Anxiety and Depression Among Women Living with HIV: Prevalence and Correlations

Introduction: It has been found that HIV positive women are becoming increasingly affected by various illnesses, including Common Mental Disorders (CMDs) such as depression. Such comorbidity escalates the disease progression to the severe stage and commonly hinders treatment adherence. This study determined the prevalence of anxiety and depression amidst women living with HIV. Methods: Based on a cross-sectional and facility-based study, 357 HIV positive women were recruited using the systematic sampling technique from two public hospitals in Jijiga town, Ethiopia. The Hospital Anxiety and Depression Scale (HADS) was administered for screening, and followed by a pre-tested questionnaire that comprised of Perceived Social Support and HIV stigma. Results: The results revealed that the prevalence of both anxiety and depression amidst HIV positive women was 28.9% and 32.5%, respectively. In the multivariate analysis, it was discovered that lack of formal education, being divorced, unemployed, and earning a monthly income less than 1400 ETB (37.5 USD) were significantly associated with depression. Women with symptomatic HIV clinical stage III (AOR =2.06, 95% C.I (0.75-5.61), with CD4 cell count below 250 (AOR = 1.14, 95% C.I (0.57-2.28), and with co-infections (AOR= 1.04, 95% C.I (0.40-2.71) also suffered from depression. Conclusion: The study outcomes show that the prevalence of depression in women with HIV was 32.5%, but they were more likely to be depressed if they were illiterate, divorced, unemployed or had a financial burden. In addition, HIV positive women with less CD4 cell count and in the final clinical stage or suffered from a co-infection were also associated with depressive symptoms. This signifies the public health implications of psychological and cognitive morbidities of the illness among these women with chronic illnesses. Hence, future mental health interventions and HIV care should be integrated with substantial emphasis given to vulnerable groups, including HIV positive women.


INTRODUCTION
Anxiety and depression are the two most common CMDs that plague society. Depression is characterised with altered mood, low energy, loss of pleasure, and lack of interest that may lead to poor concentration, low self-esteem, poor appetite, Depression alone is estimated to represent about 15% of the total global disease burden in 2020 [2].
The recurrence rates of anxiety and depression are two to four-fold amongst HIV positive patients than the rest of the population [3]. These two CMDs, which are a long-standing challenge in the public health sector, are usually unnoticed and remain a potentially risky condition that can affect not only the social aspect of their quality of life, social interactions, and adherence to treatment [4 -6], but also their life expectancy [7].
Besides, HIV positive patients with comorbid diseases, such as depression or anxiety, are more likely to acquire viral resistance, and as a consequence, develop poor health outcomes due to low treatment adherence, especially decreased CD4 level [8 -12]. Advanced and fatal trajectory has been reported when depression and anxiety occur as comorbid diagnosed in a chronic patient [13].
This may result in substantial public health risk, which raises the need for further studies and effective prevention strategies so as to ascertain better health outcomes and improved quality of life for patients in Sub-Saharan Africa diagnosed with HIV.
The burden of this comorbidity shows the high prevalence of these CMDs among HIV patients than the general population [14]. The magnitude of the problem is amplified when HIV is widespread among those with high vulnerability, stigmatised, and marginalised, such as women, racial or ethnic minority, people with disabilities, or those with a higher level of depression than the rest [15].
So far, depression is highly prevalent in Ethiopia, and it is considered as one of the common psychiatric disorders that affect women as its associates are gender-specific. Some studies conducted in Ethiopia revealed a high prevalence of depression amidst AIDS/HIV patients, whereby 38.94% of the studied participants were depressed [16].
Women diagnosed with HIV have been substantially investigated to address worldwide HIV transmission issues within the biomedical domain. Nevertheless, studies regarding the psychosocial well-being of HIV positive women across developing and developed nations are in scarcity [17].
In developing countries, depression is a serious barrier that must be addressed when discussing HIV prevention as the two diseases are fundamentally linked and aggravate each other. Some studies reported that depressive symptoms and depression reflect the poor adherence to HIV treatment, thus adversely affecting the desired clinical results [18].
Despite the long-reckoned importance of screening depression and managing HIV positive patients across healthcare facilities, their prevalence and risk factors on women have remained untapped. As such, this study exhaustively assessed the prevalence and correlations of anxiety and depression amongst HIV positive women.

