Article Text

Download PDFPDF

Improving communication in the emergency department
  1. E Redfern1,
  2. R Brown2,
  3. C A Vincent3
  1. 1
    Emergency Department, Bristol Royal Infirmary, Bristol, UK
  2. 2
    Imperial College Healthcare NHS Trust, London, UK
  3. 3
    Clinical Research Safety Unit, London, UK
  1. Correspondence to Dr E Redfern, Emergency Medicine, Bristol Royal Infirmary, Marlborough Street, Bristol BS2 8HW, UK; emredfern{at}hotmail.com

Abstract

Background: A previous study examined the communication process within the emergency department (ED) and identified a complex process with many opportunities for breakdown and error. In this paper the first two interventions in a series of studies to improve this highly vulnerable communication process are described.

Aim: To improve the reliability of two steps of the communication process identified as having a high probability of failure: (1) transfer of information between the ambulance crew and the emergency staff; and (2) preparation of written documentation following patient assessment.

Methods: Quantitative assessments of the reliability of communication were carried out to establish the extent of problems highlighted during the failures modes and effects analysis (FMEA) previously described. Improvements to the process were then introduced, and the process re-examined to assess the impact of the changes and reduction of the likelihood and severity of the failure mode.

Results: The studies demonstrated very high levels of communication failure, particularly in transfer of written information from the ambulance crew. Countermeasures were introduced which resulted in a substantial reduction in missing and incorrect information. In addition, there was a threefold improvement in the number of correct clinical documents used by doctors in the resuscitation room.

Conclusion: Observational study and audit revealed the extent of process failures identified in the initial FMEA process. With the introduction of simple changes to the communication system, a marked improvement in the availability and quality of pertinent clinical information was achieved with considerable implications for the timeliness and quality of care provided to patients.

View Full Text

Statistics from Altmetric.com

The communication process in the emergency department (ED) is complex, and a number of studies have described problems and breakdowns in communication with significant implications for patient care.1 2 3 In an earlier paper we reported a systematic mapping and analysis of the communication process using Failure Modes and Effects Analysis (FMEA) which identified a considerable number of vulnerabilities and potential for failure.4 The communication process proved to be unduly complex and there were many opportunities for error which may impact on patient care.

Based on the results from the FMEA, we are now engaged in a series of studies aimed at improving the communication at points in the pathway where significant vulnerabilities were identified. In this paper we report findings from the first two interventions, which form part of a larger project on quality improvement in the ED. We focused on the Patient Report Form (PRF) provided by ambulance crews and the standard printed document produced for the recording of clinical information by the doctor. Both of these steps had been identified by staff as having significant problems with potential impact on patient care. Our findings were in accord with those of the few previous studies in this area. Previous research on handover from ambulance providers to ED staff has uncovered gaps in communication,2 with one study showing that only 56% of verbal information is accurately retained by the ED staff.5 The perceived quality of verbal handover has been shown to be variable between different ambulance crews.6

Study 1: Improving transfer of information from the Patient Report Form (PRF)

The PRF is the written record detailing the events from initial call to ambulance control, arrival of the ambulance crew at the scene and transfer to hospital. It is completed by ambulance crews and records basic details such as unique crew identification numbers, exact times relating to movements of the ambulance both at the scene and arrival at hospital, as well as patient demographic details. Clinical data recorded include the reason for the call out, medication and allergies, as well as the patient’s clinical condition, vital signs and drugs given. Comments on the condition of the patient may also be included, particularly any signs of deterioration or improvement.

The PRF ideally serves to support the efficient transfer of care to the next healthcare professional involved with the patient. The details recorded are particularly crucial if the patient is not able to communicate effectively or where the only information available comes from bystanders who witnessed the incident leading to the ambulance call. Clinical staff working in the ED consider the PRF information as absolutely vital in these circumstances and as highly informative at other times; however, accessing this vital information can be problematic.

The PRF used in our region (London) is a carbon copy triplicate. The top two copies are retained by the ambulance trust for legal and audit purposes. The bottom copy is handed to personnel in the ED. After arriving in the ED, the ambulance crew give a brief verbal handover to the nurse receiving the patient. The crew then pass the demographic details of the patient to an ED receptionist who registers the patient onto the computer system. If the patient has attended the ED previously and the correct demographic details are entered, the records will be matched and linked up to previous episodes. A set of ED notes is then generated for the patient and the bottom copy of the PRF is scanned and attached electronically to the patient’s current episode. This enables the attending clinician to view the PRF along with copies of written documentation from previous attendances.

