Difficult airway communication between anaesthetists and general practitioners

Introduction

A patient is classed as having a ‘difficult airway’ when intubation for mechanical ventilation is difficult, or when it is hard to achieve oxygenation using basic (‘bag, valve, mask’) airway skills. Mismanagement of the airway can be disastrous. The recently published National Audit Project of the Royal College of Anaesthetists and the Difficult Airway Society (NAP4) reported that 46 patients had either died or suffered serious brain damage in the UK in the year ending August 2009, purely because of complications of airway management. ¹ If forewarned, alternative airway management plans can be put in place, ensuring that senior help and additional equipment are available at the induction of anaesthesia. This forewarning requires effective communication between anaesthetists and other healthcare professionals, in particular general practitioners (GPs).

The General Medical Council describes communication as a cornerstone of good medical practice. ² Within anaesthesia, textbooks and national guidelines place great emphasis on the communication of the difficult airway.^3–7 However, as multisite care has become increasingly common, the ability of anaesthetists to communicate directly with one another, or to be able to see an old anaesthetic record has diminished. Information provided by a general practice care summary can be invaluable to anaesthetists who do not have immediate access to a patient's complete care record.

In general practice, records are summarised using a system of diagnostic coding, Read Codes.^8,9 These translate clinical events and conditions into entries in their electronic medical records. Codes are interchangeable between the various GP computer systems throughout the UK. When a hospital referral is made, the system prints out the patient's record, listing only the coded clinical information, according to its specified priority (the Emergency Care Summary). These systems already code for difficult intubation and the Read Code is SP2y3. ⁹ Despite this, we had never seen this code (or diagnosis) used by an anaesthetist or appear in an Emergency Care Summary.

We wanted to explore communication of the difficult airway between anaesthetists and GPs. We conducted our survey in two stages. Initially, we aimed to establish how anaesthetists had been alerted to difficult airway information and how they communicated it with others. (For the questions used, please see Appendix 1.) We then put a similar set of questions to GPs (Appendix 2). Finally, we asked clinicians to identify any simple measures that could be instituted to improve this chain of communication.

Methods

We created two online questionnaires (www.zoomerang.com) (Appendices 1 and 2). The questionnaire for anaesthetists was deployed throughout the UK in 2009 via the Royal College of Anaesthetists' college tutors. Tutors were asked to circulate them on to consultant colleagues. Non-responding departments were followed up by telephone. Our second questionnaire sought to establish the views of Scottish GPs (Appendix 2). The survey link was disseminated via the National General Medical Services programme to Clinical Leads in the 14 Scottish Territorial Boards and onward to individual GPs. Prize draws were offered in both surveys to help maximise response. The results were analysed using Microsoft Excel 2007 (Microsoft, Seattle, WA, USA). Not all respondents completed all questions in either survey, but all responses have been included in the analysis, accounting for the variable denominators.

Results

Anaesthetist survey

We received 631 replies from anaesthetists in 124 different hospitals. Two hundred and eighty-three of the 631 (45%) aimed to write letters after encountering a difficult airway but only 125 (20%) would do so in every case. Of the remaining 348 anaesthetists who did not write letters, 25 (7%) stated that they would ‘always mean to’. Two hundred and sixteen (62%) respondents would record the information elsewhere: anaesthetic chart (202, 94%), patient notes (142, 66%) and departmental records (seven, 3%). One hundred (16%) would tell the patient and seven (1%) would take no action at all. Barriers to letter writing were identified as lack of time and the absence of a secretarial service.

The clinical scenarios that would trigger a difficult airway letter varied between anaesthetists and are shown in Figure 1. (Note that the difficult Bag-Valve-Mask column includes 4 (1.3%) anaesthetists who would only write if BVM was difficult, irrespective of the grade of laryngoscopy.) Anaesthetists would send difficult airway letters to the patient in 241 (89%) of cases, 225 (83%) to the GP and 34 (12%) to the surgeon. Two hundred and thirty-eight (88%) would write in the case notes, 45 (17%) in departmental files and 21 (7%) in the hospital computer system. Some individuals also kept personal records. The content of the letters are shown as Table 1.

