Generalized psychological distress among HIV-infected patients enrolled in antiretroviral treatment in Dilla University Hospital, Gedeo zone, Ethiopia

Background Psychological disorders like depression and anxiety are potentially dangerous conditions. In the context of HIV/AIDS, this can influence health-seeking behavior or uptake of diagnosis and treatment for HIV/AIDS, add to the burden of disease for HIV patients, create difficulty in adherence to treatment, and increase the risk of mortality and morbidity. The objective of this study was to assess the prevalence and correlates of generalized psychological distress among HIV-infected subjects on antiretroviral treatment (ART). Design An institution-based cross-sectional study was conducted. Interviews were conducted with 500 patients initiating ART at Dilla Referral Hospital. Generalized psychological distress was measured using the Hospital Anxiety and Depression Scale (HADS). A cutoff score ≥19 was used to identify possible cases of patients with generalized psychological distress. Multivariable logistic regression analysis using SPSS Version 20 was performed to identify factors associated with psychological distress. Results The prevalence of generalized psychological distress among the population of this study was 11.2% (HADS≥19). Factors independently associated with generalized psychological distress were moderate stress (OR=6.87, 95% CI 2.27–20.81), low social support (OR=10.17, 95% CI 2.85–36.29), number of negative life events of six and above (OR=3.99, 95% CI 1.77–8.99), not disclosing HIV status (OR=5.24, 95% CI 1.33–20.62), and CD4 cell count of <200 cells/mm3 (OR=1.98, 95% CI 0.45–0.83) and 200–499 cells/mm3 (OR=3.53, 95% CI 1.62–7.73). Conclusions This study provides prevalence of psychological distress lower than the prevalence of common mental disorders in Ethiopia and comparable to some other studies in sub-Saharan Africa. The findings are important in terms of their relevance to identifying high-risk groups for generalized psychological distress and preventing distress through integrating mental health services with HIV/AIDS care and support program.

depression and anxiety. For example, the prevalence of depression among HIV/AIDS patients ranges from 12% in south India (8) to 54.4% in Italy (7). In Brazil, it is 29.4% (5), and in the United States it is 37% (2). In African countries, it is also high; for example, in South Africa it is 25.4% (6), in Uganda it is 8.1% (13), and in Botswana it is 28% (10).
Psychological disorders like depression and anxiety are potentially dangerous conditions, in the context of HIV/ AIDS, which can influence health-seeking behavior or uptake of diagnosis and treatment for HIV/AIDS (1). They add to the burden of disease for HIV patients regarding difficulty in adherence to treatment (14); increased risk of mortality and morbidity (2Á5); reduced productivity; inconsistent use of condoms; declines in CD4 cell count that lead to rapid progression to AIDS and death; and increased risk of heart diseases (10).
Studies conducted in the general population in Ethiopia show that the situation is worse. There are 1.3 million adults living with HIV/AIDS in the country (15). Only one study shows the magnitude of common mental disorders in TB/HIV coinfected people (16). Published reports show that residence, educational status, gender, marital status, income level, behavior, psychosocial status, social status, and HIV-related clinical variables are important risk factors for depression and anxiety (4,6,7,10,13,17,18). However, previous researches do not*in one and the same study*simultaneously incorporate all correlates of psychological distress in people receiving antiretroviral treatment (ART). This study fills this particular research gap by simultaneously assessing the impact of sociodemographic factors, psychosocial, and HIV-related clinical factors on psychological distress in a population of public sector among patients receiving ART in Ethiopia.

Methodology
Study setting Dilla University Referral Hospital is found in Dilla City administration, which is located 360 kilometers from the capital city, Addis Ababa, in the south of Ethiopia. It is the public hospital that is an affiliate of Dilla University and provides training for health sciences students in a range of disciplines. Additionally, the hospital provides a higher level of clinical care for nearly a million of the catchment area populations. Since 2005, the hospital has been providing highly active antiretroviral therapy (HAART) for people living with HIV (PLHA). During the study period (January 2013 to November 2013), 4,091 subjects were enrolled in chronic HIV/AIDS care and 2,359 patients were on HAART. According to the national guideline, ART shall be initiated for eligible patients. The eligibility of the patients is determined if their CD4 cell count is B200/mm 3 or if they fulfill WHO clinical AIDS stage III or IV.