In practice, however, both clinical reports and the formal FMEA process identified a number of problems with this apparently straightforward process. The PRF might be missing or illegible in varying degrees, completely nullifying the ambulance crew’s efforts to communicate effectively with emergency staff (table 1). We therefore designed and carried out a quality improvement study with the following aims:

Table 1

Failure modes in the transfer of information from ambulance crews

  • To establish the usefulness of the information on the PRF to clinical care.

  • To quantify the number of missing or illegible scanned images.

  • To develop and evaluate an intervention to resolve any problems identified.

Method

Design and setting

The study was set in a busy central London ED. We first assessed the quality and clinical value of written information provided by ambulance crews. We then assessed measured communication efficiency before and after an intervention to improve availability of information from the PRF.

Procedure and measures

The first stage was to assess the information available from the PRF. In this phase all paper copies of PRFs submitted by the ambulance crews during a single week (n = 182) were collected and analysed. The standard of completion of the information and usefulness to clinical care was analysed by the principal author (ER). The presence or absence of the following patient information was assessed:

  • Usefulness of information regarding the presenting complaint.

  • Presence of abnormal and normal observations.

  • Medication given by ambulance crews.

  • Information about patient’s normal medication.

  • Information about allergies.

Following this, the reliability of the communication process was examined and quantified. A total of 250 sets of notes of patients brought in by ambulance were retrospectively audited to examine: (1) the presence or absence of the PRF in the notes and (2) the degree of legibility of the scanned copy. Following the intervention, a further 250 sets of notes were examined for the same factors.

Intervention

We approached members of the team seeking ideas about how to improve the legibility of the scanned document. We then discussed the feasibility of implementing the change to the system with our reception staff. A simple solution was to photocopy the top sheet of the PRF at the time of the crew arriving in reception and then to scan the photocopied document: this resulted in a totally legible electronic copy available to the departmental staff during the patient attendance. In addition, the receptionists undertook to actively ask the ambulance staff for the PRF in order to reduce the number of PRFs that were not transferred. The reception staff agreed to a trial of this system for 1 month. Emails were sent out to every receptionist, the change in practice was discussed at the monthly staff meeting, posters were made and the changes were discussed with the ambulance service.

Results

Clinical value of patient report form (PRF)

The initial assessment of the PRF sheets showed that they contained substantial amounts of important clinical information that was often essential to proper care of the patient. Of the 182 PRF sheets examined: 100% (182/182) contained useful details regarding the presenting complaint; 55% (100/182) of the documents detailed abnormal observations; 15% (28 of 182) recorded medication given by the ambulance crew; 66% (122/182) of the PRFs documented the patient’s own medication; and 88% (160/182) noted allergies or the absence of allergies.

Intervention to improve transfer of information

The simple interventions produced a dramatic improvement in the information available to ED staff. Before the intervention 82% of the scanned PRFs were completely illegible and 18% partially legible; after the intervention, 100% were wholly legible (table 2). The numbers of PRFs that were scanned to the database improved from 70% to 80%. The two-sample proportion z test was used to analyse the results, which were both clinically and statistically significant.

Table 2

Legibility of Patient Report Form before and after intervention

Study 2: Generation of correct note format for doctors

The second study followed a similar pattern, examining another issue highlighted by the FMEA process concerning documentation used by doctors to record clinical information for different types of patient. In our ED, proforma style structured notes are used to record, structure and guide the collection of clinical information. The department uses five different types of proforma, two generic that are used in either majors area (trolley assessment) or in minors area (ambulatory care). These two are printed out automatically at registration according to the area of the department to which the patient is directed. The remaining three are specifically designed for critically ill patients and include separate ones for cardiac arrest, serious illness and major trauma. These are printed by the receptionists on arrival of the patient according to their predominant complaint. The findings of the FMEA suggested that there was a high probability of the wrong document being requested or the wrong document being generated, leading to a delay and inappropriate direction of patients within the department. The immediate effects of this on patient care are that: (1) it is difficult to record vital information; (2) much time is wasted reading through notes to identify salient information; and (3) a proper management plan may not be recorded. We therefore carried out a further quality improvement study to (1) establish and quantify whether the correct documents were being generated for each patient and whether each set of notes contained a management plan; and (2) evaluate the impact of an intervention to improve this aspect of documentation.

Method

Design and setting

The study was set in a busy central London ED. We measured the efficiency of the communication using the proforma notes before and after an intervention designed to ensure the correct document was generated.

Procedure and measures

A retrospective audit of 160 ED notes was carried out to assess whether the correct template was used for the correct presenting complaint and whether there was a clear management plan recorded in the notes, regardless of the proforma used. Patients were identified by searching within the term “area  =  majors/minors” or “presenting complaint  =  trauma/cardiac arrest” on SYMPHONY database. Fifty sets of majors and minors notes were studied and 20 each of the major trauma and cardiac arrest notes in relevant patients. Following the intervention, a further set of 160 notes were audited in the same way.

