Sample and Procedure
The participants in this study were part of a research project evaluating gambling disorder among secondary school students in Uganda. The study was a cross-sectional survey conducted in 2019 at Mbarara Municipality, a town located in the South-western part of Uganda and the main administrative and commercial center of Mbarara District. To achieve a representative sample of secondary school students in the Municipality, multistage cluster sampling was used in the study. The first stage involved random selection of two schools from each of the Municipality’s six divisions with each division represented by one public and one private school. At the second stage, students were randomly selected from each class in the secondary school using the class register. A total of 921 secondary school students participated in the study. Secondary schools with both ordinary and advance classes were included in the study and students that have not been in school for at least two months or spent an academic session in the school were excluded in the study. Under the supervision of survey administrators, students completed the self-administered questionnaire in English language in a selected hall during the school period. The study was approved by the Research Ethics Committee of Mbarara University of Science and Technology. Written informed consent was obtained from the participating schools and students before the questionnaire was administered, and students that were not willing to participate in the study were allowed to leave the hall where the survey was conducted.
The survey collected the following students’ characteristics: age, gender, religion, family background, class of student, school type, and caretaker/guardian. Participants also answered questions on owning a mobile phone (yes/no), and ever failed a subject in an exam (yes/no).
Psychological distress was assessed with Kessler’s psychological distress scale (K10), a ten item with a 5 point likert scale that evaluates frequency of depressive and anxiety symptoms over past four weeks. Psychological distress was categorized into normal, mild, moderate, and severe psychological distress. Scores range from 10–50, with score of 0–19 categorized as normal, 20–24 as mild distress, 25–29 as moderate distress and scores from 30 and above categorized as severe psychological distress (30, 31). The K10 has been shown to be a reliable and valid tool for screening of adolescents in epidemiological surveys and has been used extensively in Sub-Saharan Africa (32–36).
Alcohol Consumption, Smoking and Substance Use
Substance use among the students was assessed by alcohol, smoking and substance involvement screening test (ASSIST), a tool validated by World Health Organization for assessing the use of alcohol, tobacco products and other drugs. Substance use was assessed using the second question from ASSIST (37, 38). It consists of 10 questions assessing alcohol, smoking and substance use in the past three months, with Never = 0, Once/twice = 2, Monthly = 3, Weekly = 4, and Daily = 5. A score of “0” was classified as “no use” and 2 and above classified as “substance use”.
Risky Sexual Behavior
The sexual behavior of the participants was assessed with questions on being sexually active, use of condoms and ever being pregnant. It was categorized into two: “No risky behavior” and “Risky Sexual behavior”. Respondent that answered “No” to being sexually active or “Yes” to being sexually active but “No” to her or their sexual partner ever being pregnant and “Yes” to consistent use of condoms was categorized as “No risky behavior”.
The chronic illness or condition of the participants was determined by the question “Do you have a chronic (long lasting) illness or health condition” with yes or no response.
Two questions assessed social support, “Is there advice and support provided by your family and friends for you?” and “Are there elders or community leaders that you can easily go to for support and advice? Social support was classified into “No social support” and “Have Social Support”. Any “Yes” answer to one of the question was categorized as have social support.
Data analysis was done with STATA Version 12.0. All participants that did not complete the questionnaire on psychological distress were excluded. Descriptive statistics was used to show the characteristics of the participants. The prevalence of psychological distress was determined by the scores of the students on Kessler’s psychological distress scale (K10). To identify the high risk behaviors associated with psychological distress, multinomial logistic regression was used with normal psychological distress as the reference. The variables which were significant in univariate analysis were included in the multivariable multinomial regression model. All statistical tests were two sided and all variables with p < 0.05 were considered significant at 95% confidence interval.