Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



Research Advisor

Mona Newsome Wicks, Ph.D.


Jerome L. Blakemore, PhD Ann K. Cashion, PhD. Patricia D. Cunningham, DNSc Elizabeth Tolley, PhD


Anxiety, depression, African American, HIV-positive women, inherited vulnerabilities, acquired vulnerabilities


This descriptive correlation study had four purposes: describe depressive and anxiety symptom levels in pregnant and non-pregnant African American (AA) HIV-positive women, examine the association between underlying vulnerabilities and depressive symptoms in HIV-positive AA women, determine the association between underlying vulnerabilities and anxiety symptoms in AA HIV-positive women, and determine the predictors of depressive and anxiety symptoms among African American HIV-positive women. A convenience sample of 80 African American HIV-positive women was recruited from university-based obstetrics/-gynecology and infectious disease practices in a large metropolitan city in the Mid-South. Depressive symptoms were measured with the Center for Epidemiologic Studies Depression Scale (CES-D) and Edinburgh Postnatal depression Scale (EPDS). Anxiety symptoms were measured with the Beck Anxiety Inventory (BAI). Sleep quality was measured with the Pittsburg Sleep Quality Index (PSQI). An investigator-developed demographic data form was used to obtain patient characteristics including data about inherited and acquired vulnerabilities. Inherited vulnerabilities included family history of mood disorders and family history of alcohol and/or drug use. Acquired vulnerabilities included educational level, income level, alcohol and/or use, HIV disclosure status, sleep quality, and history of sexual assault. The stress diathesis theory provided the conceptual basis for the study. The sample consisted of both pregnant (n = 20) and non-pregnant (n = 60) women. Participants were typically single (57.5%), middle-aged (33.21 ± 10.7 years), and had completed high school (30%) or some college (35%). Poverty was prevalent in the sample, with 70.5% reporting annual household incomes below $10,000. Most women (92.5%) had disclosed their HIV status. Almost half of the sample (42%) reported a history of sexual assault. Substance abuse was higher in the non-pregnant group of women (40%) than the non-pregnant group (5%) of women. Median depressive symptom scores for the total sample were mildly elevated. Median anxiety symptom scores for the total sample illustrated little or no anxiety symptoms. Poor sleep quality was prevalent in both groups and was associated with elevated CES-D, EPDS, and BAI scores in both groups of women. Family history of mood disorders was associated with elevated CES-D scores in the group of pregnant women, and elevated BAI scores in both groups. HIV disclosure was associated with elevated CES-D scores in the pregnant group, and substance use with elevated CES-D scores, in the non-pregnant group. Non-pregnant HIV-positive women had higher depressive symptom scores on both the CES-D and EPDS when compared to pregnant HIV-positive women. Regression analyses were performed to determine which independent variables predicted elevated depressive and anxiety symptom scores. Substance use and poor sleep quality were statistically significant predictors, accounting for 55 % of the variance in CES-D scores, 43% of the variance in EPDS scores, and 53% of the variance in BAI scores. Future studies should include testing interventions that improve sleep quality. Untreated mood disorders are problematic in HIV-positive patients, as depression is associated with decreased adherence to highly active antiretroviral medications. Ongoing assessments of depressive and anxiety symptoms are necessary, as clinically significant levels resulting in referrals were warranted.