OPEN IN VIEWER

Table 1.

Information included by anaesthetists in their letters to general practitioners. Values are number (proportion).

Number of anaesthetists (n = 272)
Grade of laryngoscopy	242 (89%)
Method used to secure airway	178 (65%)
Experience of anaesthetist	167 (61%)
Ease of mask ventilation	19 (7%)
Narrative description of events	5 (2%)

In the past, 578 (92%) of anaesthetists had been alerted to a difficult airway by an old anaesthetic chart and 525 (83%) by the patient. Two hundred and fifty-eight (41%) had been notified by a letter from another anaesthetist but only 35 (6%) by a letter from the GP.

Twenty anaesthetists (3%) were already aware of a Read Code for difficult tracheal intubation, though 454 (72%) thought that it would prove useful. Having been made aware of the SP2y3 coding, 256 (62%) of 414 then stated they would ask the surgeons to include it in their discharge summaries. One hundred and sixty-eight (41%) hoped to include it in their own letters and 123 (30%) felt that they would now write letters more frequently as a result.

One hundred and fifty-two anaesthetists commented on improving communication with GPs: 71 (47%) suggested standardised letters, possibly including the Read Code, 25 (16%) proposed a national database, 21 (14%) putting stickers on the patient notes and 12 (8%) giving out ‘Airway Alert’ cards or bracelets.

GP survey

We received 217 replies from 11 of the 14 Scottish GP Territorial Boards. Three Territorial Boards had a moratorium on circulating surveys and declined to take part despite lengthy discussions. One hundred and ninety-three (76%) of respondents were GP Principals, of which 142 (65%) had been practicing for longer than 15 years. Twenty-three (11%) of the GPs had had prior anaesthetic experience.

Eighty (37%) of the GPs had previously received difficult airway correspondence (Table 2) but only half of these (n = 40) had then included it in a subsequent referral. Of these, seven (18%) forwarded it to the anaesthetist, 24 (60%) to the surgeon, 10 (25%) informed the patient, whilst only 20 (50%) included it in the emergency care summary.

OPEN IN VIEWER

Table 2.

Sources of difficult airway information for general practitioners. Values are number (proportion).

Number of GPs (n = 80)
Anaesthetist	43 (54%)
Surgeon	24 (30%)
Ward doctor	15 (19%)
Patient	9 (11%)

This was despite 212 (98%) of the GPs feeling that it was important to make patients aware of their difficult airways and to store that information in the GP records. Of the five (2%) GPs who disagreed with this statement, two felt that it would frighten the patient and three that the patient would forget or confuse the information.

Twenty-one (10%) of the GPs were aware of the Read Code for difficult tracheal intubation and six had seen it used in the past.

Suggestions to improve communication were made by 194 (89%) of the GP respondents. One hundred and ninety (98%) stated the importance of diagnostic coding. Some GPs pointed out that coding is often done by clerical staff or medical students. These staff members may not appreciate the need to code for a difficult tracheal intubation if it appeared as part of a surgical discharge letter, or as a separate letter from an anaesthetist to the practice. Twenty-eight (14%) of GPs also indicated that labelling the diagnosis as ‘High Priority’ or ‘Priority 1’ would make the information appear prominently in all subsequent printed emergency care summaries, on a par with drug allergy information.

Discussion

Less than half of the anaesthetists responding to our survey even aimed to write letters after encountering a difficult airway. This was the despite the advice of the anaesthetic literature.^3–5 While some anaesthetists might record the information on the anaesthetic chart or elsewhere in the case notes, these may not always be available when needed. In the emergency setting, or even in the elective setting but at a separate site or hospital, the GP Care Summary may be the only medical summary readily available to the admitting staff. Telling patients directly, although important, can be unreliable, as they may have forgotten or be incapacitated when assessed.

