The “jaundice hotline” for the rapid assessment of patients with jaundice

Background and setting

The acutely jaundiced patient requires rapid assessment, diagnosis, and treatment. Initial assessment should include history, examination, laboratory investigations, and abdominal ultrasonography.1 One possible diagnosis is hepatobiliary malignancy, so rapid diagnosis and treatment are important to avoid evoking considerable anxiety in the patient. This is particularly relevant in the United Kingdom, because a recent government initiative has dictated that patients with a suspected diagnosis of malignant disease must be seen by a specialist within two weeks.2 Jaundice fulfils the criteria for referral under this scheme. This ruling has put considerable strain on existing health resources and has required reorganisation of services, particularly in specialties such as gastroenterology.

The Royal Cornwall Hospital serves a largely rural population of 385 000. The hospital provides endoscopic retrograde cholangiopancreatography for an additional 70 000 people in the west of the county. Transport links are poor, and travel to and from the hospital can be difficult and expensive. The county is one of the poorest in the United Kingdom.3

The gastrointestinal unit is staffed by three consultant gastroenterologists, two consultant gastrointestinal surgeons, and three gastrointestinal radiologists. There is also one specialist registrar, one senior house officer, and one preregistration house officer.

The problem

In 1996 concerns were raised by local gastroenterologists and general practitioners over the management of patients with acute jaundice. Pressure on outpatient clinics and radiology services resulted in unacceptably long waiting times for assessment. A perception among general practitioners was that patients would be better off admitted acutely, resulting in long stays in hospital for patients while they awaited appropriate investigations and treatment, often under the care of staff other than gastroenterologists. A retrospective audit of the records of 71 consecutive patients admitted with jaundice over a three month period showed that 57 of these patients had been admitted directly to hospital under a wide range of specialties, and 21 had come under the care of medical gastroenterologists. The mean duration of stay in hospital was 11.5 days. Some patients experienced unacceptable delays before appropriate diagnosis and treatment took place.

A reorganisation of the management of acutely jaundiced patients was needed. Our aims were to shorten waiting times for consultation, investigation, and treatment and minimise stay in hospital while providing a high quality service run by gastroenterologists. The gastrointestinal unit was already involved in developing a “one stop” approach to several clinical problems such as dyspepsia, dysphagia, and iron deficiency anaemia.4 We were interested in developing a similar service for the jaundiced patient. Financial and staffing constraints meant that the initiative would have to be established within the existing service framework and with a minimum of extra expenditure.

Key measures for improvement

From our analysis of the audit data and from speaking to general practitioners, hospital consultants, and junior doctors, we identified several areas for improvement.

Strategies for change

The “jaundice hotline” was established in November 1998 and piloted for one year. It was publicised initially by mailshots to primary care practices and then reinforced by several workshops. Presentations were made in hospital to junior and senior medical staff.

The service is open to patients with acute jaundice. Referrals are made through a dedicated, 24 hour telephone answering machine and fax line and are processed by a secretary. Referrals are not scrutinised by a clinician. Apart from the purchase of an answering machine, further resources were not required. Existing staff took on the extra secretarial work.

The clinics occur twice weekly. For the first year they took place in the endoscopy unit. Each clinic was designed for up to five patients. All cases are discussed with either a specialist registrar or consultant gastroenterologist. The structure of the jaundice hotline is shown in the figure.

Structure of the jaundice hotline

• Download figure
• Open in new tab
• Download powerpoint

Effects of change

We collected data on time from referral to consultation, time from consultation to endoscopic retrograde cholangiopancreatography, final diagnosis, and duration of stay in hospital. Notes of all patients seen in the first year were reviewed retrospectively. We asked 36 primary care practices and 30 consecutive patients seen in the first six months to rate the quality of the service on a scale of 1 (poor) to 5 (excellent).

Of the 107 patients seen in the first year, 57 were male. The mean age was 63 (range 15-97) years. The mean time from referral to attendance at the jaundice clinic was 2.5 (range 0-12) days. Seven patients were seen on the day of referral.

