Dr John Orchard describes the development of a demand management tool which enables GPs to prioritise patients for hip and knee replacements and reduces waiting times

The North Amber Valley locality commissioning project was one of the original 42 projects funded in October 1997. It consists of 37 GPs working in nine practices based in the towns of Alfreton, Ripley and Crich. Situated on the Northern border of Southern Derby-shire Health, the practices serve a population of 62000 spread across three local authority boundaries.

Central to the scheme is the local geography. Taking the largest town of Ripley, with its cottage hospital as the epicentre, patients are equidistant (approximately 12 miles) from the nearest district general hospitals in Derby, Chesterfield and Mansfield.

As a result, the GPs have a long history of inter-practice cooperation to provide:

  • Services at the cottage hospital
  • Out-of-hours cover
  • Postgraduate education meetings.

Of the nine practices in 1997, four were fundholders and five were non-fundholders. The fundholders brought to the group the experience of service developments within their practices, which highlighted the two-tier system that existed at the time.

Discussions between the practices identified two major areas of concern:

  • Waiting time for operations, specifically large joint replacements
  • The ability of elderly patients to access the services.

It was resolved to address these concerns by:

  • Maximising the use of existing facilities
  • Bringing services closer to patients
  • Devising a system of accessing care based on clinical need alone, in order to remove the inequities created by fundholding. Specifically, on several occasions patients living next door to each other, but registered with different practices, yet suffering from the same condition, i.e. osteoarthritis of the hip, had vastly different waiting times for operation: in one case, 3 months at the fundholding practice, and in another case, 2 years at a non-fundholding practice.

It was decided to identify the most efficient provider of orthopaedic services, to negotiate the shortest possible waiting time, and to establish a consultant-led service at the cottage hospital.

It was also thought desirable to maximise the use of existing facilities at the cottage hospital, such as physiotherapy, radiology, ECG, and phlebotomy. The cottage hospital at Ripley had sufficient capacity to accommodate an extra clinic and negotiations were undertaken with the Southern Derbyshire Acute Trust, Chesterfield, and North Derbyshire Royal Hospitals and King's Mill Hospital, Mansfield.

The best deal proved to be a 9-month wait (from seeing the consultant) and a clinic staffed by a consultant and a specialist nurse, run on a monthly basis from Chesterfield Royal Hospital.

In order to take the project forward, it was necessary for all the stakeholders to own the process. Direct meetings were arranged with the surgeons concerned, business managers, their representatives and ancillary staff, physiotherapists and GPs.

A protocol for referral of patients requiring large joint replacements was devised with input from all these groups. A subgroup for each protocol then met to identify the practicalities and ensure that the proposed service could be delivered. The formulated protocols were then presented to all the participating clinicians and arranged specifically for the purpose.

  • Access to the clinic was to be limited to members of the North Amber Valley locality commissioning pilot scheme; this was important since practices not belonging to the pilot scheme also used Ripley Hospital.
  • No referrals for large joint replacement were to be accepted unless accompanied by a completed protocol. (This would involve extra reports, physiotherapy assessments and the completion of a protocol form). Incomplete referrals were to be returned to the referring GP by the administrative staff of the cottage hospital.
  • The majority of investigative assessments were to be delegated to primary care, so that the surgeon already had the the results of the assessments when the patient was referred.
  • The patients attending a hospital would be limited to a single appointment with the consultant, who had a greater amount of time to spend with them since the number referred to this clinic had been reduced by the application of the protocol in primary care.
  • Preoperative and postoperative clinic attendance would take place at the cottage hospital and not the district general hospital.

Before referral, all patients should have:

  • An examination of the affected joint and adjacent joints. These findings were to be recorded in the referral letter.
  • X-rays of the affected joint, and corresponding joints of the unaffected limb. These X-rays were agreed with the consultant concerned so that the correct X-ray was taken on each occasion.
  • A physiotherapy assessment. This would be undertaken before referral to the consultant so that if there was any possibility of improvementgin the condition with physiotherapy, this could be carried out before the consultant saw the patient.
  • A trial of intra-articular steroid, if appropriate, and an essential needs assessment.

The scoring system for assessment and its interpretation are shown in Tables 1 and 2 respectively.

Table 1: Scoring system for physiological assessment

1. The joint has a decreased range of movement

  Score
a. No decrease Nil
b. Decreased One
c. Fixed deformity Two

2. The X-ray shows osteoarthritis

a. No osteoarthritis Nil
b. Osteoarthritis present One
c. Severe osteoarthritis with loss of joint space Two
3. The physiotherapy report
a. Good response to physiotherapy Nil
b. Some response to physiotherapy One
c. No response to physiotherapy Two
4. Local steroid injection (knees only)
a. Good response Nil
b. No response Two

5. The patient requires a walking aid

a. No aid required Nil
b. Aid required One
6. Social and economic need
a. None Nil
b. Severe Two

 

Table 2: Interpretation of the scoring system

Total score six or more Joint replacement likely, therefore refer

Score of four to six Joint replacement possible – reassess in six months

Score less than four Reassess in 12 months

Each protocol point represents an instance at which the decision to refer may be deferred. The protocol points also highlight procedures that can easily be undertaken in general practice, and guide the GP as to what constitutes an appropriate referral. For example:

  • Where the patient is shown not to have osteoarthritis on X-ray, it is unlikely that he/she is going to require a joint replacement
  • Where the physiotherapy report indicates that the patient can benefit from physiotherapy, there is no need to refer to a consultant first.
  • Where the patient can benefit from a steroid injection, clearly there is no need to refer to a hospital for this procedure to be undertaken.