From our data, only 20% of anaesthetists always wrote to primary care about patients with difficult airways and the information was only passed back by GPs half the time. If representative, the chain of communication in patients known to be difficult to intubate was intact in only approximately 10% of cases (Figure 2). This is supported by our finding that only 6% of anaesthetists had ever been alerted to a difficult tracheal intubation by a GP.

The low figure may in part be explained by GP practices using non-medically qualified coders. Furthermore, even if the significance of an event was to be recognised, the correct code may not always be applied, as there are several Read Codes relating to intubation including one for successful tracheal intubation.

A recurring theme, recently highlighted by NAP4, is the need for forward planning when managing the airway. ¹ Given this importance of forewarning, we would suggest that an intact chain of communication in only 6–10% of cases is too low and must be addressed in the interests of patient safety.

Lack of time was identified by anaesthetists as a particular barrier to writing difficult airway letters. Seventy-one also suggested that standardisation would also improve communication. Despite this, our survey revealed considerable variation in the triggers, contents and recipients of difficult airway letters.

Many respondents’ suggestions for improvements are not new. Liban ¹⁰ first proposed expanding the ‘Medic Alert’ database to include difficult airways in 1996. However, to our knowledge, promoting the SP2y3 coding with ‘High Priority’ or ‘Priority 1’ labelling in correspondence with GPs would be a novel strategy.

For the 20% of anaesthetists who write on every occasion, SP2y3 with a statement of priority would improve the effectiveness of their letters. For those who intend to write but don't always succeed, standardised letters using Sp2y3 could easily be issued with just the patient details. For those that do not write at present, asking the surgeon in theatre to include the SP2y3 coding in the operation note/ patient discharge summary might be a simple yet effective way to have the information successfully recorded in a patient's medical record.

A standardised letter including the Read Code would lack absolute clinical details. However, electronically generated care summaries do not have the capacity to contain detailed airway information. The anaesthetist receiving the information would then have to pursue it further but we argue that knowing that there has been a problem is more important than a detailed description of its management. For a junior anaesthetist working out of hours, a line in a GP letter would serve as a useful warning to call a consultant early, avoiding undertaking an unexpected difficult intubation alone.

Our surveys have many limitations. In using a cascade method of distribution, we do not have an accurate denominator and were unable to follow up specific non-respondents. Clearly, our data cannot reflect the views of those who chose not to respond. However, we distributed our questionnaires to encompass the widest possible spectrum of anaesthetists and GPs, rather than to confine ourselves to the members of one specific society or interest group. Surveying GPs was particularly difficult given their operational structure and the decision of some Health Boards to institute moratoria on surveys distribution. However, we felt the inclusion of the opinions of GPs to be essential in looking at the communication between the two groups.

We were surprised to find how many anaesthetists were unaware of the Read Code system, which could improve communication with our GP colleagues – the main beneficiaries being the receiving anaesthetists. We can only speculate how many other specialities fail to communicate effectively with other clinicians because they have not investigated the robustness of their messaging systems.

Conclusion

Our findings illustrate how poor the communication between anaesthetists and GPs might be. Read Codes may not be the natural ‘language’ of the anaesthetist nor of any hospital based practitioner and we accept that the SP2y3 coding is not a panacea for communication between anaesthetists and GPs. Communication can only be effective if firstly it is recorded, then delivered and understood by the intended recipient. Finally, the recipient must be able to take appropriate action based on the information. Our survey found deficiencies in each aspect of this chain, despite the importance given to communication by many bodies and authors.^1–7 We believe that our findings provide an important reminder that communication can be poor despite good intentions. We hope the results will inspire clinicians from all disciplines to examine every aspect of their communication pathways to ensure effective delivery of information, but more importantly to improve patient safety and enhance patient care.