The diagnosis of depression in elderly patients is rarely straightforward. Older patients often deny feeling depressed and are more likely to present with sleep and/or appetite disturbance, agitation and multiple somatic complaints.¹ Symptoms such as fatigue, poor concentration and weight loss may be caused by physical illness or ageing rather than depression,² and some medical conditions, especially malignancies and endocrine disorders, may masquerade as depression.

Current classifications of mood disorders reflect symptoms in younger people and there is no consensus on how depression in the elderly should be defined. Reported prevalence rates vary enormously, from 0.4% to 35%.³

A primary care study from England and Wales has measured the prevalence of depression in people aged 65 and over using a semi-structured diagnostic interview and computerised diagnostic system (GMS-AGECAT).

The study, a stratified random subsample of the Medical Research Council Cognitive Function and Ageing Study (MRC CFAS), investigated 2,640 participants from two rural and three urban sites. Deprivation and disability were measured using the Townsend Deprivation Index and Townsend Disability Scale respectively. Patient selection was deliberately biased to include a high number of patients with cognitive problems and was subsequently corrected by backweighting.

The overall prevalence of depression (defined as an AGECAT severity level ≥ 3) was 8.7%. If patients with dementia and a secondary diagnosis of depression were included, the prevalence increased to 9.7%. Severe depression (level 4) was rare, affecting only 2.7% of participants. There was significant variation in the prevalence of depression between the five centres, varying from 3.5% in Cambridgeshire to 13.6% in Newcastle.

After adjustment for other factors, depression was found to be associated with severity of disability, but not deprivation.

Major depression, as defined by the DSM-IV, appears to be relatively rare in the elderly, with an average prevalence of 1.8%.³ Studies using the GMS-AGECAT system report much higher prevalence rates. A comparative study found a prevalence of depression of 11.4%, whereas only 0.86% of the study population fulfilled the DSM-IV criteria for major depression and 3.6% fulfilled the criteria for minor depression.⁴

There are two possible conclusions. Either the DSM-IV fails to identify a large proportion of elderly patients with depression, or depression is much less common in the elderly than generally believed and is overdiagnosed by GMS-AGECAT.

Depression in the elderly approximately doubles risk of mortality,⁵ so a healthy survivor effect might be expected, with only the sanguine living to a ripe old age.

An audit in our practice found that we were diagnosing depression in only 2.9% of our patients aged over 65 per year, compared with a point prevalence of 15% in community studies.¹ If two-thirds of older people with serious depression do not have symptoms that match the diagnostic criteria of current classification systems,⁶ perhaps we should not be using the PHQ-9 in patients over 65.