Intended for healthcare professionals


Treating depression in later life

BMJ 2004; 329 doi: (Published 22 July 2004) Cite this as: BMJ 2004;329:181
  1. Carolyn Chew-Graham, senior lecturer in primary care,
  2. Robert Baldwin, consultant psychiatrist,
  3. Alistair Burns (A_Burns{at}, professor of old age psychiatry
  1. Rusholme Academic Unit, Rusholme Health Centre, Manchester M14 5NP
  2. Manchester Mental Health and Social Care Trust, Manchester Royal Infirmary, Manchester M13 9WL
  3. Education and Research Centre, Wythenshawe Hospital, Manchester M23 9PL

    We need to implement the evidence that exists

    The three major mental health problems affecting older people are dementia, delirium, and depression. Depressive disorders are the most common, affecting one in seven, a prevalence rate that is consistent across countries and cultures.1 Comparable rates for dementia and delirium are one in 17 and one in 25. Therapeutic optimism surrounds the management of dementia and delirium, but the development of new management approaches in late life depression has been neglected, perhaps because of stigma. Twenty years ago, a rule of three was proposed in this journal: a third of older depressed people get better, a third remain the same, and a third get worse.2 This rule is still applicable today despite the availability and efficacy of treatments that could substantially improve these ratios. Depression late in life is associated with serious morbidity and mortality, including suicide.3 The national service framework for older people targets stroke, falls, dementia, and depression, but only the first three seem to attract new funding or services commissioned by primary care trusts.

    Two thirds of older people with serious depression do not have symptoms that fit current classifications of mood disorders,1 which have been generated to reflect symptoms in younger people. Older people may have insufficient symptoms to meet the threshold for a disorder, and presentation differs from younger people because of ageing, physical illness, or both.

    Detection of depression is poor,4 and primary care providers may lack the necessary consultation skills or confidence to diagnose late life depression correctly. Although this is unproved, they may be wary of opening a Pandora's box in time limited consultations and share therapeutic nihilism with the patient.5

    The evidence base for the management of late life depression is increasing. The number needed to treat for major depression treated with antidepressants is about four and is similar to other age groups.6 Perhaps patients who are referred to old age psychiatry services do better as specialists tend to be more persistent in monitoring treatment and encouraging compliance, but 90% of older depressed people do not see a specialist.7 Older people may be distrustful of tablets, not take them, and in recent research have expressed a preference for psychological interventions.8

    Such interventions are effective in people with late life depression and anxiety, but such therapeutic options may be unavailable in primary care.9 10

    Studies have evaluated interventions such as feedback of screening results to primary care doctors, dissemination of guidelines, educational packages offered by a nurse to primary care doctors, feedback of screening test results, and recommending that antidepressants are prescribed to patients screened positive for depression.11 None of these has been shown to improve outcome in patients.11

    Innovations in the management of depression have been evaluated. The best results come from models that use multifaceted interventions and principles of collaborative care. These vary but generally are brief and include the deployment of care managers and flexible collaboration between primary and specialist care, to improve access to the psychiatrist and mental health teams for the primary care team. Important new evidence shows that this approach is very effective in older depressed patients. Unützer et al randomised 1801 depressed primary care patients aged over 60 to usual care by their primary care doctor or collaborative care.8 Collaborative care was delivered by a depression care manager (a nurse or a psychologist) under the supervision of a psychiatrist and a primary care doctor. The main components of the intervention were case management, education about depression, and management of medication. At 12 months, almost half the patients in the intervention group were at least 50% improved from baseline, compared with only one in five of those receiving usual care. In addition, patients in the collaborative care group were more satisfied, complied better with treatment, and experienced less functional impairment. The cost of the intervention was about £370 ($670; €550) per patient over 12 months. This is substantially less than a year's worth of antidementia drugs (about £1000). A similar approach was effective in minor depression and dysthymia.12

    Keeping the patient well is as important as initial treatment. Antidepressants—both tricyclics and selective serotonin reuptake inhibitors—and psychological interventions prevent relapse, and some patients may benefit particularly from combining these modalities.13 Such a need for long term management may necessitate an approach through the chronic disease model.

    New challenges arise from evidence showing that a proportion of late onset depression arises because of age associated cerebrovascular disease.7 So treatments of the future may include some not usually regarded as part of the psychiatrist's toolkit—for example, drugs to halt vascular damage. In the meantime the task for primary care must be to move away from the old model of relying on the passive prescription of pills.

    The treatment of depression in late life requires some imaginative commissioning at the interface of primary and secondary care. These arrangements should incorporate the new evidence base, favouring active care management and timely support from specialist mental health services.


    • Competing interests None declared.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    View Abstract