Letters

Guideline for primary care management of dementia

Such guidelines should consider all relevant effective treatments

Guideline should cover differential diagnosis

GPs may want to continue prescribing donepezil for patients

Some recommendations given are not based directly on evidence cited

Authors' reply

Such guidelines should consider all relevant effective treatments

EDITOREvidence based guidelines have a responsibility to consider all the relevant effective treatments and not to concentrate only on those with which clinicians are familiar. I was surprised by a serious omission from the North of England evidence based guidelines for the primary care management of dementia¹of category I evidence for the effectiveness of Ginkgo biloba extract in dementia, from a large randomised controlled trial.² The number needed to treat for a 4 point improvement in the cognitive subscale of the Alzheimer's disease assessment scale at one year of follow up has been calculated as 7.9 (95% confidence interval 4.2 to 67); for a significant improvement in the geriatric assessment by relative's rating instrument (a daily living and social behaviour score assessed by family members) it was 7.0 (3.3 to 97). The dose of G biloba extract was 120 mg a day.³

G biloba extract is available over the counter, and the cost of a year's treatment (from one major supermarket) is £85. A year's treatment with donepezil, by contrast, costs £891 for 5 mg and £1248 for 10 mg.

Tom Marshall, Honorary lecturer in public health medicine. 
Department of Public Health and Epidemiology, University of Birmingham, Birmingham B15 2TT

^a MarshaTP{at}HSRC1.bham.ac.uk

Competing interests: None declared.

1. Eccles M, Clarke J, Livingstone M, Freemantle N, Mason J, for the North of England Evidence Based Dementia Guideline Development Group. North of England evidence based guidelines development project: guideline for the primary care management of dementia. BMJ 1998; 317: 802-808[Free Full Text]. (19 September.)
2. Le Bars PL, Katz MM, Berman N, Itil TM, Freedman AM, Schatzberg AF, for the North American RGb Study Group. A placebo-controlled, double-blind, randomized trial of an extract of Ginkgo biloba for dementia. JAMA 1997; 278: 1327-1332[Abstract].
3. Dementia: diagnosis and treatment. Bandolier 1998; 5(2): 2-3.

Guideline should cover differential diagnosis

EDITORThe summary version of the guideline for primary care management of dementia is inadequate and in some respects inaccurate.¹ Not to cover differential diagnosis is a major omission; treatments for Alzheimer's disease now exist, and general practitioners and the primary healthcare team need to identify patients with vascular dementia and cognitive impairment (many such patients have treatable risk factors) as well as those with Lewy body dementia (for which diagnostic advice is in fact provided). The guideline should at least have recommended referral to specialist services, since distinguishing between the dementias is one of our most important clinical problems and an area where guidelines are most urgently required. Advice about when to refer to social services is essential.

The section on physical screening failed to recommend a physical examination, which is an essential part of the assessmentto identify treatable vascular risk factors, for example. Recommended routine screening tests should have included vitamin B-12 assay and liver function tests. The search strategy and synthesis included only the findings of studies published before 1996, apart from two 1998 references. Thus many of the statements about drug treatment are inaccurate, and the recommendations are misleading. Rivastigmine, recently licensed for Alzheimer's disease, is not mentioned; tacrine is not available in the United Kingdom; and velnacrine was never licensed.

Adequate evidence now exists that donepezil improves cognitive and global functioning in some patients with Alzheimer's disease and that these benefits may be maintained with long term use.² Cost analysis models suggest that donepezil is cost neutral,³ and therefore treatment should not be withheld on financial grounds. Specialists should start and supervise such treatment,⁴ but no evidence based support exists for the recommendation that general practitioners should not continue to prescribe it if it seems beneficial.

The guideline states that respite care does not reduce the burden of caring for a person with dementia. Research commissioned by the Department of Health found that almost four fifths of carers reported that respite care had made their life better.⁵

Finally, the guideline should reflect the fact that in many practices the lead professional for detecting and managing dementia will be a nurse rather than a doctor.

