Korten & Henderson (Reference Korten and Henderson2000) described findings from a 1997 national survey, in which a “household sample of 10 641 individuals representative of the adult population of Australia” was interviewed. They reported that “the prevalence of a diagnosis of any ICD-10 anxiety or depressive disorder… declines for both men and women after the age of 55 years”, and noted a trend for psychological symptoms to be fewer among the older age groups. Before accepting the findings as evidence that depression is less prevalent in old age (a conclusion that might affect decisions about allocation of resources), the following points should be noted (see Reference Snowdon, Draper and ChiuSnowdon et al, 1998).

First, the (approximately) 1600 subjects aged over 65 years were not truly representative of the older population. The survey excluded the 10% of older Australians who were temporarily or permanently residing in institutes (e.g. hospitals, nursing homes, boarding houses), or homeless at the time of the survey. It also excluded those with moderate or severe dementia (Mini-Mental State Examination score <18). The prevalence of depression is considerably higher among those with physical disability, those in residential care and those with dementia.

Second, the instrument forming the core of the interview was the automated version of the Composite International Diagnostic Interview (CIDI), which discounts symptoms attributable to physical illness (Reference JormJorm, 2000). Studies that rely on subject-reported symptoms may underestimate the severity of depression in old age, since older patients with depression are less likely than younger patients to acknowledge having affective symptoms (Reference Lyness, Cox and CurryLyness et al, 1995).

Third, the response rate in this survey was 78%, but the response rate of different age-groups was not known. In other surveys (e.g. Reference Kramer, German and AnthonyKramer et al, 1985), older subjects have been twice as likely as younger adults to decline involvement. Refusers are more likely to be depressed.

Finally, the report did not differentiate prevalence rates in ‘young-old’ and ‘old-old’ individuals, yet various researchers have found a progressive increase in rate from 55 to 85 years. Jorm (Reference Jorm2000) commented on the lack of consistency between researchers regarding whether or not depression becomes less prevalent in old age.