings there is an absence of routine and patients may present to either area at any time and with varying degrees of acuity. Both areas of practice involve autonomous decision making by nurses based on patient assessment and the setting of priorities related to patient acuity. It may be postulated that the assessment and clinical decision making skills of midwives may be applied to any practice setting, however more research in this area is warranted. The positive correlation between triage decisions and tertiary qualifications may be explained by the higher degree of educational preparation in this group of participants. This notion is supported by the literature. It is stated by del Bueno (1983) that educational preparation is positively correlated with
© 2001 Harcourt Publishers Ltd
Clinical decisions using the National Triage Scale
acceptable decision-making and that nurses with higher educational preparation had a higher incidence of correct decisions. Pardue (1987) and Holl (1994) also state that critical thinking ability and decision-making were better in nurses who had a higher standard of education. If formal education within a nursing specialty does not show any relationship with improved decision making in that specialty, questions should be raised as to how nurses should be educated for specialty practice, specifically the quality of clinical and formal educational preparation of triage nurses. It may be hypothesized that education in areas such as critical thinking, clinical decision making and problem solving are a more adequate preparation for clinical practice than simply imparting specific knowledge about specialist nursing, for example emergency nursing or triage.
Limitations This study was limited by its sample size of 31 participants. This limitation occurred because of the relatively small number of both emergency nurses and triage nurses. Scenario based assessment of triage decisions may be seen by some as a limitation, although the validity of the use of scenarios has been acknowledged in the literature (Holzemer et al 1981; Carroll and Johnson 1990). With regard to the statistically significant findings concerning participants with a qualification in the midwifery category, it should be noted that there were only six (6) participants (19.3%) in this group. To overcome this limitation, replication of the study with a larger sample size and a larger proportion of participants with a midwifery qualification is required.
Recommendations There is a need for a review of the educational preparation of triage nurses with specific content, length and scope of nursing specialty education. Clinical decision-making at both undergraduate and postgraduate levels should be examined, as should the differences in the decision-making skills of beginning and specialist practitioners. There is a need for further research into the effect of educational preparation on decisions made by triage nurses and, more specifically, their application of the National Triage scale.
© 2001 Harcourt Publishers Ltd
Critique of the clinical decision making abilities of those responsible for educating triage nurses is of paramount importance and questions regarding the clinical credibility and recency of practice of those teaching nursing specialty courses should be raised. There is a need for clear definition about the difference in expected performance of generalist and specialist nurses, and in more specific terms, the decision-making skills of these groups. Consideration should be given to the clear definition of specialty practice education in emergency nursing as opposed to generic critical care courses. In nursing, the importance of chronological experience alone as valuable criteria for advancement should be questioned and other ways of assessing the quality and value of clinical experience need to be explored. The use of clinical fellowship and mentors in clinical practice for developing specialty practitioners needs to be explored as a means of ensuring that clinical experience has value in improving practice. References Australasian College for Emergency Medicine 1993a Policy document: triage Australasian College for Emergency Medicine 1993b Position paper: a National Triage Scale for Australian Emergency Departments Brillman J, Doezema D, Tandberg D, Sklar D, Davis K, Simms S, Skipper B 1996 Triage: Limitations in predicting the need for emergency care and hospital admissions. Annals of Emergency Medicine 27(4): 493–500 Carroll J, Johnson E 1990 Decision research: a field guide. Sage Publications: Newbury Park Commonwealth Department of Health and Family Services and the Australasian College for Emergency Medicine 1997 The Australian National Triage Scale: a user manual 1997 Considine J, Ung L, Thomas S 2000 Triage nurses’ decisions using the National Triage Scale for Australian emergency departments. Accident & Emergency Nursing 8 (4): 201–209 del Bueno D 1983 Doing the right thing: nurses’ ability to make clinical decisions. Nurse Educator 8 (3): 7–11 Doherty S 1996 Application of the National Triage Scale is not uniform Australian Emergency Nursing Journal 1 (1): 26 Edwards B 1994 Telephone triage: how experienced nurses reach decisions. Journal of Advanced Nursing 19: 717–724 Fitzgerald G 1996 The National Triage Scale. Emergency Medicine 8: 205–206 George S, Read S, Williams B 1993 Nurse triage. British Medical Journal 306 (Jan): 208
Accident and Emergency Nursing ( 2 0 0 1 )
9, 101–108
107
Clinical decisions using the National Triage Scale
Geraci E, and Geraci T, 1994 An observational study of the emergency triage nursing role in a managed care facility. Journal of Emergency Nursing 20 (3): 194–198 Holl R 1994 Characteristics of the registered nurse and professional beliefs and decision making. Critical Care Quarterly 17 (3): 60–66 Hollis P, Sprivulis P 1996 Reliability of the National Triage Scale with changes in emergency department activity levels. Emergency Medicine 8: 231–234 Holzemer W, Schleutermann J, Farrand L, Miller G 1981 A validation study: simulations as a measure of nurse practitioners’ problem-solving skills. Nursing Research 30 (3): 139–144 Jelinek G, Little M 1996 Inter-rater reliability of the National Triage Scale over 11,500 simulated occasions of triage. Emergency Medicine 8: 205–206 Mallett J, Woolwich C 1990 Triage in accident and emergency departments. Journal of Advanced Nursing 15: 1443–1451 McDonald L, Butterworth T, Yates D 1995 Triage: a literature review 1985–1993. Accident and Emergency Nursing 3: 201–207
108
Accident and Emergency Nursing ( 2 0 0 1 )
9, 101–108
Pardue S 1987 Decision-making skills and critical thinking ability among associate degree, diploma, baccalaureate and masters prepared nurses. Journal of Nursing Education 26 (9): 354–361 Purnell L, 1991 A Survey of emergency department triage in 185 hospitals. Physical facilities, fast track systems, patient classification systems, waiting times and qualifications, training and skills of triage personnel. Journal of Emergency Nursing 17 (6): 402–407 Rowe J 1992 Triage assessment tool. Journal of Emergency Nursing 18 (6): 540–544 Sims K, Fought S 1989 Clinical decision making in critical care. Critical Care Nursing Quarterly 12 (3): 79–84 Standen P, Dilley S 1997 A review of triage nursing practice and experience in Victorian public hospitals. Emergency Medicine 9: 301–305 Williams G. 1992 Sorting out triage. Nursing Times 88 (30): 34–36 Zwicke D, Bozien W, Wagner E 1982. Triage nurse decisions: A prospective study. Journal of Emergency Nursing 8 (3): 132–138
© 2001 Harcourt Publishers Ltd