Depression, Anxiety, and Suicide Risk among Ugandan Youth in Vocational Training

The current study examines the prevalence of depression, anxiety, suicide risk, and PTSD in Ugandan youth (13-25 years) attending vocational training programs. Youth from five urban (n=224 females, 81 males) and four rural (n=153 females only) vocational training centers operated by a non-governmental organization completed demographic and mental health questionnaires. Nearly half of the youth reported moderate or severe depression and/or anxiety. More than half reported anxiety and depression-related impairment. Nearly a quarter of youth had considered or attempted suicide. More than half screened positive on the PC-PTSD screen. Rural female youth reported the most food insecurity (56.9%), trafficking (37.9%), severe depression (35.9%), depression-related impairment (56.9%), severe anxiety (26.1%), and anxiety-related impairment (55.6%). Results from this study suggest that Ugandan youth have exceedingly high rates of depression, anxiety, suicide risk, and probable PTSD. Rural female youth may be especially at risk. Relevant treatment interventions are needed that can be adapted to youth in vocational training centers.

future adult life [7]. However, a major gap in knowledge involves the prevalence of depression and associated factors in these settings [8].
In Uganda, more than 75% of the population is under 30 years of age, and 22.5% is between ages 18-30 [9,10]. Youth may face multiple psychosocial stressors, including exposure to physical, emotional, interpersonal, and sexual violence [11][12][13]. For example, millions of children have been orphaned by the AIDS epidemic and suffer high levels of psychosocial and psychological distress as a result [14]. Furthermore, children living in the slums of Kampala have reported high rates of abuse (N=380, 34.0%) and commercial sexual exploitation (N=593, 39%), [15][16][17]. Though country-wide poverty has declined significantly in the recent decades, the unemployment rate is disproportionately higher for youth versus adults and even higher among urban and female youth when compared to their rural and male counterparts [10]. Poverty and unemployment, in turn, make youth susceptible to forced-labour and sex trafficking [18]. Cumulatively, these stressors can contribute to significant psychiatric morbidity, including depression, anxiety, and post-traumatic stress disorder, especially among more vulnerable young people [5,19,20].
The current literature documenting the need for mental health care among groups of Ugandan youth is very limited. The few existing studies have focused on Ugandan adolescents with HIV or exposed to war trauma, including being abducted to become child soldiers, and found they experienced elevated psychological distress [21][22][23][24][25][26]. High rates of suicidal ideation among youth (N=1134, 23.54%), depressive symptoms among adolescents (N=519, 21.0%), and anxiety disorders among children and adolescents (N=1680, 26.6%) have also been reported [27][28][29].
These findings underscore the importance of determining the prevalence of common mental illnesses among groups of Ugandan youth in settings that could provide mental health care. Enabling young people to enter the workforce is essential for reducing the burden of poverty in Uganda. Identifying key groups of youth likely to enter the workforce and in need of mental healthcare is key to capitalizing on their full socioeconomic potential. The present study attempts to address this critical gap in knowledge by describing the prevalence of depression, anxiety, and PTSD among impoverished and unemployed Ugandan youth participating in vocational training programs. We compare rates based on youths' gender because of depression's well-established disproportionate prevalence among post-pubertal adolescent and adult females across diverse settings [5,30], as well as the greater risk Ugandan female youth face for unemployment and sexual exploitation [10,18]. Due to the unique stressors facing rural youth, including proximity to health care, lacking daily necessities, and greater likelihood of unemployment, this study also included an urban versus rural analysis [10,31].

Study Sample and Setting
The study sample is comprised of youth ages 13-25 drawn from nine youth drop-in centres. The drop-in centres are operated by Uganda Youth Development Link (UYDEL) a non-governmental organization started in 1993 with the aim of enhancing socioeconomic opportunities for vulnerable and disadvantaged young people. UYDEL targets young people who have been victims of commercial sexual exploitation and child labour trafficking, who have suffered from alcohol, drug, and substance abuse, and who have little ability to obtain gainful employment without training and toolkits to begin gainful employment. UYDEL provides evaluation of and support for these vulnerable youth, including vocational skills training and emotional support at 13 youth drop-in centres located in eight districts in the central region of Uganda (Kampala, Wakiso, Mukono, Mubende, Mityana, Gomba, Sembabule, Bukomansimbi, and Rakai). The study sample was drawn from five drop-in centres in urban settings and four drop-in centres in rural settings. Because the rural drop-in centres operate a project exclusively for young women, no males were drawn from the rural drop-in centres. All youth who visited the drop-in centres on the days the research team held screenings were invited to participate. A total of 224 women and 81 men from the urban drop-in centres and 157 women from the rural drop-in centres participated.

Data Collection Procedure
Prior to participating all youth completed informed consent. Staff members at the youth drop-in centres interviewed each participant individually and entered data into tablets directly. Data was gathered through October 2017 to May 2018. Less than 2% declined to participate.