Rebecca Eastley, Clinical research fellow. 
Judy Haworth, Clinical research fellow. 
Gordon Wilcock, Professor in care of the elderly. 
Department of Care of the Elderly, Frenchay Hospital, Bristol BS16 1LE

Deborah Sharp, Professor of primary health care. 
University of Bristol, Division of Primary Health Care, Bristol BS8 2PR

1. Eccles M, Clarke J, Livingstone M, Freemantle N, Mason J, for the North of England Evidence Based Dementia Guideline Development Group. North of England evidence based guidelines development project: guideline for the primary care management of dementia. BMJ 1998; 317: 802-808. (19 September.)
2. Rogers SL, Friedhoff LT. Long-term efficacy and safety of donepezil in the treatment of Alzheimer's disease: an interim analysis of the results of a US multicentre open label extension study. Eur Neuropsychopharmacol 1998; 8: 67-75[Medline].
3. Stewart A, Phillips R, Dempsey G. Pharmacotherapy for people with Alzheimer's disease: a markov-cycle evaluation of five years' therapy using donepezil. Int J Geriatr Psychiatry 1998; 13: 445-453[Medline].
4. Standing Medical Advisory Committee. The use of donepezil for Alzheimer's disease. London: Department of Health , 1998.
5. Levin E, Moriaty J, Gorbach P. Better for the break. London: HMSO , 1994.

^b Competing interests: None declared by RE or DS. JH is partly employed to undertake research into new treatments for dementia and has acted as an adviser to a pharmaceutical company developing treatments for dementia. GW has been reimbursed by Novartis, which makes drugs for Alzheimer's disease, for attending a symposium and has received a fee for speaking; he has also received funding from the pharmaceutical industry for clinical trial work on drugs for Alzheimer's disease, mostly on a "per patient" basis, and this funds staff salaries. The pharmaceutical companies include and have included Janssen, Eisai, and Shire Pharmaceutical Development.

GPs may want to continue prescribing donepezil for patients

EDITORThe guideline for the primary care management of dementia seems to confuse the lack of evidence of benefit with evidence of lack of benefit regarding the effects of donepezil in improving function in patients with dementia.¹ The recommendation that general practitioners should not continue donepezil started in hospital is given a strength of A (that is, it is directly based on evidence from well designed randomised controlled trials, meta-analyses, or systematic reviews). This is clearly inconsistent with the authors' acknowledgment of the limitations of current knowledge.

We have one patient aged 63 with severe dementia who was prescribed donepezil and whose wife reported considerable benefits. The responsible consultant withdrew it after one year to assess whether it was still useful. The patient's wife reported deterioration in his functioning, which has been reversed by restarting the drug. Although this is not a placebo controlled individual trial, it is one means of assessment in the real world. As general practitioners we have not yet been asked to prescribe this drug, but if secondary care were to stop prescribing it we would continue it. Failure to do so would seriously damage our relationship with the patient's wife, who believes that she has convincing evidence of its benefit. To use the guidelines to claim that "we know better" in the absence of relevant evidence seems the height of professional arrogance.

Evidence based medicine loses some credibility if it attempts to control behaviour in the vast number of clinical situations for which hard evidence is lacking. As professionals we must be prepared to admit to the limits of our knowledge and share these limitations with our patients.

William G Tapsfield, General practitioner. 
Diana M Jelley, General practitioner. 
Collingwood Surgery, North Shields NE29 0SF collingwood_surgery{at}cableinet.co.uk

1. Eccles M, Clarke J, Livingstone M, Freemantle N, Mason J, for the North of England Evidence Based Dementia Guideline Development Group. North of England evidence based guidelines development project: guideline for the primary care management of dementia. BMJ 1998; 317: 802-808. (19 September.)

^c Competing interests: None declared.

Some recommendations given are not based directly on evidence cited

EDITORWe are concerned that some of the recommendations in the evidence based guidelines for the primary care management of dementia are not based directly on the evidence cited, thus undermining the evidence based approach.¹ Some recommendations are tentative ("General practitioners should consider using formal cognitive testing") despite good evidence to support them. Others are quite categorical ("General practitioners should not initiate treatment with donepezil") but not based on any published studies of the prescription of donepezil in primary care. The advice that general practitioners should not initiate treatment with donepezil (which is out of step with the Standing Medical Advisory Committee's guidelines²) is based only on a judgment made by the authors.

