Context
Large numbers of people in Ethiopia are affected by common mental disorders, particularly by depression and anxiety, and these conditions are slightly higher among women compared to men [24]. These disorders are highly prevalent in the community that accounts for the pooled prevalence of 22% [24]. Depression alone contributes to about 6.5% of the burden of diseases in Ethiopia [25]. On the other hand, only a few people receive formal mental health treatment. In Ethiopia, a meta-analysis reported that the pooled prevalence of formal mental health help-seeking intention and behaviors of people with depression is 48% and 38% respectively [26]. Modern mental health treatments, including psychotherapy and pharmacology, are mainly concentrated in the capital city of Ethiopia, Addis Ababa [27].
Higher education in Ethiopia began in 1950 with the establishment of the University College of Addis Ababa [28]. At present, within the nine regions and the two city self-administrations, 45 public universities train students in regular, summer, extension, and distance programs. According to the Annual abstract report of the Ethiopian Ministry of Education in 2017, there were 679,299 undergraduate students enrolled in all programs in all public universities. Of these, 392788 students (255657 male and 137131female) enrolled in regular undergraduate programs for three, four, five, and six years. Concerning the mental health status of undergraduate students in Ethiopia, the prevalence of mental health problems including anxiety and depression ranged from 21.6 - 49% [11, 12].
Study setting
The current study was conducted at Wolaita Sodo University (WSU) which is one of the 45 public universities in Ethiopia. The University is located in the Sodo town of Wolaita Sodo Zone in Southern Nations, Nationalities, and Peoples’ Regional State (SNNPR) of Ethiopia. Sodo town is located 320 km south of Addis Ababa, the capital city of Ethiopia on the road crossing Butajira and Alaba towns heading to Wolaita Sodo. WSU was established in 2007 as a result of the rapid expansion of access to higher education in Ethiopia over the last twelve years. The university began with an intake of 801 students (609 male and 192 female) in four faculties and sixteen departments. Currently, the university runs undergraduate and graduate programs in six colleges and five schools. During the study period, a total of 12,028 (7321male and 4707 female) regular undergraduate students were registered. WSU has two counseling offices that aimed to provide mental health intervention for students with mental health problems and two health care centers (Ottona hospital and students clinic) that provide both physical and mental health care service. Ottona hospital provides health care service for the community and for the students with severe mental health problem by providing medication.
Study design, objectives and study period
An institution-based cross-sectional survey was conducted among WSU regular undergraduate students from December 4 of 2017 to January 5 of 2018. The objective was to estimate the prevalence of mental health problems, perceived need, and to identify barriers for professional mental health care.
Sample size
Sample size of 1135 was estimated with assumed prevalence rate of mental distress 40.9% [29], precision of +3%, 95% confidence interval, and 10% non-response considered. For the other two objectives (perceived need and barriers to receiving professional mental health care) separate sample size was not estimated. All participants who were screened positive for mental distress were used as the denominator to estimate the proportion of students having perceived need to receive professional mental health care. Those participants who had mental distress symptoms but who did not receive mental health services from professionals in the past three months were eligible to be part of the study.
Sampling and procedures
A stratified multistage sampling technique was used to recruit study participants. List of students’ names was obtained from the main registrar office of the University. The first name of the first participant was selected randomly, and then the remaining participants were selected using systematic random sampling. To accomplish this process, the first step was stratifying regular undergraduate students by their schools/colleges (six colleges and five schools). In the second step, the total sample size was allocated into the 11 strata using probability proportional to the number of the students as a measure of size. The third step was selecting participants from each school and college based on the proportion of the size of each department. The fourth step was selecting participants from each level of study year based on the proportion to each year. Then, the final step was randomly selecting the first participant and systematically selecting the rest participants from each level and section of the study year.
Measurements
The survey questionnaire consisted of four parts, namely, demographic variables, mental health measure, the perceived need for professional mental health care, and barriers to access the service. Participants’ sex, age, religion, ethnicity, marital status, current place of living, area of growing up, department, level of the study year, family history of mental illness and substance use are all documented in the demographic characteristics questionnaire.
Self-Reported Questionnaire (SRQ-20): It is a screening tool for mental distress developed by WHO [30]. SRQ-20 is a self-report instrument with 20 binary responses (yes/no) questions. It has the potential of detecting cases and non-cases with sensitivity ranging from 63-90 and specificity ranging from 44-95 [31]. WHO recommends SRQ-20 as a reliable and valid instrument to detect general Common Mental Disorders [31]. It was developed specifically for use in LMICs [30]. SRQ-20 has been previously translated into Amharic language in Ethiopia, locally validated [32, 33], and used in different community surveys [34-36] and institution-based surveys [11, 12, 29, 37] with cut-off points > 4 [37], > 7 [12], > 8 [29] and > 11 [11]. SRQ-20 has good psychometric properties (i.e. sensitivity 90% and specificity 83% ) for detecting individuals with mental distress in the Ethiopian population with an optimal cut-off point at >7 [38]. To identify cases in the current study, a cut-off point of eight and above was used based on the previous study on ducted among university students in Ethiopia [29]. In the pilot data collected from 38 undergraduate students in a similar population but in a different setting to the current study, the internal consistency was 0.77.
