Red X iconGreen tick iconYellow tick icon

Annabel Begg 2005


Depression in later life is associated with increased morbidity, increased utilisation of both mental health and non-mental health services, and increased suicide and non-suicide mortality. The increasing proportion of older people in the New Zealand population emphasises the importance of addressing late life health issues, and the over-representation of elderly among primary care attenders points to an important opportunity for detection and management of late life depression. To date, three New Zealand studies have measured levels of depressive symptoms among patients attending General Practitioner appointments, although none have focused on the elderly.

This thesis reports a descriptive study of cross-sectional design carried out over a nine-week period in a large primary care clinic in urban Christchurch. The primary aim was to estimate the prevalence of depression among elderly attenders (aged 65 years and above) using a validated symptom scale (the 15-item Geriatric Depression Scale, or GDS15). Additional aims included investigating the levels of depressive symptomatology by sociodemographic variables, comparing General Practitioner report with GDS15 scores, and exploring the development of shortened versions of the GDS15 in the study setting.

Study interviews were carried out on site following the patients' General Practitioner appointments. Over the data collection period a total of eighty out of ninety nine eligible patients (81% response rate) consented to and completed the study interview, which included sociodemographic questions, the GDS15, and the Folstein Mini Mental State Examination (MMSE). All eighty participants consented to their General Practitioner being asked to complete a brief questionnaire asking about the participant's current mental state.

The male-to-female ratio of participants was similar to that in the practice age-sex register. The median age was 74.5 years (interquartile range 70.0 to 83.0 years) and the oldest age group (85 years and above) was under-represented. The mean GDS15 score was 2.0 (SD 2.5), with a range of 0 to 12 out of 15. Using a cut-off score of 5 or above out of 15 to designate caseness for depression, the prevalence of depression was 10% (95% confidence interval 4.4% to 18.8%). The sensitivity of General Practitioner diagnosis (compared with caseness on the GDS15 ) was 62.5% (95% CI 30.6% to 86.3%) and the specificity was 79% (95% CI 68.4% to 87.0%). All participants scored 24 or more out of 30 on the MMSE. There were no differences in GDS15 scores according to sociodemographic variables, although the small sample size permitted only large differences to be detected.

Key limitations of the study were the small sample size and reduced generalisability due to the narrow setting of the study. Strengths of the study included the proportional sampling method used, the consistency of administration of the GDS15, and the assessment of cognitive function to aid interpretation of GDS15 scores.

This study complements the existing literature on the prevalence of depression among adults of all ages attending primary care in New Zealand. The prevalence of caseness on the GDS15 was lower than that reported in overseas studies which have used this scale in the primary care setting. This may be due to any combination of: methodological differences, differences in primary care systems, and differences in the prevalence of depression.

The evidence for both negative impacts of late life depression and for the treatability of the condition support the need for further research. The methodology of the present study could translate into a larger study in primary care, residential care, or a community setting.

Back to top