A survey of local palliative care guidelines revealed wide variations in quality. Professor Ilora Finlay, Julie Bowdler and Peter Tebbit give practical advice on adapting national guidelines

There are important distinctions to be made between guidelines and protocols. In essence, protocols are mandatory, whereas guidelines are just what they say they are – a series of steps or statements to guide the management of a particular situation or ondition.

With the increasing move to evidence-based medicine, there is a need for clinical practice to incorporate findings from research and, where no such findings exist to guide practice, to draw on documented best practice to underpin what is done.

Clinical guidelines are a summation of the state of knowledge on a particular condition at a particular time.1 They have been defined as 'a comprehensive set of best practice guidance relating to the management of all aspects of a particular condition.1

Guidelines are also seen as a way of rationalising, or rationing, drug prescribing. They can therefore:

  • Reduce the number of drugs likely to be prescribed for a particular condition
  • Guide pharmacy with regard to stocking of medication
  • Considerably reduce variations in practice and hence variations in outcomes and costs of care.2

There is evidence that guidelines do improve the care that patients receive. Du Pen and colleagues3 demonstrated benefit from the use of algorithm-guided decision-making in the management of cancer pain. In their study, cancer patients randomised to a pain algorithm group achieved a significant reduction in their usual pain intensity compared with controls, who received standard community practice.

Thus guidelines on cancer pain should disseminate up-to-date information on the assessment and diagnosis of pain in cancer patients, and on analgesic prescribing. Producing guidelines involves a great deal of work, underpinned by systematic review, and so the production of nationally agreed guidelines is encouraged, as it allows elements of the development process to be shared, and avoids duplication of work.4

The production of an agreed set of definitive guidelines also avoids the confusion that can arise among clinicians when there is proliferation of guidelines,5 particularly when they appear to be giving slightly different messages.

In 1998 the National Council for Hospice and Specialist Palliative Care Services (NHC) produced guidelines on the management of cancer pain. 6

Since then there had been some discussion in the Clinical Issues Committee of the Council as the guidelines appeared to be patchily implemented and many other guidelines were known to be in local circulation. It was therefore decided to undertake a benchmarking review of locally derived guidelines on control of cancer pain.

A total of 116 different documents were reviewed. Core findings demonstrated that local guidelines were popular. Of the different types, algorithms seemed to present the information most clearly. A typical algorithm is shown in Figure 1.

Figure 1: A typical algorithm on the management of cancer pain

Grimshaw et al7 noted that "Evidence based national guidelines produced by experts are more likely to be valid but least likely to be implemented". Indeed, for cancer pain guidelines this appeared to be the case. Few centres said they used the NHC document, although it was apparent that they may well have referred to it when drawing up their own guidelines. Nationally produced guidelines still need local adaptation to suit local circumstances, and to ensure that local physicians feel a sense of ownership, as this is an important factor in uptake and use.8

In the process of adoption, the quality of content or presentation should not be compromised. However, some of the documents reviewed by the National Council group contained spelling errors, including misspelt drug names, or conflicting advice. These findings support Grimshaw and Russell's statement that "local guidelines, where there is user involvement in their creation, will increase the likelihood of their use, but not their validity".9

Others have had the same experience. Grilli and colleagues5 have been critical of the quality, reliability and independence of practice guidelines, which they obtained via a Medline search. They assessed the quality of a guideline by:

  • Description of the type of professionals involved in developing the guideline
  • Description of the sources of information used to retrieve the relevant evidence
  • Explicit grading of the evidence in support of the main recommendations.