Intervention to improve information transfer and documentation

Staff members were invited to make suggestions on how to improve the generation and use of the correct documents. Solutions generally focused on raising awareness of the different documents among the reception and medical staff at teaching and staff meetings. The importance of documenting a management plan was discussed with the medical staff after presentation of the initial findings of the audit. We also made another simple but crucial change by placing an electronic copy of each one of the cardiac arrest and major trauma templates on the computers in the resuscitation room. This meant that the correct document could be printed even if an incorrect document had initially been generated.

Results

The initial data collection confirmed the problems identified in the FMEA; use of the correct template ranged from 86% for majors to only 10% for patients with cardiac arrest. Only 60% of patients with serious conditions had a documented management plan. Again, the two-sample proportion z test was used to analyse the results, and the intervention produced both clinically and statistically significant improvements (tables 3 and 4).

Table 3

Use of correct template before and after the intervention

Table 4

Documentation of management plan before and after the intervention

Discussion

Both the FMEA and the studies to quantify the extent of the problems identified serious weaknesses and inefficiencies in the system of communication with the ED. We fully expect that, when further quantitative studies are carried out on different points in the process, we will see further evidence of routine failure and further opportunities for improvement.

However, simple interventions produced significant results. Clinical staff now have improved timely access to relevant information from ambulance crews, the correct template is more frequently being used and documentation of management plans has increased. The process will be audited to judge the longevity of the change once the intervention has become established, but it is possible that there will be a reduction in the benefit seen unless there is some continued feedback and monitoring of the system. Involving staff in all aspects of diagnosis and management of the problems with communication will hopefully ensure that the team understands why changes to the system need to be made.

Solutions to problems need to be time- and resource-neutral or to require little additional effort. Hospital and ambulance staff have seen the benefit of accessibility and judge that the small amount of time taken to copy the PRF is worth the effort. The solution offered for the legibility and retention of the document gave an unexpected benefit of having the PRF available at the point of contact with the patient rather than as a loose copy that is somewhere in the department.

Looking further ahead, we hope that information technology will bring further improvements to communication systems. The ambulance service is working towards integration of a prehospital electronic patient record with the hospital record and with the National Care record service, but this technology will not be available until 2013. Electronic patient record systems offer solutions for printing documents according to logic modules or rules in the system. However, the system still relies on humans entering data to drive the rules, and that is prone to error. The electronic patient record will require structured data entry, and early adoption of this approach—albeit on paper—allows evaluation of the structure and best way to present data.

The findings of this study have several implications for future practice. First, EDs may be able to use this methodology to examine prospectively their own areas of potential systems failure and improve the design of their care process; the complex issue of communication can be simplified using a formal technique. Second, involving staff in the study drew on their knowledge, experience and creativity, generated many good suggestions and ensured acceptance of the eventual implementation. Third, it shows that making simple changes, when introduced and evaluated systematically, can lead to large improvements in the care provided to patients.

Conclusion

We conclude these two papers with a final reflection. What took us so long? Clinical staff had been using a highly inefficient and frustrating system producing illegible transfer forms for years, yet no action had been taken. We believe that this is a common experience throughout health care, both in the UK and in other countries. Many reasons might be advanced for this, with varying degrees of criticism of healthcare staff, but three particular issues emerge. The first is the extraordinary tolerance of healthcare staff for inefficient and failure-prone systems and the degree to which they cope with frustrating working conditions rather than attempting to change them. Secondly, it took a formal study to bring home the full extent of the problems; even though we are very familiar with the environment, we did not fully appreciate the complexity of the communication process or the extent of the failures. Finally, dedicated time is needed even to bring about simple, if far reaching, changes. When a clinical member of staff was allowed sufficient dedicated time to assess and improve the system, substantial gains were made in a relatively short time. We believe that there is a more general lesson here. We will, in the end, provide better care for our patients if we begin to allocate some proportion of clinical time to system improvement rather than direct patient care. This has already happened to some extent with the requirement to carry out audit, but we need to go further in both allocating more time and being more ambitious, not simply auditing but actually effecting improvement.

Acknowledgments

The authors thank Dr Robert Redfern for help with the statistical analysis in this paper.

REFERENCES

View Abstract

Footnotes

  • Funding The Clinical Safety Research Unit is affiliated with the Centre for Patient Safety and Service Quality at Imperial College Healthcare NHS Trust which is funded by the National Institute of Health Research. This research described here was supported by the National Institute of Health Research.

  • Competing interests None.

  • Provenance and Peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • Primary survey
    Steve Goodacre