Details

The Recognition and Determinants of Depression at a South African Primary Care Clinic

by Triant, Virginia Athena

Abstract (Summary)
Depression leads to substantial morbidity and mortality on a global scale, but it is frequently underrecognized and inadequately treated in primary care settings. The detection of depression is particularly challenging in under-resourced countries. This study attempts to determine rates of recognition and treatment of depressive disorders at a community clinic in the Cape Flats, South Africa. The diagnostic instrument PRIME-MD was administered and charts reviewed for a sample of 222 patients presenting to the Lotus River Community Health Centre. Outcomes sought were (1) the prevalence of depressive disorders and (2) rates of detection and treatment as indicated by antidepressant prescription through chart review. The prevalence of depressive disorders in the group was found to be 32% (N=70), with 13% meeting criteria for major depressive disorder (N=29). Depressed patients tended to be younger (p<.001) and female (p=.026) and were more likely to describe somatic symptoms than were non depressed (p<.001). There was a statistically significant correlation between a diagnosis of depression and prescription of a tricyclic antidepressant (TCA) (p=.028). The sensitivity and specificity of a TCA prescription for depression were 20% and 91%, respectively. Multiple logistic regression analysis demonstrated that the strongest factor contributing to either a diagnosis of depression or a prescription of an antidepressant was the presence of somatic complaints. In this primary care setting, patients with depressive diagnoses were reliably recognized as indicated by a significant correlation between depression and antidepressant receipt. Physicians appeared to respond primarily to somatic rather than psychological presentations. While depressed patients received antidepressants more frequently than non-depressed, only a minority of depressed patients was recognized as such. Analyzing patterns of recognition requires an understanding not only of physician practices but also of the cultural setting of the health care system.
Bibliographical Information:

Advisor:Jeffrey Levine MD

School:Yale University

School Location:USA - Connecticut

Source Type:Master's Thesis

Keywords:multivariate analysis

ISBN:

Date of Publication:02/24/2003

© 2009 OpenThesis.org. All Rights Reserved.