Similar criteria have been developed at St George's Hospital Medical School for judging the acceptability of a guideline development procedure.10

An important part of the recommendations on developing a guideline is that it must be underpinned by systematic review.11 When based on a review, the guidelines can be updated whenever new relevant evidence is published, although it is better to specify on the guideline the review date for the systematic review that underpins it.11

Jackson and Feder12 concluded that an essential component of a successful guideline is the presentation of evidence and recommendations in a concise, accessible format. Decision-makers must be able to retrieve and assimilate information quickly. Moreover, information must be presented in a flexible format that is applicable to specific patients or circumstance.12

In palliative care, a number of drugs are given for indications or by routes for which they are not licensed.13 In the UK, drugs can be prescribed outside their licensed route, but at the discretion and with the responsibility of the prescriber; thus there is potentially a legal responsibility in recommending non-licensed use. There are legal implications in making recommendations in guidelines, potentially involving author and sponsor liability.14

The National Council document clearly emphasises the importance of diagnosing the cause of each pain experienced by a patient with cancer. Each pain should be assessed and the status of the patient reviewed regularly to ensure that pain control is obtained.

The document outlines the management of mild pain with group 1 non-opioid drugs (Figure 2, below) and the use of co-analgesics at all stages.

Figure 2: NHC guidance on the management of mild cancer pain with group 1 non-opioid drugs

Step 1: the NHC position


  • 1000mg/4 hours (elderly may require only 6-hourly) up to maximum recommended 24-hour dose of 4 g
  • Tablets, soluble tablets, suppositories
  • Cheap, safe – liver metastases not an absolute contraindication

Limited effectiveness

Non-steroidal anti-inflammatory drugs (NSAIDs)

  • Wide variety, choice depends upon safety, cost and convenience to patient
  • Should be once or twice daily, easy to swallow (i.e. small and coated) or suspension
  • Try other NSAID if partial relief obtained
  • Caution with increasing age, renal and/or liver impairment, bleeding peptic ulcer
  • Discontinue if no effect after 5 days. If effective, continue by non-oral routes when too ill to swallow
  • May have morphine-sparing effect

Ibuprofen is noted as having lowest GI risk; diclofenac, naproxen and piroxicam are intermediate risk

Evidence of significant increase in frequency of peptic ulceration in patients who die of advanced cancer taking both NSAIDs and steroids. Evidence that misoprostol protects against both gastric and duodenal ulceration;
H2 antagonists only protect against gastric ulceration. Further research needed to clarify guidelines for clinical practice in this area.

Combining paracetamol and NSAID more effective than either alone

For moderate pain the mild opioids, such as codeine- or dihydrocodeine-containing compounds, are recommended (Figure 3, below), and opioids should be titrated to achieve relief of severe pain. The opioid recommended by the HC document is normal-release morphine for titration (Figure 4, below).

Figure 3: NHC guidance on the management of moderate cancer pain with group 2 weak opioid drugs

Step 2: the NHC position


  •  Usually given in combination with paracetamol (co-proxamol) although some doubt whether this is more
    effective than paracetamol alone
  • Equivalent to 1000mg paracetamol and 100mg dextropropoxyphene 3 tablets 4-hourly – elderly may require only 6-hourly – up to a maximum recommended dose of 12 tablets per 24 hours
  • Must be given for 24 hours to allow for time to steady state and full clinical benefit


Codeine, dihydrocodeine

  •  30-60mg codeine 4-hourly
  • Slow-release preparations of dihydrocodeine are more convenient

Side-effects of weak opioids:

  • nausea/vomiting – usually pass off in 3-4 days, but haloperidol 1.5 mg at night is effective short term
  • drowsiness – again for the first few days only,
  • confusion – again for the first few days only,
  • constipation – as for strong opioids, appropriate laxatives should be prescribed when weak opioids are started
Figure 4: NHC guidance on the management of cancer pain not controlled with weak opioids or weak opioid/paracetamol combination

Step 3: the NHC position

Morphine is the treatment of choice

Use oral route

Starting dose

  • usually 10mg of immediate-release morphine
  • 2.5-5 mg for elderly patients, patients with severely impaired renal or liver function, or patients who are very frightened of starting morphine despite reassurance

Dosing frequency

  • usually 4-hourly with immediate-release morphine
  • less frequently in renal or liver impairment (not just liver metastases) and in elderly people – depending on the duration of analgesic effect