Study population and study design
This facility-based, cross-sectional study was undertaken at Dilla University Referral Hospital in Dilla town, Ethiopia. Five hundred HIV-seropositive individuals aged 18 years and above on ART were included in this study. From 2,359 registered patients, 500 study subjects were selected by systematic sampling techniques with a k factor of 5 from the ART registration database. The study subjects were drawn by a systematic sampling technique from ART registration database from January 2013 to November 2013. Subjects were excluded if the following conditions were found: (1) significant chronic systemic illness; (2) significant neurologic disorders, including traumatic brain injury; and (3) a history of schizophrenia or severe psychotic disorder.

Data collection tools
The data collection tools contain various semistructured questionnaires prepared in English but translated into Amharic (a lingua franca used for interethnic communication in Ethiopia). The tools were originally developed by the West. The tools were administered by a trained nurse using face-to-face interview techniques. These include the following:  (19). Response categories ranged from one (strongly disagree) to five (strongly agree). Summative composite scores were created for perceived stigma (range 08Á40). The scores were scaled in the positive direction (the higher the score, the higher the stigma); overall stigma scores were categorized into three categories such as no or mild, moderate, and severe stigma using the 33rd and 66th percentile cutoff values from the distribution of scores. The scale reported a Cronbach's alpha of 0.90 in this study. ii. Perceived stress scale: the 10-item Perceived Stress Scale (PSS) (20) was used to assess the degree to which situations in life are perceived as stressful (reliability in this sample, Cronbach's alpha 00.78). The possible response ranges from zero (never) to four (very often). PSS scores were obtained by reversing responses to the four positively stated items and then summing across all items. Individual scores on the PSS can range from 0 to 40 with higher scores indicating higher perceived stress; the overall scores were divided into three groups: scores ranging from 0Á13 were considered low stress; those from 14Á26, moderate stress; and those from 27Á40, high stress. High perceived stress category was excluded from the analysis because none of the respondents had high perceived stress. iii. Negative life events score index: 20 life events specifically related to being HIV-positive were constructed for this study. Respondents were asked to indicate whether they experienced such an event on a dichotomous scale (yes/no). The items considered for the negative life events in this study were bereavement, severe illness, and severe interpersonal conflict in significant social relationships, such as parent, sibling, spouse/ lover, and child/children. Individual-related negative life events included severe sickness; interpersonal conflict; feelings of isolation and abandonment; lack of the basic requirements of food, shelter, and medicine; job loss; discrimination; and worries about personal finances. A total score was generated to reflect the total number of life events reported, and this scale had a Cronbach's alpha of 0.71 in the present study. iv. Social support index: constructed from 19 functional support items (21) and used to measure five dimensions of social support: (1) emotional support (the expression of positive, affective, empathetic understanding, and the encouragement of expressions of feelings), (2) informational support (the offering of advice, information, guidance, or feedback), (3) tangible support (the provision of material aid or behavioral assistance), (4) positive social interaction (the availability of other persons to share enjoyable activities), and (5) affectionate support (involving expressions of love and affection). For each item, patients were asked to indicate how often each kind of support was available to them if they needed it. Response choices were as follows: 10none of the time, 20a little of the time, 30some of the time, 4 0most of the time, and 50all of the time. A total score was generated for social support with higher scores reflecting better social support. The Cronbach's alpha of this scale was 0.96 and was categorized into low, medium, and high social support based on 33rd and 66th percentile distribution of scores. v. HIV disclosure status: disclosure refers to whether participants disclosed to the most important people in their lives, with response categories ''yes/no.'' 3. Clinical factors: ART side effects, number of previous opportunistic infections (OIs), and CD4 cell count.
i. ART side effects: refers to whether the participant experienced side effects from ART. ii. Number of previous OIs: refers to the number of OIs the respondents had been diagnosed or experienced in the past 6 months. Respondents were categorized into three groups: none, 1, and 2 or more previous OIs. iii. CD4 cell count: the patient's record was reviewed to obtain recent CD4 cell count. In this study, the CD4 cell count was used to classify the patients into three categories; B200 cells/mm 3 as severe, 200Á499 cells/mm 3 as moderate, and 500 cells/mm 3 as mild. 4. Generalized psychological distress: the Hospital Anxiety and Depression Scale (HADS) was used to measure symptoms of generalized psychological distress. This scale serves to identify individuals who are likely to meet the formal definition of anxiety and/or depression disorders. The scale shows good reliability among HIV-infected subjects. The Cronbach's alpha was 0.87, and the intraclass correlation coefficient (ICC) was 84% for Ethiopian HIV-seropositive patients (22). The Cronbach's alpha result for HADS of this study was 0.89. Moreover, we used cutoffs ]19 points to measure generalized psychological distress among the HIV-seropositive patients (22).