METHODS
This study involved two hospitals based on their comprehensive service provision. In accordance with the national HCT guideline, secondary-and tertiary-level healthcare facilities offer minimum curative and preventive packages. Hence, healthcare facilities were purposively selected for their key service delivery and availability of chronic patients. As such, a cross-sectional study was conducted in antiretroviral treatment clinics of two hospitals located in Jijiga town of eastern Ethiopia in June 2019. The study population comprised of women aged 18 years and above attending HIV treatment service at the two selected hospitals. To estimate the required sample size, a single population proportion formula had been applied. The prevalence of depression (P) from a baseline study in Ethiopia was 38.9% with 5% significant level, Q = 0.05,/2 = 1.96, and 5.0% absolute precision or margin of error tolerance (d = 0.05). The Epi Info software suggested 357 women as the study sample size. The systematic sampling method was applied to select the participants for this study. To determine the sampling interval, those with appointments during the data collection period were divided by the total sample size, wherein the initial point was selected in a random manner.
A trained nurse conducted the interview with women attending the antiretroviral treatment centres using a pretested structured questionnaire. Apart from the socio-demographic profile, their clinical and laboratory investigation results were retrieved as well. Women participants were screened with the HADS tool. This tool is composed of seven questions each for anxiety and depression. Although these 14 questions are interspersed in the questionnaire, they must be scored in a separate manner. A score of 8-10 signifies mild depression, while 11-14 is moderate, and 15-21 indicates severe depression. Besides, the PSS-HIV scale was used to assess the level of perceived support in women with HIV. This 12-item scale has examined women's self-esteem, self-development and belongingness. Moreover, the HIV stigma scale was used to measure the extent of stigma by examining the negative selfimage, concerns about public attitude, disclosure concerns and personalized stigma.
The gathered data were analysed using IBM SPSS statistics version 22. Multivariable logistic regression was performed by using forward stepwise logistic regression to identify the independent variables linked with the outcome variable. Variables with p-value 0.005 or less were incorporated into the model. Odds Ratio (OR) and Adjusted Odds Ratio (AOR), along with their Confidence Interval (95%CI), were calculated to identify the strength of the correlations.
Ethical approval of the study was obtained from the International Islamic University of Malaysia Research Ethics Committee (IREC). A permission letter was also obtained from the University's board of review. Written informed consent was collected from the participants after explaining the study aims and other information relevant to this study. The confidentiality and privacy of the participants had been strictly maintained.

Socio-demographic Characteristics
In total, 357 participants were involved in this study and the mean value of their age was 34.2 years (SD=8.8 years). Most of the participants were urban dwellers (n=332, 93%) and derived from the Amhara ethnicity (n=130, 36.4%). Among the total participants, 134 (37.5%) were married, and 165 (46.2%) belonged to Islam. Around half of the participants (46.2%) did not attend any formal education, while 129 (36.1%) were unemployed, and 173 (48.5%) reported that their family earned less than 1400 Ethiopian Birr or 37.5 US dollars based on the Ethiopian poverty assessment. Besides, most of them were nulliparous 173 (48.5%). Table 1 tabulates the demographic profile of the participants.

Clinical Characteristics of the Study Participants
The clinical characteristics of the participants were assessed, including clinical staging level, viral load and CD4 level, as well as the comorbidity of the common opportunistic infections.
Out of the 357 participants interviewed, 127 (35.6%) of them were in WHO clinical stage I, and followed by 93 (26.1%) at clinical stage III. As for virological examination, 151 (42.3%) of the participants had a CD4 count above 475 cell/l, and followed by 111 (31.1%) with a CD4 count ranging from 251-475 cell/l. A total of 175 respondents (49.0) complained suffering from at least a common opportunistic infection.

Depression and Anxiety Prevalence
The HADS was used to screen the study participants. The prevalence of depression and anxiety of the participants was 32.5% and 28.9%, respectively, while comorbid prevalence was 22.7%. The scores of anxiety and depression revealed that 18.2% of the participants had symptoms of severe depression, 12.0% experienced moderate depression, and 1.7% suffered from mild depression.
In HADS, a higher rating denotes a higher level of anxiety and depression. For the scales of anxiety and depression, scores under 7 suggest that the participants were not depressed, scores 8-10 reflect mild cases, 11-14 are moderate, and 15-21 display severe cases.