Dose increase

  • usually allow 24 hours before considering dose increase to allow time to steady state
  • up to 3-4 days with renal or liver impairment and with elderly people
  • increase dose by 30-50%

No maximum dose. Dose is titrated upward until pain is relieved. Some patients, particularly young adults, may require more than 1000mg of morphine per 24 hours

Breakthrough pain – pain which occurs before next regular dose of analgesia is due

  • usually due to inadequate regular dose
  • give extra dose of morphine equivalent to 4-hourly dose when pain occurs
  • consider increasing regular morphine dose
  • may occur when patients are taking slow-release morphine preparations. These patients should have immediate- release morphine available for breakthrough pain

Incident pain – pain which occurs with movement, weight-bearing, dressing change or some other incident. Patient is usually pain-free at rest

  • avoid increasing regular morphine dose
  • exclude surgically correctable lesion such as impending or actual bone fracture
  • give 4-hourly equivalent dose of morphine 30 minutes before the incident which produces the pain
  • alternatives include dextromoramide 5-20mg 15 minutes before incident or patient self-administration of nitrous oxide/oxygen mixture at time of incident

NHC alternative routes of administration

 Although most patients will only require oral medication, alternative routes include:

  • Rectal
  • Subcutaneous (do not use intramuscular route)
  • Sublingual

Diamorphine is the opioid of choice for subcutaneous administration. Its high solubility permits delivery of high

  • concentrations in low volumes
  • one third of oral dose
  • 4-hourly when given by intermittent injection
  • continuously by portable syringe driver

Indications for parenteral administration of opioids

  1. Persistent nausea/vomiting, e.g. inoperable bowel obstruction
  2. Inability to swallow oral medications, e.g. oesophageal obstruction
  3. Altered consciousness level, e.g. coma when patient is dying

Epidural and intrathecal route – opioids and alternative drugs

  • indications still to be clearly defined
  • considered if appropriate doses of opioids and another analgesics cause unacceptable side-effects when given
    by other routes
  • requires experienced personnel
  • preparations must be prepared under aseptic conditions

Guidance on the use of adjuvants (Figure 5, below) is also given in the document.

Figure 5: NHC guidance on the use of adjuvants in the management of cancer pain

Adjuvants – the NHC position

 Adjuvants defined in NHC as drugs which contribute to pain relief in certain situations without being classical

analgesics. They are used in combination with the medications described above.

Steroids – can be considered for patients with

  • nerve compression pain
  • neuropathic pain
  • bone pain
  • painful hepatomegaly
  • headache from raised intracranial pressure (dexamethasone)

Antidepressants – for neuropathic pain (amitriptyline, desipramine, imipramine)

Anticonvulsants – for neuropathic pain (clonazepam, carbamazepine, sodium valproate)

Antiarrhythmic drugs – for neuropathic pain (lignocaine, flecainide, mexiletine) seek specialist advice

Antispasmodics – for intestinal colic (e.g. hyoscine, butylbromide, hyoscine hydrochloride)

Bisphosphonates – for bone pain (pamidronate clodronate) seek specialist advice

Although the NHC guidelines do not recommend a specific option when pain is controlled, several preparations are described in the section on severe pain. Fentanyl, phenazocine, methadone and hydromorphone are included as alternative strong opioids to morphine, but pethidine, buprenorphine and nefopam are to be avoided.

The submitted local guidelines indicate a widespread use of slow-release morphine or transdermal fentanyl for maintenance therapy, suggesting that these are the most commonly prescribed sustained-release preparations.

The NHC guidelines are due for review. They must take account of the newer opioids available, such as oxycodone. In the meantime, some local guidelines should be revised urgently to avoid errors in drug use.

Some general recommendations can be made for the production of clinical guidelines in palliative care (see Table 1 below).

