Background

Mental illness is among many non-AIDS complications that limit the HIV/AIDS patients’ quality of life (QOL) [1, 2]. Mental illness and HIV infection are linked in many ways including (I) HIV infection often result in serious emotional disturbance due to its malignant course and associated stigma [3]. (II) HIV has direct effects on a central nervous system which may lead to cognitive, perceptual and behavioral changes [3]. (III) Mental illness can be a consequence of opportunistic neurological and systemic infections and their treatments [4]. (IV) Some Highly Active Antiretroviral Therapy (HAART) have psychiatric side effects [4]. (V) Patients of severe mental illness are at risk to HIV infection [5]. (VI) HIV and psycho-active Substance use is connected in direct ways (IV use) and in indirect ways influencing sexual behavior [3,4,5]. There are many reported impacts of mental health problems among HIV/AIDS patients including speeding up the disease progression, reducing adherence to treatments, affecting willingness or ability to access health care and increased risk of transmission of other sexually transmitted infections (STI) by increasing high-risk behaviors [6].

Common mental disorders (CMDs) is set of signs and symptoms of non-psychotic depression, anxiety, and somatoform disorders and it is frequently reported among HIV infected people [7,8,9,10,11,12]. The magnitude of major depressive disorder (MDD) ranges from 16.2 to 36% among HIV patients in the USA [13]. This is as compared to the general population’s prevalence 4.2%, it is 4 to 7 fold greater [14]. Another study in LAMIC (Kenya, Democratic Republic of Congo and Thailand) again reported that depression is higher in HIV patients than in HIV negative individuals [15]. However, these studies on co-occurring of mental illness and HIV-infection is still limited in sub-Saharan Africa including Ethiopia [16,17,18]. The study on Anxiety Disorders among adult HIV/AIDS patients in a Sub-Saharan Africa revealed that the magnitude of anxiety disorders and mixed anxiety-depressive disorder among adult PLWHA was 21.7, and 5.3% respectively. Lack of family support, unemployment and being unmarried were factors significantly associated with anxiety disorders among participants [19]. Study in Nigeria showed that individuals with HIV had significantly higher rates of common mental disorder (OR = 3.58, 95% CI = 1.44–8.94) than healthy people and stage of the HIV was significantly associated with common mental disorder [7]. A systematic review and meta-analysis result in sub-Saharan Africa revealed that prevalence estimates of depression ranged between 9 and 32% in PLWHA on Antiretroviral Therapy (ART) and in untreated or mixed (treated/untreated) ones. Low socio-economic conditions in PLWHA on ART, female sex and immunosuppression in mixed/untreated PLWHA were reported associated factors [20].

The aim of the current study is to determine the magnitude of common mental disorders and factors associated with it among HIV positive individuals undergoing HIV services in Harar town, Eastern Ethiopia. It is important as baseline information for the concerned stakeholders including health professionals and policymakers and in general to improve the quality of care for HIV/AID patients.

Methods

Study Area

We conducted the study in Harar town, Harari regional state, Eastern Ethiopia, at Hiwot Fana specialized University and Jugel governmental Hospital. Location of Harar is 527 km from the capital city of Ethiopia, Addis Ababa to the direction of the east. In Harar town, there are 19 health posts, 2 private hospitals, 8 health centers, 1 FGAE (Family Guidance Association of Ethiopia) clinic, 3 government hospitals and one university specialized hospital. During our study, a total of 7558 HIV patients were enrolled at ART clinic in the town. Hiwot Fana and Jugel hospitals are the major ART sites in the region.

Study design

We conducted through facility-based Cross-sectional study design.

Source population

Adult Sero-positive individuals in Harar town, Eastern Ethiopia.

Study population

The 420 randomly selected adult HIV patients enrolled in HIV services (Pre-ART and on ART) at Hiwot Fana specialized University Hospital and Jugel governmental Hospital in Harar town, Eastern Ethiopia and who were 18 years or older were included in the study and patients who were critically ill during data collection period were excluded.