Factors of Depression and Anxiety Among HIV Positive Women
Binary logistic regression was employed in this study to identify the correlations of socio-demographic profiles, as well as clinical and virologic factors, with anxiety and depression.
In order to control the influence of a potential third variable, and to avoid non-causal association due to this confounding factor, a step-wise logistic regression was carried out, and the adjusted multivariable was assessed by including factors that were significant in the binary regression analysis.
The outcomes revealed that marital status, education level, occupation, and monthly income of the participants were significantly related to both anxiety and depression. Other socio-demographic variables, such as the age of the participants, residence, ethnicity, faith, and the number of living children, had no association with anxiety and depression.
The clinical characteristics of the participants, such as HIV clinical staging, number of CD4 cell counts, and presence of opportunistic infections, were significantly linked with depression and anxiety, as presented in Tables 2 and 3, respectively.
The findings displayed a higher prevalence of anxiety (AOR = 4.09, CI (1.56-10.8) and depression (AOR =2. 17,) among unemployed women and they were significantly more prevalent among those who earned less than 1400 ETB a month.
Spearman's coefficient test was used to determine the correlation among perceived social support, HIV stigma, and depression.  A significantly negative correlation was found between perceived social support and depression (r = r = -0.11 p = 0.037) with BCa 95% confidence interval [-0.21 to -0.02] and anxiety (r = -0.15 p = 0.005) with BCa 95% confidence interval [-0.26 to -0.04] among the participants.
The lower score of perceived social support among the participants suggested a higher level of anxiety and depression.
Meanwhile, a positive correlation was found between HIV stigma, and depression (r = 0.37 p < 0.001) with BCa 95% confidence interval [0.26 to 0.46] and anxiety (r =0.36 p<0.01) with BCa 95% confidence interval [0.24 to 0.46]. This showed that higher scores of HIV stigma among the study participants denoted a higher level of anxiety and depression, as tabulated in Table 3.