Table 1: A checklist for locally derived guidelines
The document should meet the following criteria:
Titles etc
Has clear title indicating the guideline content
Identifies the intended user in a prominent position
States its source
States a publication date and review date
States any sponsor's name
The size is appropriate for the user, e.g. fits a doctor's white coat pocket
The binding is not too tight or bulky
Readable font (12 point is recommended)
Colour assists information retrieval, even when photocopied
Index, contents and headings assist information retrieval
Content decided through consultation with intended users
Notes unlicensed recommendations of drugs or dosages
Gives all generic drug names
References the evidence, and strength, and on which it is based
Indicates where/when specialist help can/should be obtained
Has been proof-read meticulously

It is essential that any guidelines state clearly who has produced them and who the intended users are. The evidence base that underpinned their production must be clearly referenced, and the strength of this evidence stated. Guidelines must be dated, and state the date they are due for revision.

The presentation style requires attention: the typeface must be easily read, uncluttered by unnecessary colour or graphics, printed in a format that is easily carried around, and preferably fit in a doctor's white coat pocket. The binding is also important, as booklets bound too tightly are difficult to use.

Grouping information in a logical manner, and a contents list and index, will improve access to the information. All drugs should be quoted by their generic name, even when the trade name is used, and any recommendations for uses other than those stated in the manufacturer's Summary of Product Characteristics/Data Sheet must be clearly stated.

Ways of accessing specialist advice should be stated. Finally, meticulous proof-reading is essential.

  • Note: See 'In Reply', October 2000 for a letter from Professor Finlay regarding this article

  1. NHS Executive. Faster Access to Modern Treatment; how NICE appraisal will work. Leeds: NHS Executive, January 1999.
  2. Norheim OF. Healthcare rationing – are additional criteria needed for assessing evidence based clinical practice guidelines? Br Med J 1999; 319: 1426-9.
  3. Du Pen SL, Du Pen AR, Polissar N et al. Implementing guidelines for cancer pain management: results of a randomised controlled clinical trial. J Clin Oncol 1999; 17(1): 361-70.
  4. Miller J, Petrie J. Development of practice guidelines. Lancet 2000; 355: 82-3.
  5. Grilli R, Magrini N, Penna A, Mura G, Liberati A. Practice guidelines developed by specialty societies: the need for a critical appraisal. Lancet 2000; 335: 103-6.
  6. Working Party on Clinical Guidelines in Palliative Care. Guidelines for Managing Cancer Pain in Adults. 2nd edn. London: The National Council for Hospice and Specialist Palliative Care Services, 1998.
  7. Grimshaw J, Eccles M, Russell I. Developing clinically valid practice guidelines. J Eval Clin Pract 1995; 1(1): 37-48.
  8. Grimshaw J, Russell I. Effect of clinical guidelines on medical practice. A systematic review of rigorous evaluations. Lancet 1993; 342: 1317-22.
  9. Grimshaw J, Russell I. Achieving health gain through clinical guidelines: II. Ensuring guidelines change medical practice. Qual Health Care 1994; 3: 45-52.
  10. Littlejohns P, Cluzeau F, Bale R, Grimshaw J, Feder G, Moran S. The quantity and quality of clinical practice guidelines for the management of depression in primary care in the UK. Br J Gen Pract 1999; 49: 205-10.
  11. Shekelle PG, Woolf SH, Eccles M, Grimshaw J. Clinical guidelines: developing guidelines. Br Med J (Clinical Research Ed.) 1999; 318: 593-6.
  12. Jackson R, Feder G. Guidelines for clinical guidelines (editorial). Br Med J (Clinical Research Ed.) 1998; 317: 427-8.
  13. Atkinson CV, Kirkham SR. Unlicensed uses for medication in a palliative care unit. Palliat Med 13(2): 145-52.
  14. Hurwitz B. Legal and political considerations of clinical practice guidelines. Br Med J (Clinical Research Ed.) 1999; 318: 661-4.

Guidelines in Practice, July 2000, Volume 3
© 2000 MGP Ltd
further information | subscribe