Intended for healthcare professionals


Drugs for Alzheimer's disease

BMJ 2001; 323 doi: (Published 21 July 2001) Cite this as: BMJ 2001;323:123

Cholinesterase inhibitors have passed NICE's hurdle

  1. John T O'Brien (j.t.o'brien{at}, professor of old age psychiatry,
  2. Clive G Ballard (c.g.ballard{at}, professor of old age psychiatry
  1. Institute for the Health of the Elderly, Wolfson Research Centre, Newcastle General Hospital, Newcastle upon Tyne NE4 6BE

    Alzheimer's disease, the commonest cause of dementia in older people, affects 4% of the over 65s and 20% of the over 80s, 1 2 with around 400 000 sufferers in the United Kingdom.2 The prevalence of the condition will double over the next 50 years. As well as causing immense distress to patients, their carers, and families, dementia is estimated by the Audit Commission to cost the United Kingdom £6.1bn a year (at 1998-9 prices), with £3.3bn of this direct spending by health and social services.3 The National Institute for Clinical Excellence (NICE) has reviewed the available drugs for Alzheimer's disease and declared them clinically effective in reducing the burden of disease in some patients.

    The aetiology of Alzheimer's disease remains unknown, and no treatments reverse or stabilise the disease. Current management focuses on establishing an accurate clinical diagnosis, ensuring appropriate services are provided, supporting carers, and treating associated non-cognitive problems. Within the past three years three cholinesterase inhibitors (donepezil, rivastigmine, and galantamine) have been licensed in the United Kingdom for use in mild to moderate Alzheimer's disease. These drugs are a rational therapy based on the core deficit in the disorder, that of cholinergic deficit.4 These compounds represent symptomatic treatments and have been shown in several large, multicentre, randomised, double blind, placebo controlled trials to improve cognitive function, global outcome, and activities of daily living.5-9 There is also accumulating evidence that they may improve non-cognitive symptoms such as psychosis and apathy.9 The mean effect of drug over placebo represents an improvement in cognition roughly equivalent to stemming 6-12 months of natural decline in untreated patients. When the drug is withdrawn the clinical gain is reversed, and there are no convincing clinical data that these drugs modify the disease.

    Group means hide a marked heterogeneity of response as 40-50% of patients show a definite clinical improvement (4 points on the Alzheimer's disease assessment scale-cognitive subscale (ADAS-cog)), while 20% show a strong response (7 points on the ADAS-cog, equivalent to stemming a year or more of natural cognitive decline). Responders are maintained above baseline for 12-18 months on both cognitive and non-cognitive measures.10 Numbers needed to treat for significant clinical improvement are 3-7.11 No reliable predictors of response have emerged, and in each patient careful assessment of benefit needs to be undertaken after two to four months of treatment. Both efficacy and side effects (mainly gastrointestinal problems) are similar between compounds.

    Since the licensing of the first drug (donepezil) in the United Kingdom in 1997 inconsistent availability within clinical practice has resulted in geographical inequities in availability. Initial scepticism over these compounds was fuelled by the late publication of key trial data, initial lack of clear effects on activities of daily living scores, and difficulties in determining cost effectiveness. The latter remains a problem as economic analyses have been forced to use short term trial data to predict long term outcome (for example, delay in admission to institutional care). However, most economic analyses clearly show either cost neutrality or cost effectiveness in favour of these agents.12 Though more data are needed, the evidence to date suggests that the modest cost of these agents (£800-£1200 a year) would be more than offset by savings elsewhere, though not necessarily within the healthcare system.

    Summary of NICE guidance on antidementia drug use

    • All three drugs (donepezil, rivastigmine, and galantamine) should be available in the NHS for those with mild and moderate Alzheimer's disease

    • The diagnosis must be made by a specialist according to standard diagnostic criteria

    • Cognition, global and behavioural functioning, and activities of daily living must be assessed before prescription, which is limited to secondary care. Cognitive function (mini-mental state examination (MMSE)) score must be >12

    • Compliance must be assured

    • Drug should be continued only after assessment at 2-4 months showing: improvement or no deterioration in MMSE score; evidence of global improvement on behavioural or functional assessment

    • Patients to be reviewed every 6 months and treatment continued only while MMSE score remains >12

    • Drug costs may be about £42m/year, which may be offset by delay into residential care

    • Specialised secondary care services need expanding, particularly memory clinics

    • Guidance No 19, January 2001.

    The National Institute for Clinical Excellence's recent guidance recommends when and how cholinesterase inhibitors can be prescribed (see box). These guidelines are sensible and will facilitate equitable availability for patients with mild to moderate Alzheimer's disease across the United Kingdom while providing structure for clinical practice which requires standardised monitoring of progress and discontinuation of treatment in the absence of benefit. Nevertheless, it may be hard to observe the guidance's recommendation not to prescribe these agents to patients when their mini-mental state examination score falls below 12—that is, when they enter a more severe stage of illness.

    Preliminary evidence indicates that these agents may have value in other dementias, such as dementia with Lewy bodies, and for people with severe dementia and psychiatric and behavioural problems, though the evidence is as yet inadequate for a clear treatment recommendation. We support the institute's view that these are priority areas for further research, as are rigorous studies to determine whether cholinesterase inhibitors can modify the course of disease. However, the wider availability of these drugs will make research into the latter difficult if not impossible.

    The wider availability of these treatments will probably have other benefits. Early referral of patients will no longer be seen as pointless, and this may lead to a paradigm shift from care towards treatment. The guidance recognises this by emphasising that secondary care services, particularly memory clinics, need to be developed, though expansion of services needs to be in line with evidence of effectiveness. Although only the first step in the development of treatment for dementia, these drugs have led to a new mood of optimism about the management and future prospects for this devastating illness.


    • JOB and CB were external experts to the NICE appraisal determination and both have accepted hospitality and honoraria for lectures from Novartis, Pfizer, Shire, and Janssen. JOB is on advisory boards for Janssen and Pfizer, CB for Novartis and Shire.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    View Abstract