One of the key roles of ambulance call receivers is to obtain accurate information concerning the location of the patient and their complaint. The automatic tracing of calls to their source can accurately determine the location of most patients. Obtaining information on the patient's condition, however, depends on effective communication between the call maker and the call receiver. Ambulance services in the United Kingdom now use priority dispatch systems to categorise calls according to the level of urgency for medical treatment and to provide first aid advice (for example, cardiopulmonary resuscitation).1,2 The safety and effectiveness of priority dispatch has been assessed3; however, whether call receivers are able to procure accurate and reliable information has not been established. This study aimed to determine the nature and extent of communication problems encountered during 999 calls.
Participants, methods, and results
The study population comprised 999 calls received by an urban (West Midlands Ambulance Service) and a rural (Derbyshire Ambulance Service) ambulance service. A stratified systematic sample of the calls received over one week (early December 1998) was selected (n=1830). During this time, the West Midlands Ambulance Service received approximately 830 calls per day and the Derbyshire Ambulance Service received approximately 240 calls per day. An assessor listened to tape recorded calls to determine the caller's identity (for example, husband), location in relation to the patient, type of telephone, and communication difficulties. Cases were excluded if recording quality was very poor.
The method was piloted.4 A conversation was noted as having communication difficulties if the call receiver had to repeat the question, the caller gave an inappropriate response more than once, or misunderstanding occurred between caller and call receiver.
We used univariate analysis to assess the characteristics of callers and compare the two ambulance services, and stepwise discriminant analysis to establish the factors associated with communication problems.
Of 1830 calls, 482 (26.3%) were associated with a communication problem that may delay ambulance dispatch or prevent delivery of first aid advice (table). Fewer problems were observed in Derbyshire than in the West Midlands.
The most common reason for communication problems occurring was the emotional state of the caller (161/482, 33.4%). Almost 10% (45/482) of problems were related to the ambulance service call receiver missing information or failing to be understood by the caller. Communication problems were associated more with calls made from payphones or mobile phones (110/221, 49.8%) than with those made from land lines (372/1609, 23.1%) (χ2=71.2, df=1; P<0.001). The type of telephone (land line, payphone, mobile phone) and communication problem were similar in Derbyshire and the West Midlands. Communication problems were less likely if the call was made by a health professional (for example, general practitioner, carer, nurse) rather than by someone else (21/220, 9.5% v 461/1610, 28.6%; χ2=26.4; P<0.001).
Comment
More than a quarter of emergency ambulance calls in this study had communication problems. Calls from mobile phones and payphones generated a higher rate of communication problems than those from land lines. Mobile phones, which are used increasingly,5 may help to reduce the time taken to notify the emergency services butthe advantages of this must be weighed against the high rate of communication problems.
The occurrence of communication problems related to the emotional state of the caller highlights the need to train call receivers in dealing with people in emotional states. Use of medical/technical terms, some of which can cause considerable confusion (for example, “unconscious”), as well as talking too quickly and without clarity, have been identified as areas in which training of call receivers is needed.
Use of a standard land line, appropriate training of public service personnel, such as police and fire services, and further public education about the information required when making 999 calls may reduce the extent of the communication problem.
Table.
Communication difficulties associated with 999 calls to ambulance services. Values are numbers (percentages)
| Calls to ambulance service
|
||||
|---|---|---|---|---|
| Derbyshire (n=600) | West Midlands (n=1230) | Total (n=1830) | Significance (comparison between sites) | |
| Problem associated with call | ||||
| Caller: | ||||
| Abusive | 0 | 2 (0.2) | 2 (0.1) | |
| Breathless or ill, causing difficulty with explanations | 6 (1.0) | 24 (2.0) | 30 (1.6) | |
| Slow, vague, or deaf | 5 (0.8) | 21 (1.7) | 26 (1.4) | |
| Child | 1 (0.2) | 4 (0.3) | 5 (0.3) | |
| Emotional, excitable, upset, or speaking too quickly | 58 (9.7) | 103 (8.4) | 161 (8.8) | |
| Not understanding questions asked by call receiver | 0 | 4 (0.3) | 4 (0.2) | |
| Not wanting to answer, or giving misleading information | 0 | 5 (0.4) | 5 (0.3) | |
| Strong accent or dialect | 4 (0.7) | 18 (1.5) | 22 (1.2) | |
| Unclear or slurred speech, difficult to understand | 11 (1.8) | 17 (1.4) | 28 (1.5) | |
| Use of non-English language or poor command of English | 2 (0.3) | 9 (0.7) | 11 (0.6) | |
| Call receiver: | ||||
| Speaking too fast or using phrases not understood by caller | 2 (0.3) | 8 (0.7) | 10 (0.5) | |
| Information from caller not absorbed | 10 (1.7) | 25 (2.0) | 35 (1.9) | |
| Difficulties with address or location* | 15 (2.5) | 74 (6.0) | 89 (4.9) | |
| Technical problems | 13 (2.2) | 25 (2.0) | 38 (2.1) | |
| Other | 5 (0.8) | 11 (0.9) | 16 (0.9) | |
| Number of communication problems | 132 (22) | 350 (28.5) | 482 (26.3) | χ2=8.66, df=1; P=0.003 |
| Nature of call | ||||
| Calls from patients, relatives, and friends | 292 (48.7) | 669 (54.4) | 961 (52.5) | χ2=5.3, df=1; P=0.012 |
| Calls from health workers, carers, and police | 76 (12.7) | 144 (11.7) | 220 (12.0) | χ2=0.35, df=1; P=0.3 |
| Calls from police and fire personnel | 26 (4.3) | 83 (6.8) | 109 (6.0) | χ2=4.2, df=1, P=0.024 |
| Caller in close proximity to patient† | 158 (26.3) | 373 (30.3) | 531 (29.0) | χ2=14.26, df=1; P<0.001 |
Caller did not know or could not spell the street name, or could only provide local landmarks or road number.
Caller and patient are within hearing distance of each other.
Acknowledgments
We thank all the control room staff at West Midlands Ambulance Service NHS Trust and Derbyshire Ambulance Service NHS Trust for their help in this study; Tracey Cooper, Steve Elliker, Claire Caswell, and the control room supervisors at both ambulance services for their assistance; and Mrs Teresa Allan for her statistical advice during the design and analysis phases of the project.
Footnotes
Funding: Laerdal Foundation for Acute Medicine.
Competing interests: None declared.
References
- 1.Marsden AK. Getting the right ambulance to the right patient at the right time. Accid Emerg Nurs. 1995;3:177–183. doi: 10.1016/0965-2302(95)90002-0. [DOI] [PubMed] [Google Scholar]
- 2.McNaughton GW, Wyatt JP. Telephone guided CPR—it's good to talk! Pre-hosp Immed Care. 1997;1:71–72. [Google Scholar]
- 3.Nicholl J, Coleman P, Parry G, Turner J, Dixon S. Emergency priority dispatch systems—a new era in the provision of ambulance services in the UK. Pre-hosp Immed Care. 1996;3:71–75. [Google Scholar]
- 4.Cooke M, Wilson S. Are 999 callers in a position to give triage information and receive first aid advice? Pre-hosp Immed Care. 1998;2:193–196. [Google Scholar]
- 5.Mobile Phones and Health Independent Expert Group on Mobile Phones—National Radiological Protection Board April 2000. www.iegmp.org.uk/IEGMPtxt.htm (accessed 18 August 2001)