The recommendation (cited as based on category I evidence) that general practitioners should not continue prescriptions of donepezil started in hospital is again not founded on any published evidence. In fact there is evidence from category I studies that, in patients who have responded to donepezil, stopping the drug leads to clinical deterioration.³ This is supported by our experience in Southampton Memory Clinic, where we have used discontinuation of donepezil as a useful adjunct to clinical assessment and rating scales in some patients when response was uncertain.

If evidence is lacking this needs to be made clear; the authors seem to confuse no evidence of effectiveness with evidence of no effectiveness. Statements that are said to be based on evidence but are not lend a spurious scientific respectability to the recommendations. This blurs the boundary between evidence and opinion, which the article aims to keep separate.

H Matthews, Consultant in old age psychiatry. 
Western Community Hospital, Southampton SO16 4XE

D Wilkinson, Consultant in old age psychiatry. 
C Holmes, Senior lecturer in old age psychiatry. 
Moorgreen Hospital, Southampton SO30 3JB

1. Eccles M, Clarke J, Livingstone M, Freemantle N, Mason J, for the North of England Evidence Based Dementia Guideline Development Group. North of England evidence based guidelines development project: guideline for the primary care management of dementia. BMJ 1998; 317: 802-808. (19 September.)
2. Standing Medical Advisory Committee. The use of donepezil for Alzheimer's disease. London: Department of Health , 1998.
3. Rogers SL, Farlow MR, Doody RS, Mohs R, Friedhoff LT and the Donepezil Study Group. A 24 week, double-blind placebo-controlled trial of donepezil in patients with Alzheimer's disease. Neurology 1998; 50: 136-145[Abstract/Free Full Text].

^d Competing interests: HM has received a grant from Eisai and Pfizer towards running costs of the Southampton Memory Clinic. DW has undertaken phase III trials for Eisai, Pfizer, Novartis, and Bayer of cholinesterase inhibitors and has received sponsorship from these companies for attending conferences and as a member of their advisory boards. CH has no competing interests.

Authors' reply

EDITORMarshall and Eastley et al identify potentially relevant studies that were published after the period covered by the systematic review on which the guideline was based. Such studies will be incorporated when the guideline is reviewed.

Three of the letters question the evidence for and the subsequent recommendations on the use of donepezil. The meta-analysis of the effectiveness of donepezil within the guideline is based on all the data that were available (data from 1102 patients). It shows a significant effect on cognitive function, no measurable changes in quality of life, and inadequate evidence of effects on activities of daily living. The average effect on the cognitive subscale of the Alzheimer's disease assessment scale (an improvement of 2.8 points) was remarkably consistent, with only 1 patient in 40 achieving a benefit of more than 3.4 points. The guideline development group was thus unconvinced of the clinical importance of the effect of donepezil on cognitive function and questioned the widespread use of the drug by general practitioners. It also felt that responsibility for prescribing should lie with clinicians likely to prescribe the drug regularly. This is compatible with the Standing Medical Advisory Committee's guidance.

Eastley et al raise the issue of whether we should have included differential diagnosis. We chose explicitly not to do so, as we regarded this as a function of secondary care. Dementia with Lewy bodies was specifically mentioned because of the need to avoid neuroleptic drugs in patients with dementia of this type. Eastley et al further suggest that recommended routine screening tests should have included vitamin B-12 assay and liver function tests. We could not identify any evidence to support such recommendations. Their final point, about who is likely to be the lead responsible clinician in a practice, is an implementation issue to be decided at a local level.

Martin Eccles, Professor of clinical effectiveness. 
Centre for Health Services Research, University of Newcastle upon Tyne, Newcastle upon Tyne NE2 4AA

Julie Clarke, Royal College of General Practitioners and Alzheimer's Disease Society educational fellow. 
Department of Primary Care, University of Newcastle upon Tyne, Newcastle upon Tyne NE2 4HH

Moira Livingstone, Consultant in old age psychiatry. 
North Tyneside Healthcare Trust, North Shields NE29 8NH

Nick Freemantle, Senior research fellow. 
James Mason, Research fellow. 
Centre for Health Economics, University of York, York YO1 5DD

^e Competing interests: None declared.