The data of the adolescent study of Dodoma Health and Demographic Surveillance System (HDSS) were used for the current study. Dodoma HDSS is a longitudinal study that was developed by the University of Dodoma (UDOM) in collaboration with Africa Academy for Public Health (AAPH), the Harvard T. H. Chan School of Public Health, and the Dodoma Regional Health Management Team with the intent of providing information on longitudinal data regarding socio-demographic and health of the individuals.
Study design and set up of the Dodoma Adolescent Study
A community-based cross-sectional study among adolescents was done in Dodoma rural at Chamwino District from April to June 2017. The District has a total population of 330,543 individuals of which 71,208 are adolescents . The survey was nested within the platform of the HDSS site in Chamwino, Dodoma. The site involved two wards with a total of five villages. The HDSS site has registered 5256 households with a population of 23785 and adolescents aged 10-19 years are 6162. The adolescent survey included adolescents aged 10-19 years.
Sample size and sampling technique
The current study adopted the sample size and sampling procedures of the Dodoma adolescents study. The sample size was obtained through the Cochran formula holding these assumptions: 50% prevalence (p) of malnutrition, 5% desired precision and adolescent population (N) of 6162 individuals. Based on the given information, a minimum sample size required was 362 adolescents. To avoid non-response that could occur during data collection, an adjusted sample size of 517 respondents was calculated by using assuming a minimum response rate of 70%. However, the sample size was collected beyond the expectation and the final sample size obtained was 1,226 adolescents. The respondents of the study were selected from a stratified sample of the adolescent population of Dodoma HDSS. The SAS PROC SURVEY SELECT procedure was employed to select the sample while villages were the sampling strata.
Survey administration and instrumentation
Dodoma adolescents' study used a standardized structured questionnaire across seven countries.
The questionnaire was amended in tablets and administered in electronic form. The questionnaire
included variables of socio-economy, socio-demography, food security, teeth and hand wash, food diversity, feelings (mental health) and friendship, school, physical activity, and home activities, physical fighting, physical attacks, sexual practice, injuries, health status assessment, health care, life satisfaction, substance use, exposure to sexual explicit materials, pregnancy, media use, sexual transmitted infection, and nutrition status.
Open Health and Demographic Surveillance (Open HDS) data system was used to collect data. This uses android based tablets to enable the collection of data. This web-based application allows research assistants to electronically register and collect information from households. The information is then linked to android based tablets through wireless synchronization and then sent to the central data store based on the Open Data Kit (ODK) [24, 25].
The current study mainly focused on 3 of the questionnaire-units of the Dodoma adolescent survey namely sexual practice, mental health, and exposure to sexual explicit materials. The items from the questionnaire units are presented as follows:
This module consisted of four items: “Have you ever had sexual intercourse?”,” For the first time you had sexual intercourse, how old were you? ” “The last time you had sexual intercourse, did you or your partner use a condom?” and “, with how many people have you had sexual intercourse during your life time?” To assess the magnitude of sexual contact, which is the outcome variable for this study, a dichotomous answer to item one “Have you ever had sexual intercourse” was used. The response options were 1=yes and 2=no. The response was then recoded as 1=yes and 0=no.
Psychosocial distress was measured across 4 mental health indicators of distress: anxiety or worried, loneliness, sadness, and attempt to suicide. The items options were as follows:
Loneliness: “During the past four months, how often have you felt lonely?” (The options were: 1=never, 2= rarely, 3= sometimes, 4=most of the time, and 5=Always). The response were recoded as 1 = most of the time or always and 0 = never, rarely, or sometimes.
Anxiety or Worried: During the past four months, how often have you been so worried about something that you could not sleep at night? (Response options were 1=never, 2= rarely, 3= sometimes, 4=most of the time, and 5=Always.) (Coded 1 = most of the time or always and 0 = never, rarely or sometimes).
Sadness: Over the last week, how frequently have you experienced, depressed, low mood, sadness, and feeling blah or down, just couldn't be bothered? (Response option were: 1 = hardly ever, 2= sometimes, 3=much of the time, 4=most of the time, to 5= all of the time.). The responses were recoded as 1 = much of the time, most of the time or all of the time, 0 = Hardly ever or sometimes.)
Suicide attempt: “Have you ever seriously consider attempting suicide during the last 12 months?” (Response option was 1 = yes and 2 = no, coded 1 = 1, 2 = 0).
The total frequency of psychosocial distress indicators was computed to determine if the respondent had 0, 1, 2, and 3 or 4 indicators. Subjects with 1 or more psychosocial distress indicators were considered to be psychosocially distressed.
Exposure to sexually explicit material
This module included 5 items: “Have you ever watched a sexually explicit film?” “With whom do you usually watch sexually explicit films?”, “How often did you watch such materials in the last month?”, “where do you find the sexually explicit material you watched?” and “Have you ever sent /received a sexually suggestive text or picture?” The subject was considered to be exposed to sexually explicit material if he/she ever watched a sexually explicit film or ever sent or received a sexually suggestive text or picture.
All statistical analysis procedures were conducted using the SAS Version 9.4 software. To make statistics computed from the data more representative of the adolescents’ population for each village in Dodoma HDSS, the sampling weights were used in data analysis. The sampling weights were computed as the inverse of selection probabilities. In the results, weighted percentages were reported. The contribution of psychosocial distress and sexually explicit materials on sexual intercourse, while adjusting for social-demographic variables and alcohol use was evaluated using weighted multiple logistic regression models. The strength of association was construed through the adjusted odds ratio (AOR) and p-value. A bivariate weighted logistic regression model was first used for assessment of the crude impact of explanatory variables. All variables with p-values less or equal to 0.25 in bivariate analyses were counted in the multivariable models. The criterion for statistical significance was set at a p-value of 0.05.
Ethics approval and consent to participate
Informed verbal consent of participation was sought from both parents and each adolescent before measurements or interviews. The ethical clearance was sought from the ethical review board of the University of Dodoma. To ensure adherence to ethics and safety, confidentiality was maintained at all times. An informed consent form that included all elements of informed consent was translated into Kiswahili, a language that is well understood by the participants of the study.