Statistical analyses
The data collected from the respondents were entered into Epi Info version 7 and imported to SPSS-20. First, descriptive statistics of the generalized psychological distress were generated with the aim of assessing the prevalence of generalized psychological distress in the study population using cutoffs ]19. Logistic regression models were used to assess univariate associations between the dependent variable generalized psychological distress and independent variables, grouped into sets of demographic, psychosocial, and HIV-related clinical risk factors, with unadjusted and adjusted odds ratio (adjusted for sex and age group) reported. After adjusting within each set of risk factors, those associated with generalized psychological distress at a level of significance of 0.1 were entered into a multivariable model using forward stepwise methods to determine their independent effect on psychological distress. A confidence interval of 95% was used to see the precision of the study, and the level of significance was taken at aB0.05.

Ethical clearance
The study protocol was reviewed and approved by Dilla University ethical clearance committee. Moreover, consent was obtained from the study subjects after adequate information was given about the study. Respondents found to have significant psychiatric After adjusting for age and sex, HIV-seropositive persons with moderate stress events, severe perceived stigma, low social support, and those not disclosing their HIV status and experiencing six or more negative life events in the past 6 months were significantly at higher risk of developing psychological distress (Table 3).

Factors independently associated with generalized psychological distress
Moderate stressful events, low social support, number of negative life events of six or more, not disclosing HIV status, and severe and moderate levels of CD4 cell count were independently associated with generalized psychological distress (Table 5).

Discussion
This study is one of few studies conducted to investigate prevalence and factors associated with generalized psychological distress in HIV-positive patients enrolled on an ART program in developing countries like Ethiopia.
The prevalence of generalized psychological distress obtained in this study was 11.2% (HADS ]19). The HADS was found to be a reliable and acceptable tool in a previous study conducted in Ethiopia among HIVseropositive individuals (22). Because there is a large overlap between HIV manifestation and somatic symptoms of common generalized psychological distress, like anxiety and depression, it is important to use measures that do not contain somatic items to assess psychological distress validly and reliably in HIV-seropositive persons. The prevalence of psychological distress in this study is lower than prevalence of common mental disorders (21.8%) reported from a study conducted in Ethiopia among HIV-seropositive individuals (23), and it is comparable to figures of 14.2% of mental distress according to a study from Zambia conducted in the general population (11) and 15.5% of mixed prevalence proportion of anxiety and depression found in rural Tanzania (24). The differences in prevalence of generalized psychological distress could be attributable to several factors, including the population being studied, the study periods, and the diagnostic tools. The cutoff point we used is arbitrary because of uncertainty of the correct cutoff points of HADS. This may be one of the limitations of this study, and further research needs to be conducted that addresses the issues of sensitivity and specificity of HADS in HIVseropositive persons.
Unlike studies from Botswana (10) and Brazil (18) that have found illiteracy significantly associated with increased symptoms of depression, we did not found any association between educational status and generalized psychological distress, which is similar to some previous studies that have reported no association between educational status and depression symptoms (6,13,25,26).
It is well established that widows and unstable marital relationships increase the risk of having depression and anxiety symptoms (27). Moreover, widowed subjects might have stress when one loses the beloved one (28), but contrary to the aforementioned two findings, we have not found any association between marital status and psychological distress.

