Gender Differences in Depressive Symptoms in Kumasi, Ghana
Embargo until
Date
2013-09-02
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Johns Hopkins University
Abstract
Background: Depression is an important public health issue globally, and accounts for an increasingly large proportion of disease burden in low-income countries. Depression is twice as common in women as men. Despite the public health importance, very little research has been done on depression in low-income countries, especially outside the post-partum period. This dissertation uses a mixed methods approach to assess gender differences in depressive symptoms in a peri-urban area near Kumasi, Ghana.
Methods: This research was embedded within the Family Health and Wealth Study, an ongoing cohort study. Results and analyses are presented in three manuscripts. The first manuscript addresses the results of focus group discussions with FHWS participants and in-depth interviews with community care providers. Qualitative data analysis included reading, coding, displaying data, and data reduction. The second manuscript contains the results of mixed methods analyses to assess the internal consistency reliability, face validity, content validity, and construct validity of a ten-item version of the Center for Epidemiologic Studies Depression (CES-D) Scale. In the third manuscript, we present the results of a series of Tobit regression models identifying factors associated with depressive symptoms for men and women, and statistical differences between groups.
Results: Our qualitative data analyses reveal a possible conceptual model for depression in the study community, including description of potential causes, consequences, and treatments for depression. Common causes described for men and women differ, but both are rooted in a combination of family/relationship and economic factors. Our quantitative results indicate adequate reliability and validity of the ten-item CES-D Scale in the study community. The addition of locally-relevant symptom groups would improve the scale’s validity. The following characteristics were associated with more depressive symptoms in combined regression models for men and women: having a larger education difference with one’s partner, higher commitment, lower communication, and partner’s higher CES-D score. There were significant interaction effects between sex and relative SES and self-rated health.
Conclusions: The results of this dissertation provide information on the experiences of depressive symptoms in peri-urban Ghana. The findings can be used to inform future research and interventions to address depression in sub-Saharan Africa.
Description
Keywords
Mental health: Women's health