Sample size determination

We calculated Sample size using single population proportion formula taking p = 46.7% from previous study “Prevalence of common mental disorders among HIV/AIDS patients in Ethiopia” [8], d of precision 5, 95% confidence interval and 10% non- response rate.

$$ {\displaystyle \begin{array}{c}\mathrm{n}=\frac{{\left(\mathrm{z}\;\upalpha /2\right)}^2\mathrm{p}\left(1-\mathrm{p}\right)}{{\mathrm{d}}^2}\kern1em \mathrm{where}\kern1em \mathrm{n}=\mathrm{sample}\kern0.5em \mathrm{size},\kern0.5em \mathrm{Z}\upalpha /2=\mathrm{Z}\kern0.5em \mathrm{score}\kern0.5em \mathrm{at}\kern0.5em 95\%\kern0.5em \mathrm{CI}=1.96,\kern0.5em \mathrm{p}=46.7\%\kern0.5em \mathrm{and}\kern0.5em \mathrm{d}=5\%\kern0.5em (0.05)\\ {}\mathrm{n}=\frac{(1.96)^20.467\left(1-0.467\right)}{(0.05)^2}=\frac{3.8416\times 0.467\times 0.533}{0.0025}=\mathbf{382}\end{array}} $$

Final sample size with 10% non-response rate was 420.

Sampling technique

During our study period, there were 7558 registered HIV/AIDS patients enrolled in HIV services at both Hiwot Fana University specialized and Jugel governmental hospitals. Final sample size (420) was proportionally allocated to each hospital 281 from Hiwot Fana and 139 from Jugel hospital. At each hospital patients were stratified into pre-ART and ART based on their ART status. Then we selected eligible patients by simple random sampling technique based on their card number (Fig. 1).

Fig. 1
figure 1

Sampling procedure of adult patients attending HIV follow up service in Harar town, Eastern Ethiopia, 2015

Full size image

Data collection

We collected data through patient interview and review of medical records. Data regarding socio–demographics, CMDs, majority of clinical and psychosocial variables like (perceived HIV stigma, partner’s Sero-status, Children’s Sero status, death of either partner or family members due to HIV/AIDS, hospitalized in the last month, decreased level of functioning than usual, lost job due to HIV/AIDS, faced severe stress in the last 6 months, having family history of mental illness and having any comorbid chronic physical illness) and Substance-related variables (current drinking alcohol, chewing Khat, smoking cigarette, and using illicit substances) were retrieved by Interviewing patients using the adopted standardized questionnaire through face to face interview using interviewer-administered a standardized questionnaire. However, Respective medical Records like (HIV/AIDS stage, CD4 count and started ART) were obtained from patients’ medical record. We used Self Reporting Questionnaire (SRQ-20) to get data about CMDs. We used standardized questionnaire in interview-form instead of self-report because of the 16.2% (68) of our study participants were unable to read and write and they could not fill the questionnaire. The questionnaire was translated into local languages Afan Oromo and Amharic and validated to Ethiopian context. As common mental disorders (CMDs) refers to the set of signs and symptoms of non-psychotic depression, anxiety, and somatoform disorders, considering how challenging it can be to diagnose these disorders in primary care practice, the World Health Organization (WHO) developed the Self Reporting Questionnaire (SRQ-20), a screening instrument to identify suspected CMDs cases in primary care settings [21]. The SRQ-20 has 20 yes/no questions and it has been validated in Ethiopia, with high sensitivity (85.7%) and specificity (75.6%) [22]. Cut point of SRQ-20 is different in institution based and community based and usually cut-point of ≥7 is used in institution based [22]. Data collectors were 5 diploma nurses with previous data collection experience and supervisors were 3-graduate nurses and all of them were given one-days training prior to the data collection period. The study was conducted after approval obtained from the College of Health and Medical Sciences, Institutional Research Ethics Review Committee (IRERC) of Haramaya University. Written informed consent was obtained from the study participants after providing a clear explanation of the objective of the study. The collected data were kept confidential. Participants’ right to refuse and the chance to ask anything about the study was respected. The names of the participants were not written.

Study Variables

Dependent variable

Status of common mental disorder (yes/no).