In 62 patients the jaundice was due to obstruction. The cause of obstruction was common bile duct calculi in 26 patients and biliary malignancy in 30 patients. In 59 patients endoscopic retrograde cholangiopancreatography was subsequently performed. One patient was immediately given percutaneous trans-hepatic cholangiography, and in two patients further investigation was deferred at the patient's request. In five patients whose final diagnosis was parenchymal liver disease endoscopic retrograde cholangiopancreatography was performed as part of the initial investigation. The average time between initial consultation and endoscopic retrograde cholangiopancreatography was 5.7 (range 0-35) days. Where a decision to proceed to endoscopic retrograde cholangiopancreatography was made at the time of the patient's attendance at the clinic, the mean wait was 4.6 (range 0-10) days.

In 45 patients the cause of jaundice was non-obstructive or unknown. Parenchymal diagnoses included alcohol related liver disease (nine patients), hepatitis B (six), autoimmune liver disease (six), drug induced hepatitis (three), and metastatic disease (four). There were two cases of Gilbert syndrome and one case of congestive cardiac failure.

Six patients were admitted directly after attendance at the clinic. Seventy patients were admitted for at least one day during their further investigation and treatment. Of these patients, 46 were inpatients for 3 days, and most patients attending for endoscopic retrograde cholangiopancreatography stayed one night only. The overall mean duration of hospital stay for all patients admitted through the jaundice hotline was 5.5 (range 1-33) days, compared with a mean of 11.5 days obtained in the previous audit. Gastroenterologists managed all inpatients.

Of the 36 primary care practices sent a questionnaire, 34 rated the service as above average or excellent (4 or 5 on the scale). Of 30 consecutive patients who were asked to rate the service at the end of their first visit to the clinic, 29 gave a score of 4 or 5.

We believe we achieved our aims to provide rapid access to high quality consultation, investigation, and treatment for jaundiced patients and to reduce their stay in hospital. All patients are seen by a gastroenterologist, and times from referral to consultation are short. The reduced stay in hospital may save a substantial number of bed days a year, and this saving may increase as more patients with jaundice are referred to the service. It must be emphasised, however, that only when all patients with jaundice are seen by the service can an accurate estimate of the number of bed days saved be made.

Initially referral rates were low, which is reflected in the relatively small number of patients seen in the first year in comparison with the numbers predicted from the initial audit. However, as publicity has increased, most jaundiced patients in our district are now being assessed in this way, and the service is now fully established and in its third year.

Short waiting times and reductions in hospital stay do not, however, necessarily lead to a higher quality of clinical care. These are surrogate markers only. Unfortunately, no data on medical outcome and patient satisfaction are available from the initial audit, so a direct comparison is not possible.

Lessons learnt

We encountered several problems in developing the jaundice hotline. Services offering referrals that are not vetted by a clinician are open to inappropriate referrals. To address this we specified that patients must have acute (<4 weeks), clinically evident jaundice. Cases where patients did not meet this criterion would need to be discussed with a clinician. This approach, combined with continuing education of our clinical colleagues, has minimised inappropriate referral.

The service is particularly well suited to patients whose jaundice is due to an obstruction, as management of these patients is relatively structured. Hepatocellular jaundice, however, is more complex, and requires expert assessment, especially in the context of hepatic failure. Our initial experiences were that some patients with hepatocellular jaundice were not managed as well as we would have liked. This prompted the introduction of stricter guidelines for admission and the immediate involvement of a gastroenterology consultant or registrar in the management of all such patients.

As the jaundice hotline became more popular it was no longer practical to hold the clinics in the endoscopy department and so the clinic was moved to the short stay ward. The nursing philosophy and skills mix of the staff on this ward have been ideally suited to the hotline and its further development. We now actively involve our nursing staff in the assessment and management of our patients, further streamlining the service.

Clear and effective communication is essential in any outpatient service. Initially, this took the form of a letter dictated by the consultant after outpatient review or endoscopic retrograde cholangiopancreatography. This produced the kind of delay that we had hoped to avoid. We now forward a copy of the hand-completed proforma immediately to the general practitioner after the initial consultation. This is followed by a formal report once all investigations have been completed.

Applying an open referral approach such as used in the jaundice hotline to other clinical problems may be more difficult. Jaundice is a very distinct presenting complaint. Patients are either jaundiced or not. Presenting symptoms of other gastrointestinal problems may be less discrete, making an open access service far less practical.

Although such a service needs a structure and guidelines to function efficiently, patients rarely conform to protocol. Flexibility has been essential to allow our service to accommodate the vagaries of patients' presentations at the clinic.

However, the most important lesson we have learnt is that, with a combination of reorganisation and, more importantly, teamwork, major improvements can be made in health service without the need for significant extra resources.