In order to assess the outcome of this intervention, it was decided to review the scheme one year after the first outpatients had been referred to Ripley Hospital. In addition, the last 50 cases that had been operated on at Chesterfield Royal Hospital were assessed to compare the following important parameters:

The number of times that patients attended the outpatient clinic before being placed on a waiting list.
How priority for a patient's operation was decided.

Whether patients had been:

a) X-rayed

b) assessed by a physiotherapist

and if this influenced the decision to operate.

The total length of time from the patient being referred by the GP to the time that he/she was operated on (referral to done time).

A survey of the GPs involved was undertaken after 10 months. With only one exception, those who replied felt that using the protocol had enabled them to make a more effective decision about whether or not to refer a patient for surgery.

The protocol enabled the GPs to decide to defer referral of a patient for surgery on the grounds that their condition was simply not bad enough.

Several GPs commented that it did require the patient to visit them four times before referral was made: initially with their complaint, then for X-ray, then physiotherapy, and subsequently to assess all the results before referral. It should be empsasised, however, that this was very convenient for the patient, and not particularly time-consuming for the GP.

Number of outpatient attendances:

The average number of outpatient attendances at Chesterfield Royal Hospital was 2.7 per patient, including physiotherapy but excluding preoperative assessment. The range was 1-4. Attendances at Ripley Hospital were 1.0 per patient.

Indications of priority:

At Ripley Hospital it was agreed that by using the protocol the patients would be operated on within 9 months. All 37 patients referred to Ripley Hospital in the period of the trial were listed for operation within 9 months.

A listing of 9 months at Chesterfield Royal Hospital constituted a 'soon' listing. The average intended wait at Chesterfield Royal Hospital was 16 months; of the patients who attended Chesterfield Royal Hospital, 14 of 49 were listed as 'soon' and three were listed as 'urgent' referrals. All of these would be operated on in less than 16 months (range 4-13 months).

The GPs who referred patients to Chesterfield Royal Hospital had written specifically to expedite 11 cases, but in only three of these did the consultant concur with the request, and agree that they merited moving to a soon or urgent status.

Only one of the patients at Ripley Hospital was operated on more urgently than 9 months and none were refused operation at 9 months, i.e. all were found to be of a soon or urgent status.

Radiology and physiotherapy:

X-rays were taken in only 23 of 49 patients attending Chesterfield Royal Hospital before their referral. At Ripley Hospital, 100% of patients had been X-rayed before referral according to the protocol.

Of the patients seen in outpatients at Chesterfield Royal Hospital, only two had had physiotherapy assessments before referral and they had come via the rheumatology department. In contrast, at Ripley Hospital 100% of the patients had had a physiotherapy assessment, which indicated that they would not benefit from further intervention.

The average referral to done time at Chesterfield Royal Hospital was 16.5 months (range 4-58 months). At Ripley Hospital it was 10 months (this represents a 9-month wait from seeing the consultant, as agreed).

Fifty-seven patients were referred for assessment by the physiotherapists. Twenty patients were not referred on for the consultant's clinic because:

  • 17 improved with physiotherapy
  • two were referred back to the GP for joint injection
  • one did not attend.

Of the referrals, 35.1% were therefore identified as amenable to treatment without the necessity for attendance at an outpatient clinic. This was cost-effective. The basic cost of a physiotherapy assessment is approximately 25% of that of an orthopaedic outpatient attendance.

  • The average wait of the control group was 16.5 months from the date of referral.
  • The average wait of the protocol group was 10 months from referral.
  • The waiting time for physiotherapy assessment in general practice was 2 weeks.

The results are summarised in Table 3.

Table 3: Summary of results

Parameter measured CNDRH Pilot
Outpatient attendance (nos) 2.7 1
Priority agreed by GP and specialist 6% 100%
X-ray before referral 48% 100%
Physiotherapy assessment 5% 100%
Physiotherapy treatment, not referred 0% 35%
Referral to done time (average) 16.5 months 10 months

CNDRH = Chesterfield & North Derbyshire Royal Hospital

This scheme demonstrates that:

  • GPs can adequately prioritise patients for surgery, using local facilities, and that the priority assigned agrees with that given by a consultant orthopaedic surgeon.
  • Use of the full range of facilities available to primary care enables GPs to establish appropriate criteria for referral.
  • By the application of a protocol, it is possible to ensure that people arriving in an outpatient department are likely to be those suitable for surgery. This reduces the number of outpatient attendances required, making more efficient use of the consultant's time.
  • It is much more economic for patients to be seen in the community where the cost of specific X-rays and physiotherapy intervention is likely to be considerably less than in a district general hospital. In this respect, the use of this kind of demand management tool is likely to lead to a more efficient health service overall.
  • It is possible to obtain cooperation between primary and secondary care. It requires a degree of trust on both sides, but overall will lead to gains in efficiency for both patients and doctors.

The 9-month wait for hip replacements in this study was negotiated as part of the study and does not represent the normal waiting time at Chesterfield Royal Hospital.

  • Acknowledgements: Mr KA Ennis and Mr IR Scott, Consultant Orthopaedic Surgeons, Chesterfield Royal Hospital.

Guidelines in Practice, October 1999, Volume 2
© 1999 MGP Ltd
further information | subscribe