Solomon H. Tesfaye and Girma T. Bune
In this study, moderate stress events were independent risk factors for psychological distress. This is similar to other studies that revealed that high stress (13) and a current stressful event (29) were found to be associated with depression, and in HIV-positive women from South Africa the degree of stressful event was significantly associated with psychological distress (30). Common psychological distresses like anxiety and depression can cause stress, which, in turn, can trigger a psychological distress. In fact, further research is recommended, such as a longitudinal study to examine the cause-and-effect association of stress and psychological distress. Several studies (6,8,10,16,31) have found perceived stigma to be a risk factor for psychological distress, which is contrary to the findings of this study. Because it is well known that people with perceived stigma may have a low self-image and be socially isolated and predisposed to develop psychological distress (32), further research is needed in this area with a large sample size and strong epidemiological study designs.
According to this study, lower social support increased the risk of psychological distress, which strengthened the findings of many other studies. Previous research (33,34) conducted in the United States in PLHA receiving ART pointed out similar results. In other words, the results show that poor social support was associated with psychological distress. Additionally, a case-control study from Nigeria (35) found that poor social support was associated with anxiety. A study from India (8) also revealed higher social support was associated with lower depression. Similarly, functional social support significantly reduced symptoms of depression and anxiety (36). The study findings suggest the need to strengthen social support networks for PLHA so as to reduce psychological distress. The findings of association between negative life events with psychological distress correspond to three crosssectional studies (13,17,37) and one longitudinal study (29). These studies found that the risk of developing depression was increased as the number of life events increased in the last 6 months. However, the cross-sectional study design precludes any conclusion on temporality of this association. But one longitudinal study conducted in a cohort of HIV/AIDS subjects found no significant difference in depression rates among those who experienced negative life events and those who did not (38).
Unlike two cross-sectional studies (13, 32) conducted among PLHA receiving ART, this study found that CD4 counts ]200 cells/mm 3 and between 200 and 499 cells/ mm 3 were both associated with psychological distress.
But the present study finding is similar with reports from Uganda (39); controlling for sociodemographic variables, their analysis showed that those respondents who have CD4 cell counts B50 cells/microliter were more likely to develop depression. Another cross-sectional study from Italy found also that those who report anxiety symptoms have lower CD4 cell counts than those with no anxiety (7). Moreover, one longitudinal study from the United States (40) reported that chronic depressive symptoms were associated with lower CD4 cell counts and concluded that depressive symptoms among women are associated with HIV disease progression. But another longitudinal study conducted among HIV-infected women found no association between CD4 cell count with depressive and anxiety symptoms (41). Since CD4 cell count has been used to measure disease progression in HIV, this, together with the findings of the present study, suggests further research is needed in low-income countries to assess the potential relationship of depression, immunity, and HIV disease progression in PLHA on ART.
In our study, we have found that not disclosing one's own HIV status is an independent risk factor of psychological distress. Patients who choose not to disclose their HIV status to anyone may suffer because they lack social  support and have difficulty maintaining close personal relationships.
One of the strengths of this study is the use of HADS, which was validated and reported good reliability among HIV-seropositive individuals initiating ART in Ethiopia.
There are a few limitations in this study. First, study participants were selected only from those HIVseropositive individuals who were attending public health facilities for ART. Patients suffering from psychological distress may be less likely to seek care and initiate treatment. They are then underrepresented in this study. Second, because the empirical analysis was based on selfreported information, the measures may not be completely accurate. Particularly problematic variables include self-reported side effects. Third, this was a cross-sectional study, so that all the data were measured at the same time. Consequently, causality between symptoms of psychological distress and their correlations could not be firmly established.

Conclusions
This study provides prevalence of psychological distress that is comparable with findings of other studies in sub-Saharan Africa. Moderately stressful events, low social support, negative life events, not disclosing HIV status, and low and moderate levels of CD4 cell count are independent risk factors associated with psychological distress.
The impact of psychosocial and HIV-related clinical factors on mental health of HIV-seropositive individual draws attention for researchers to further investigate the interrelationship between psychosocial, mental, and physical aspects of the public sector providing ART. Based on our results, we can recommend that, among ambulatory HIV-seropositive individuals in this sociocultural environment, appropriate screening and treatment for psychological distress should be considered for comprehensive HIV care. Second, establishment of a social support group mechanism is needed to reduce stress and the negative life events associated with psychological distress. The findings are important in terms of their relevance to identifying high-risk groups for generalized psychological distress and, thus, preventing distress through integrating mental health services with HIV/AIDS care and support program.

Authors' contributions
Solomon Hailemariam contributed in the generation of the topic, preparation of the proposal, data collection, analyses, and development of the manuscript. Girma Tenkolu contributed in reviewing the proposal, assisting in data collection, analysis, and critical review of final manuscript.  would like to thank the respondents of this study for their kind participation in providing the relevant data.