General Practice

Influence of symptoms of anxiety on treatment of depression in later life in primary care: questionnaire survey

BMJ 1999; 318 doi: (Published 27 February 1999) Cite this as: BMJ 1999;318:579
  1. Michael Kirby, research registrara,
  2. Aisling Denihan, research registrar,a,
  3. Irene Bruce, research nurseaa,
  4. Alicja Radic, epidemiologistd,
  5. Davis Coakley, professorb,
  6. Brian A Lawlor, professorc
  1. aMercer's Institute for Research on Ageing, Saint James's Hospital, Dublin 8, Ireland
  2. bDepartment of Medicine for the Elderly
  3. cDepartment of Psychiatry for the Elderly
  4. dHealth Research Board, Dublin 2, Ireland
  1. Correspondence to: Dr Kirby
  • Accepted 15 September 1998

While studies of the prevalence of major depressive disorders in elderly people have produced rates of 1% to 2%, depression that is clinically significant has been shown to have a prevalence of at least 10% among older people and represents the most common mental disorder in later life. Most of these depressed older people, however, do not receive any treatment for their depression.1 Prominent symptoms and syndromes of anxiety commonly accompany late life depression in the community2 and may contribute to the low level of detection of the primary depressive disorder and to inappropriate treatment with benzodiazepines. As part of a naturalistic study of mental disorders among elderly people living in the community in Dublin3 we studied the influence of concurrent anxiety symptoms on the likelihood of them receiving pharmacological treatment for depression.

Subjects, methods, and results

People aged 65years and over on the practice lists of five urban general practices and not living in residential care were identified. We interviewed 1737participants (82%) with the geriatric mental state and automated geriatric examination for computer assisted taxonomy instrument, which generates “cases” and “subcases” of mental disorder (subcase level representing symptoms not reaching the criteria for case level disorder).The level of depression among cases has been shown to correspond with what psychiatrists usually recognise as a depressive disorder and has been validated against the combined categories of major depression and dysthymia from the Diagnostic and Statistical Manual of Mental Disorders, third edition (DSM-III), with good agreement.4 In addition to the primary diagnosis each subject is allocateda level of confidence on all (eight) diagnostic clusters and, therefore, the presence of symptomsor disorders comorbid with the principal diagnosis is recorded. Current use of psychotropic drugs was recorded by direct inspection of medications.

There were 184(11%) cases of depression among the elderly people. Of these, 84(46%) were receiving a psychotropic drug, with a similar proportion of depressed men (22/53, 42%) and women (62/131, 47%). Sixty four (35%) depressed participants were taking a benzodiazepine and 34(19%) were taking antidepressant medication. Of the 184depressed people, 36(20%) had a comorbid anxiety disorder (case level anxiety or phobia), 115(63%) had concurrent anxiety symptoms (anxiety, phobic, or obsessional symptoms at subcase level), and 33(18%) were free of anxiety. The table compares the use of psychotropic drugs in these three subgroups. The presence of concurrent anxiety in depression was significantly associated with the use of any psychotropic drug (χ2 8.0; df=2; P=0.02), a benzodiazepine (χ2 9.3; df=2; P=0.01), or an antidepressant (χ2 6.6; df=2; P=0.04).

Numbers (percentages) of elderly people who received psychotropic drugs for depression classed according to geriatric mental state and automated geriatric examination for computed assisted taxonomy questionnaire

View this table:


Unless elderly people with depression have concomitant symptoms of anxiety they are less likely to receive pharmacological treatment in primary care. To our knowledge no previous study has examined this issue. Sartorius et al showed that the presence of comorbid anxiety disorders with depression increased the chance that depression would be recognised and some treatment offered but excluded elderly patients and did not consider depression with concomitant anxiety symptoms of less severity than anxiety disorder.5

Whereas depression in later life was undertreated, depression with prominent anxiety did tend to receive some pharmacological intervention, whether appropriate treatment or otherwise. It would seem that the concurrent symptoms of anxiety rendered the depression “loud” and attracted the attention of the doctor. Having recognised this important aspect of the presentation of depression in later life we may be able to direct attempts to improve diagnosis and to alter the practice of symptomatic treatment with benzo-diazepines in a more focused way. The corollary to anxious depression being loud is, of course, that depression which is not flagged by symptoms of anxiety remains “silent” and is at particular risk of being missed.


We thank our colleagues in general practice for their cooperation.

Contributors: BL and DC initiated the project. MK and BL discussed the core issue and existing literature and were responsible for the design of the study. Data were collected and analysed by IB, AR, AD, and MK. The paper was written by MK and revised and approved by BL. MK and BL are the guarantors.

Funding: The Health Research Board contributed to the funding of this project.

Competing interests: None declared.


View Abstract