Measures
Sociodemographic characteristics included age, sex, and education (less than secondary, some secondary, and secondary school graduates). Food security was determined by a single question asking about going without food for a whole day (24 hours) in the past month (never, at least once). Housing status was evaluated by the number of rooms (single room, two rooms, three or more rooms) or homelessness. Living situation was based on whether youth were living with their parents or with other adults. Participants were also asked about a history of being trafficked for work (yes/no).
The PHQ-9 is a brief nine-item self-report assessment for screening, diagnosing, monitoring, and measuring the severity of depression. The PHQ-9 asks about the degree to which the respondent has been bothered by depressive symptoms over the past 14 days. Response options are measured from 0 ("Not at All") to 3 ("Nearly Every Day"). Scores range from 0-27 and are categorized as minimal (0-4), mild (5-9), moderate (10)(11)(12)(13)(14), or severe (15 or higher). Good construct, criterion, and external validity have been demonstrated [32].
Each study participant was asked to answer yes or no to the following questions about suicide risk: 1) "Has there been a time in the past month when you have had serious thoughts about ending your life?"; 2) "Have you ever in your whole life tried to kill yourself or made a suicide attempt?"; and 3) "Have any of your blood relatives committed suicide?".
After completing the PHQ-9, participants reported depression's associated degree of functional impairment by responding to "How difficult have these problems made it for you?" ("not difficult at all," "somewhat difficult," "very difficult," "extremely difficult"). They reported the degree of anxiety-related impairment in a similar fashion after completing the GAD-7.
The Primary Care PTSD (PC-PTSD) Screen is a 4-item, self-report screening tool. It includes an introductory sentence to cue respondents to traumatic events followed by four symptoms items pertaining to a DSM-IV diagnosis of PTSD. Screens are considered positive if the participant answers yes to any three items. A cut-off score of 3 is considered optimally sensitive for a probable diagnosis of PTSD [34].

Data Analyses
We conducted bivariate analyses to compare gender (female vs. male) from urban and geographic area (rural vs. urban) among females with sociodemographic factors and depression variables. We present means with standard deviation for continuous variables or percentages for categorical variables. These analyses used t-tests for continuous variables and Chi-square tests for categorical variables. For variables with small cell sizes (expected values less than 5), we conducted sensitivity analyses using the Fisher exact test, with no change in results. Table 1 presents the demographic similarities and differences between male (n=81) and female (n=224) participants drawn from the five urban youth drop-in centres. Males (17.9±2.3, 13-24) and females (17.9±2.2, 13-25) were of similar age. They reported similar educational background, food insecurity, housing, and living arrangements. For example, only a very small percentage completed secondary education (8.6% for males and 4.5% for females), and most lived in either one or two-room homes with only 1.2-1.3% reporting homelessness. Nearly a third reported living with adults other than their parents (30.6-31.9%), while a lesser though still notable number reported a history of being trafficked for work (9.9% for males and 14.7% for females). Male youth (53.1%) reported greater food insecurity compared to female youth (33.9%).

Depression, Suicide Risk, Anxiety, and PTSD of Female Urban Versus Rural Youth
The majority of youth reported moderate or severe depression and depression related impairment, but rural youth reported a greater burden of severe depression (35.9% vs 11.2%) and impairment (56.9% vs 35.7%). Both groups reported similar lifetime histories of suicide attempts (21.6-22.8%) and family histories of suicide (15.2-19.6%). However, rural youth were also more likely to report suicidal ideation in the past month (37.9% vs. 26.8%). While the overall rates of moderate (29.4%) and severe (12.7%) anxiety were high, rural youth were much more likely to report severe anxiety (26.1% vs. 3.6%). Rural youth experienced greater anxiety-related impairment (55.6% vs 35.7%). Over half (54.5-62.1%) screened positive on the PC-PTSD.