Independent Variables

Socio-demographic variables (age, sex, religion, ethnicity, marital status, occupational status, income, and educational status), clinical and psychosocial variables (HIV/AIDS stage, CD4 count, started ART, perceived HIV stigma, partner’s Sero-status, Children’s Sero status, death of either partner or family members due to HIV/AIDS, hospitalized in the last month, decreased level of functioning than usual, lost job due to HIV/AIDS, faced severe stress in the last 6 months, having family history of mental illness and having any comorbid chronic physical illness), Substance-related variables (current drinking alcohol, chewing Khat, smoking cigarette, and using illicit substances).

Data processing and analysis

After checking for completeness data were entered into a computer using EpiData 3.1 then exported to SPSS for analysis. In statistical analysis we used categorical variables. Our dependent variable (CMD) is dichotomous which is categorized as yes (having CMD) or no (have no CMD). Since our dependent variable was dichotomous and we used cross-sectional study design, in order to explore the relationship between dependent variable and each independent variable we run a bivariate analysis using odds ratio (OR) of chi square. All independent variables that have a significant association with dependent variable (CMD) during bivariate analysis were adjusted together into multivariate analysis specifically binary logistic regression in order to control confounders. A p value of < 0.05 was considered as statistically significant association at 95% of confidence interval. We used Descriptive analysis to describe variables using frequencies, percentages, tables, and figures.

Data quality control

To control the quality of data, we used the standard questionnaire in the Ethiopian context. The pretest was done at another Hospital and necessary adjustments were done before the actual data collection time. Data collectors collected data under the close supervision of trained supervisors. A one-day training about the objective of the study and other related issues was given for data collectors and supervisors.

Operational Definition

Common mental disorders: Patients who scored ≥7 of SRQ-20 items were considered as having CMD [8]. Current substance use: Is defined as participants who had used substances at least once in the last one month prior to the study period [23].

Ethical considerations

The College of Health and Medical Sciences Institutional Health Research Ethics Review Committee of Haramaya University had ethically approved the study. Permission from health facilities managers and written informed consent from patients were sought before administering questionnaires. We used anonymous questionnaires to keep patients’ confidentiality and the interview was conducted in a private room to assure the patients’ privacy. Patients with CMD were referred for professional care in the hospitals.

Results

Socio-demographic characteristics of respondents

From the total 420 patients, all were interviewed (providing response rate 100%); 26.0% of the respondents were males, and 71.0% were age less than 40 years. From the total participants, 53.6% were Oromo by ethnicity, 67.9% were Orthodox Christians, 37.6% were married and living together, 39.0% get a monthly pocket money of 500–999 Ethiopian birr, 44.3% educated primary school, and 17.4% were Government employee (Table 1).

Table 1 Socio-demographic characteristics of adult patients attending HIV follow up service in Harar town, eastern Ethiopia, 2015
Full size table

Prevalence of CMD among HIV/AIDS patients

The result revealed that among the 420 HIV/AIDS patients those were interviewed; Using SRQ-20 questionnaire (28.1%; 95% CI; 26.14, 30.06) of them had CMD.

Factors associated with CMD among patients living with HIV/AIDS

The association between CMD and independent variables were determined using bivariate and multivariate (binary logistic regression) analyses. In the final model of multivariate logistic regression analysis HIV/AIDS stage and current drinking alcohol were found having statistically significant association with CMD. However, Socio-demographic variables like: Marital status and income were statistically significant in the binary logistic regression analysis but not significant in the final model of multivariate logistic regression analysis (Table 2).

Table 2 Binary logistic regression: Socio-demographic factors independently associated with CMD among adult patients attending HIV follow up service in Harar town, eastern Ethiopia, 2015
Full size table

Likewise, Clinical, psychosocial and substance use factors like: perceived HIV stigma, lost job due to HIV/AIDS, having any comorbid chronic physical health problems, and current smoking cigarettes, chewing Khat and using illicit substances were again statistically significant in the binary logistic regression analysis but not significant in the final model (Table 3).