The perceived need for professional mental health care measuring questionnaire: Perceived need for professional mental health services in the past three months was measured using a question used in the previous studies [39, 40]. The question is phrased as follows: ‘Was there a time when you thought you should see a doctor, counselor or other health professionals for your mental distress, but you did not go in the past three months?’ with the response options of Yes/No. “Yes” response implies the need for mental health care but not received in the past three months; whereas “No” response implies no need for mental health care for mental distress.
Barriers to Access to Care Evaluation (BACE-III): BACE was originally developed to identify barriers to receiving professional mental health service for people with mental health problems [41]. It has 30-items to be completed by the participant (self-complete measure). This instrument has good psychometric properties (i.e. validity, reliability, and acceptability) [41]. BACE-III has three dimensions of potential barriers of stigma (12-items), attitudinal (10-items) and instrumental (8-items) related which asks about a range of issues that have ever stopped, delayed or discouraged an individual from receiving or continuing with professional care for a mental health problem. The response scale ranges from 0 (not at all) to 3 (a lot) and higher score indicates a greater barrier. Six of the thirty items contain a fifth option: “Not applicable”. Findings for each barrier are presented in three ways: mean score for the item, barrier to any degree (the percentage circling 1, 2 or 3) or major barrier (the percentage circling 3) based on BACE-III manual for researchers.
For the current study, BACE-III was translated into the Amharic language by two Amharic language experts whose first language is Amharic and their second language is English. One expert who knows the subject matter translated the instrument based on the BACE-III translation guide. The masked back-translation was made by two English language experts and one mental health expert. The research team compared the back-translated instrument with the original version of BACE-III and agreed upon the consistency of the translation. The translated BACE-III was piloted on 40 undergraduate students in a similar population but in a different area of the current study setting resulting in the internal consistency measure of 0.85.
After the pilot study, the authors examined the applicability of each question in the university set-up and noticed that item number 27 and 28 need some modifications. Discussion was made with a mental health expert who has experience of adapting mental health instruments. Then question number 27 which says ‘difficulty taking time off work’ was modified as ‘difficulty taking time off education’ and question number 28 which says ‘concern about what people at work might think, say or do’, was modified as ‘concern about what students might think, say or do’. The final version of the instrument was administered to students who scored eight and above on the SRQ-20 and who had a need to receive professional mental care in the past three months of the study period. Participants who answered “No” for the perceived need for mental health need measuring questionnaire were asked the following question: ‘In the past three months, did you receive help from psychologist, doctors, friends, family, religious leaders or traditional healers?’.
Training of data collectors and data collection procedures
Classroom representatives were served as data collectors. Half a day training was given by the principal investigator for data collectors about the aim of the research, the contents of data collection tools, how to approach participants, ethical issues and responsibility of controlling missing data. The classroom representatives both males and females contacted by the researcher through the help of their department heads, because they have cell phone numbers of each classroom representative. Then, by the assistance of the classroom representatives, the participants come to the selected lecture halls and class rooms and, then the data collectors explained the aim of the study. Finally, after the verbal agreement was received, the data collectors started to collect the data by explaining the instructions of all questionnaires with the close supervision of the principal investigator.
The data collection was carried out before the arrival of students’ final examination that would probably inflate the prevalence of mental distress. Those who scored eight and above on SRQ-20 were asked to answer the questions about the perceived need for professional mental health care whereas, the rest who scored bellow eight on SRQ-20 could not be allowed to pass the next questionnaires. Those who answered “Yes” to the perceived need for professional mental health questionnaire were asked to answer BACE-III questionnaire. Participants who answered “No” were skipped BACE-III and answered why they did not need mental health treatment in the past three months. Finally, after the participants completed the self-administered questionnaires, the data collectors immediately checked the existence of incomplete and missed information before the participants left the room.
Data management and analysis
Data cleaning and cross-checking were done before analysis using Statistical Packages for the Social Sciences (SPSS version 20). Descriptive statistical measures (i.e. percentage, frequency, mean, and standard deviation) were employed to summarize demographic characteristics of the participants and to identify barriers to mental health care service. Multiple linear regression was used to model the association between demographic variables with a mean score of attitudinal, instrumental, and stigma-related barriers with receiving professional mental health service. Univariate regression analysis was used to examine the association of each demographic variable with the total mean scores of attitudinal, instrumental, and stigma-related barriers to receive professional mental health care. Multivariate regression analysis was used to examine the interaction of all demographic variables together to identify variable which are significantly associated with each mean total score of the three dimensions of BACE-III after adjusting for the other variables. The result was reported as being statistically significant whenever the p-value is less than 0.05.
Ethical considerations
Ethical clearance for the conduct of the study was obtained from the Institutional Review Board (IRB) of Addis Ababa University College of Health Sciences. Information sheet containing details of the research and rights of the participants was attached to the questionnaire. Oral informed consent was obtained from the respondents. Finally, the obtained data were kept anonymous and confidential during all stages of the research.