Discussion
High rates of depression, anxiety, and probable PTSD were reported among Ugandan youth attending vocational training programs. Furthermore, just under a quarter of these youth considered or had attempted suicide in the past. Over half of these youth felt that depression and anxiety made it very or extremely difficult for them to function. This strongly suggests that evidence-based interventions, such as cognitive behavioural therapy, should be provided within these settings to support these youth as they gain vocational skills and gain employment. Several factors likely contribute to the high rates of emotional disorders and impairment noted in these youth. They have low levels of educational attainment, and many experience extreme stressors such as food insecurity. Nearly a quarter have been subjected to human trafficking. These factors may help explain depression differences based on gender and geographical residence. The current study found no gender differences in rates of depression or related impairment, or rates of anxiety or PTSD. This is surprising given that post-pubertal females are typically more likely to be depressed than males [5,30]. It is possible that the occurrence of major stressors experienced by both female and male youth in our sample may account for the high rates of depression among both genders. In contrast, rural female youth were more likely than urban female youth to endorse severe depression and anxiety, as well as related impairment. Rural youth in the current sample were also more likely to experience major stressors and a history of trafficking, which may have contributed to their heightened rates of depression.
Gender and geographic differences also emerged in suicide risk among youth in the current sample. While female youth were more likely to endorse active SI, no differences emerged in rates of past suicide attempts. Several studies have documented that females are more likely to experience SI and attempt suicide than males [35]. However, much of this research is based on high income countries. Research examining suicide risk in African LMICs suggests that gender differences may not be as pronounced as in high-income countries [8,35]. Similarly, rural vs. urban female youth were more likely to endorse SI but not SA. More population-level data is needed to understand suicide in Uganda. The current findings indicate that among largely disadvantaged youth who are preparing to enter the workforce, suicidal ideation and prior attempts are a significant problem.
Existing studies have suggested that rates of psychopathology are high among Ugandan youth. However, little is known about youth attending settings where services could potentially be provided. NGOs and other organizations that provide services to youth often prepare them for the workforce, but maybe miss important psychological problems that can further improve the chances youth will be able to maintain work and stable life. Furthermore, youth who are already linked to a service, such as the drop-in centres in our study, may be easier to engage in mental health care. Youth who are already engaged with a type of service setting may experience fewer logistical barriers related to accessing mental health care. The current study recruited youth from drop-in centres providing vocational training and emotional support to disadvantaged youth. Despite being connected to some support at the drop-in centres, youth in the current study still reported high rates of psychopathology. Fortunately, these youth have an opportunity to develop vocational skills that could improve their lives significantly. The social work staff in these settings are potentially available to provide evidence-based care for depression and anxiety within the vocational training program. Future research should examine the effectiveness of evidence-based mental health treatments, such as cognitive-behavioural therapy, for disadvantaged youth in these settings.
Several limitations should be noted with this study. The participants are a volunteer sample, although nearly all youth attending the programs were screened. The instruments were screening instruments and might not be indicative of need for treatment for all youth who screened positive. Nonetheless, this study clearly identifies a need for mental health treatment to be embedded within vocational training programs for Ugandan youth to improve the quality of their lives, as well as help them to potentiate the training.
Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Dr. Badru Bukenya is a Uganda based academic, policy analyst and development practitioner. He is currently a Senior Lecturer in the Department of Social Work and Social Administration, Makerere University Kampala. Badru's research focuses developmental social work more broadly and specifically on statecivil society engagement, public service delivery, public sector performance, and social protection. Badru employs mixed methods research approaches with expertise in data management packages such as NVivo and fuzzy set qualitative comparative analysis (fsQCA).
Mr. Kasirye is a founder of an NGO called UYDEL, acted as lead NGO expert for UNODC on drug abuse prevention work and research in sub-Saharan Africa. He is African regional trainer in drug abuse prevention. He has been involved in the planning, implementation, advocacy and evaluation of programs related alcohol and drug abuse, street and slum children in. He is a scientific advisor to the Mentor Foundation, an organization that facilitates the use of evidence-based programming among NGOs working on substance abuse prevention. Formerly a chairperson of the East Africa Policy Alliance. He is a US Fulbright Humphrey Fellow 2011 on new innovations in substance prevention. He has participated in NIDAsupported collaborative research on substance abuse and run away youth in Uganda. His past research has focused on juveniles, drug abuse and sexual risk behaviors. He is an author of the book on children and vulnerability in Uganda.
Joanitah Lunkuse is a highly experienced and motivated psychologist passionate about children and young people from Kampala working with Uganda Youth Development Link. She has interacted and worked with, as well as interviewed vulnerable young people, aged 13-24 years, affected by sexual abuse, drugs, alcohol and substance abuse and other insecurities alongside crime. She is currently implementing a project targeting slum young people to strengthen their socio-economic capacity and improve their access to drugs and substance information, sexual reproductive health rights information and services. She has also served as a youth representative and consultant in development of positive youth engagements with SVRI,UNODC among others.
Moses Kinobi is a social worker with Uganda Youth Development Link and has got a six years working experience with children and youth between ages of 13-24years who are vulnerable to exploitation such as commercial sexual exploitation, victims of human trafficking, worst forms of child labor, out of school slum youth, street children, youth from poor families and young people having a drug abuse problem which makes them prone to depression, stress and trauma. Moses is a trained depression and trauma counsellor using the model building resiliency and increasing Community Hope. Dr. Lingqi Tang is a principal statistician at the UCLA Health Services Research Center since 1998 and has worked on major center projects. She received her Ph.D. in Statistics and Applied Probability from the University of Alberta in 1992. Her interests include study design, finite population sampling, and analysis of incomplete data. She is highly skilled at advanced programming and has extensive knowledge and experience with implementing advanced statistical methods, including multiple imputation methods and hierarchical modeling.

Dr. Sylvanna Vargas
Dr. Jeanne Miranda is a Professor in the Department of Psychiatry and Biobehavioral Sciences at UCLA. She is a mental health services researcher who has focused her work on providing mental health care to lowincome and minority communities. She holds a Ph.D. in Clinical Psychology from University of Kansas and completed postdoctoral training at University of California, San Francisco. Dr. Miranda's major research contributions have been in evaluating the impact of mental health care for ethnic minority communities.