Table 3 Binary logistic regression: Clinical, psychosocial and substance use factors independently associated with CMD among adult patients attending HIV follow up service in Harar town, eastern Ethiopia, 2015
Full size table

As revealed in final multivariate logistic regression analysis, clients who have stage 4 HIV/AIDS were 3.37 times more likely to have CMD than clients with stage 1 HIV/AIDS (AOR 3.37, 95% CI: 1.45, 7.83) and Patients who have family history of mental illness were 2.65 times more likely to develop CMD than clients who have no family history of mental illness (AOR 2.65, 95% CI: 1.26, 5.54). Those individuals currently drinking alcohol, were more than five times more likely to develop CMD than those who are currently not drinking (AOR 5.1, 95% CI: 2.04, 12.79) (Table 4).

Table 4 Multivariate logistic regression: Socio-demographic, Clinical, psychosocial and substance use factors independently associated with CMD among adult patients attending HIV follow up service in Harar town, eastern Ethiopia, 2015
Full size table

Discussion

This study showed that the overall prevalence rate of common mental disorder among HIV/AIDS patients is 28.1% with the range between 26.14 and 30.06%. Associated factors identified in the final model of this study were stage of HIV/AIDS, Family history of mental illness and current drinking alcohol.

In the current study the prevalence of common mental disorder among people living with HIV/AIDS is closer to the finding in south west regional hospitals of Cameroon, which shows 26.7% of HIV/AIDS patients on HAART have depression [24]. On the other hand, the prevalence is greater than that found in Tanzania and South Africa [25, 26] in which the prevalence was 15.5 and 14.2% respectively. This variation could be due to different possible reasons like different method used to assess the condition. The study in Tanzania conducted using ICD-10 common mental health diagnosis and in South Africa they applied DSM diagnoses. But in the current study we applied SRQ-20 standardized questionnaire. The present study result is lower than the studies done in Zimbabwe and Indian setting [27, 28] in which the findings were 67.9 and 58.75% respectively. This difference could be the slight difference of study population nature like in the study conducted in Zimbabwe 92% of the participants were on HAART, which indicates that they could have severe illness. As the severity of HIV/AIDS increased the patients could develop CMD more [27]. But in the current study only 71.7% were on HAART and the left were pre-ART. The other possible reason could be in the present study majority of the participants were young age as 71.0% of them were at age less than 40 years. Study on common mental disorder among general population showed that older age is significantly associated with higher prevalence of common mental disorder [29].

Our study finding shown that patients who have stage 4 HIV/AIDS were more likely to develop common mental disorder (AOR 3.37, 95% CI: 1.45, 7.83) and it is in line with the results of studies done in India and Ethiopia [27, 30]. Patients those are currently drinking alcohol are also more likely to develop common mental disorder (AOR 5.1, 95% CI: 2.04, 12.79) and this result is supported by study conducted on common mental disorder among general population in Ethiopia [29]. As more alcohol enters the bloodstream, the areas of the brain associated with emotions and movement are affected, often resulting in exaggerated states of emotion (anger, withdrawal, depression or aggressiveness) and uncoordinated muscle movements [31]. Also, our study revealed that patients who had family history of mental illness were more likely to have common mental disorder as compared to patients who have no family history of mental illness (AOR 2.65, 95% CI: 1.26, 5.54).

Limitations of the study

The study was hospital based and some patients with severe common mental disorder might unlikely to be available in the hospital due to their severity of common mental disorder during data collection period. This could influence the prevalence of common mental disorder and it might not be generalized to the total population of people living with HIV/AIDS in the region. Also, our study was cross-sectional and it cannot show the cause-effect relationship between common mental disorder and independent variables.

Conclusions

In this study the prevalence of CMD is relatively high. HIV/AIDS stage, having family history of mental illness and current drinking alcohol were the identified factors those have significant association with common mental disorder among patients living with HIV/AIDS. The study finding will provides information to form rational foundation for prevention and planning to bring change in contributing factors for developing CMD among patients living with HIV/AIDS and also will be base line information for further study. The study recommended that it is important to control the progression of HIV like early detecting and treating opportunistic infections.