DISCUSSION
This study had assessed the prevalence and correlations of depression and anxiety among HIV positive women from two selected hospitals in eastern Ethiopia. The work is part of an extended mixed-method study with an emphasis on women living with HIV. This study verifies the results of past research work that reported a higher prevalence in women than men [19 -23]. In this study, the symptoms of depression were measured using the HADS screening tool. The other scales used in this study were the PSS-HIV scale to assess social support and HIV stigma. As a result, the prevalence rate of depression among the participants was 32.5%. This is lower when compared to other studies from Sub-Saharan Africa, such as 48% in Kenya [24] and 48.7% in South Africa [25]. This is ascribed to the difference in the study population recruited in these studies, which comprised of women in postpartum/pregnancy with a period of high susceptibility to depression. However, the prevalence of this study is higher when compared to other studies also conducted in developing countries, such as 11% in Malawi [21].
A number of reasons can be attributed to the prevalence of depression in these countries. Firstly, there is a difference in the screening tools used. For instance, studies in Malawi [21], Kenya [24], and South Africa [25] used the EPDS screening tool, whereas a study in Uganda applied the CES-D for screening. The other reasons refer to the methodological variations and the sample size difference.
The comorbid prevalence of anxiety and depression was 22.5%. According to Dell'Osso and Pini [26], the possible clarifications for these comorbid mental illnesses might be that both disorders are likely to share the same phenomena or common symptoms or vulnerabilities. Upon close examination, the findings revealed that the point prevalence of these comorbid illnesses is marginally higher than the institutionalbased cross-sectional study conducted on pregnant women in Tanzania (18.1%) [27].
A number of factors have been identified related to the development of anxiety and depression among women living with HIV. For instance, in this study, divorcees were found to be more likely to develop both depression and anxiety. Similar findings have been reported in studies conducted in Ethiopia [28] and Uganda [29], thus revealing that women who were not living with their partner or divorced were at higher risk of acquiring depression. A study conducted in Ukraine by Bailey [30] discovered that non-cohabiting and single mothers were more likely to be depressed. This justifies that broken-up families or insecure marital engagements intensify the risk of developing depressive symptoms. This is commonly observed among abandoned partner, who usually suffers from psychological and emotional disturbances when they miss their loved one(s) or when feeling the burden of family-related workload alone with no or less support.
It was revealed that women who were illiterate, unemployed, and whose monthly household income was less than 1400 Ethiopian Birr were at higher risk for anxiety and depression. The results are consistent with those reported in a study that involved South African HIV positive women, which presented that low-income and unemployment were related to depression [25,31]. The reason could be that in low-income countries, women are pressured to default academics for poverty-related factors, which later result in their more prominent engagement on domestic work, as well as the lack of access to health education and awareness. This is ascribed to the possible negative interaction between mental disorders (e.g., depression) and poverty, primarily because, in principle, people with depression commonly perform poorly in their daily tasks [32]. Moreover, a number of studies carried out in highincome countries revealed that people with social and economic disadvantages are more likely to develop depression [33], indicating the substantial relationships between social adversity, HIV, and depression. Based on the world federation on mental health, women tend to suffer from depression twofold more than men due to financial issues, lack of formal education, and potential exposure to violence [34].
This study found that women with symptomatic HIV clinical stage IV for anxiety and clinical stage III for depression, CD4 cell count below 250 cells/mm, and opportunistic infection were more likely to suffer from anxiety and depression. Similarly, other studies conducted in the Russian Federation [35], the USA [36], China [9], and Ethiopia [37] reported that a decrease in Cd4 cell counts and an increase in viral load were significantly related to both depression and anxiety. This is attributed to the effect of depression on HIV disease progression by suppressing the immune system, and this immunosuppression, may, in turn, be a predictor for the development of depression among HIV patients [10,38]. Additionally, the growing evidence suggests the significant role of immunological factors, such as CD4 cells, on the pathophysiology of depression development, particularly among patients with chronic illnesses.
Regarding women's expected support, a significantly negative correlation was reported between perceived social support and depression. Similarly, a study on HIV positive Malawian women indicated that depressive symptoms were associated with decreased perceived social support [39]. In line with this, numerous studies conducted in Sub-Saharan Africa have reported the significant relationship between depressive symptoms and social support [40]. This manifests that social support represents the scope of one's level of satisfaction in relation to the necessity of support, information, and interaction, apart from being a key element for coping with depression. Brittain et al. [41] asserted that the substantial moderator between social support and depression was the stigma perceived by women with HIV.

CONCLUSION
This study found a positive correlation between HIV stigma and depression, which is consistent with previous findings (Benoit et al., 2014;Gonzalez et al., 2012;Logie et al., 2013). Apprehension and depression are attributable to one's expected fear of social acceptance. This emotional disturbance may be related to feelings of not receiving the deserved public attention due to their illness or their depressive circumstances.
Hence, as this study is a cross-sectional one, the possible limitation is the lack of possibility to examine the causal association between social support and stigma with the dependent variable; depression. The other possible limitation is, the interview was not carried out by a psychiatrist who could sub-standardize the quality of the data collection. However, the findings suggest the need for further longitudinal studies in the future by taking into account several potential risk factors.
In conclusion, this study reveals that depression and anxiety are indeed prevalent among women living with HIV. These gender-specific mental health gaps need to be addressed in order to develop effective strategies that enhance the integration of mental health in the HIV care setup.

LIST OF ABBREVIATIONS AIDS
= Acquired Immune Deficiency Syndrome

ETHICS APPROVAL AND CONSENT TO PARTI-CIPATE
Ethical approval of this study was obtained from the International Islamic University of Malaysia Research Ethics Committee (IREC). A permission letter was also obtained from the Jijiga University board of review.

HUMAN AND ANIMAL RIGHTS
Not applicable.

CONSENT FOR PUBLICATION
Written informed consent was attained from each study participant after clarifying the purpose of the study. The privacy and confidentiality of the participants had been strictly maintained.

AVAILABILITY OF DATA AND MATERIALS
The data that support the findings of this research are available from the corresponding author [G.T] upon request.

FUNDING
This study did not receive any funding from the public or private sectors.