Menu
Menu
Menu
Menu
Background: Adolescents are a priority group in HIV prevention and treatment. This study sought to determine the prevalence and correlates of HIV testing services (HTS) among adolescents in the pastoralist post-conflict area of Karamoja sub region, Uganda. Methods: A cross sectional study of 1439 adolescents aged 10-19 years, attending nine public health facilities in five of the seven districts of Karamoja, was conducted between August to September 2016. Adolescents were consecutively selected and interviewed using structured interviewer administered questionnaires. All respondents who had never tested for HIV were offered HTS. The main outcome was ever tested for HIV. Correlates of ever tested were analysed using multivariate logistic regression model. Results: Of the 1439 adolescents, 904 (62.8%) were females, 1203 (83.6%) were aged 15-19 years, 618 (43.0%) had attained primary education and 885 (61.5%) had ever had sex. Overall 1177 (81.8%) had ever tested and received HIV results. Older age (15-19 years) (adj.OR = 2.71, 95% CI: 1.85-3.96), secondary level education or higher (adj.OR = 2.33, 95% CI: 1.33-4.10), and ever had sex (adj.OR = 2.03, 95% CI: 1.42-2.90) were associated with higher odds of HIV testing. Of the 262 who had never tested, 169 (64.5%) accepted testing and 2.4% were HIV positive. Reasons for not accepting the test included fear of being tested and not ready for an HIV test because of perceived suffering HIV positive clients go through. Conclusion: Awareness of HIV status and uptake of HTS among adolescents in this hard-to-reach post-conflict region was high and close to the global UNAIDS target of 90%. However, the HIV prevalence of 2.4% among the non-testers who accepted to be tested was high and emphasises the need for targeted testing to reach the undiagnosed HIV infected adolescents in this region.
Karamoja sub-region is located in the north eastern part of Uganda and is occupied by a pastoralist community that is dependent on animals for survival and security. In this setting, males do much of the animal rearing while females do the housework. This society has over the years had conflicts with other tribal neighbourhoods fighting for land, water and animals, a practice that has of recent stabilised when the government initiated a disarmament program and provided other logistical and humanitarian support in terms of food, shelter, health care and education. According to the 2011 Uganda AIDS indicator survey, general population HIV prevalence in Karamoja was 3.5% [25]—prevalence among female adolescents 15–24 years was 3.5% compared to 2.6% in their male peers [26]. The UNICEF annual report of 2013 showed Karamoja lagging behind compared to western Uganda and Acholi regions regarding access to HTS and antiretroviral therapy (ART) for prevention of mother to child transmission (PMTCT) [27]. This was a cross-sectional survey involving 1439 adolescents (10–19 years) receiving primary health care services at the outpatient department (OPD) and maternal child health (MCH) clinics at public health facilities between August to September 2016. Bennett’s sample size formula [28] was used considering health facilities as the clusters and expected daily number of adolescents attending the facilities estimated at 140. Since the coverage of HIV testing among adolescents in this region was not known at the time of planning this study, a conservative prevalence of 50% and a design effect of 2.0 were used to yield a minimum sample size of 1375 participants from nine clusters. The highest volume facilities within the region—facilities that contributed 80% of all OPD attendance for adolescents in the previous year were sampled. The 80% mark was chosen to allow adequate representation of the districts and facilities within the region to be included in the study. The 9 facilities were located in five of the seven districts in the sub-region. The number of adolescents to be interviewed at each of these nine facilities was then determined using probability proportional to size (PPS), based on the numbers of adolescents who sought HTS in the prior quarter (April–June 2016) before data collection. Within each facility, adolescents were consecutively sampled until the required number per facility was obtained. All adolescents were approached for participation irrespective of their reason for coming to the facility. Ethical approval for this study was obtained from Makerere University School of Biomedical Sciences Higher Degree Research and Ethics Committee (SBSREC) and the Uganda National Council for Science and Technology (UNCST) before data collection. Additionally, permission was sought from district and health facility heads. After the selection, adolescents were screened for eligibility. Written informed parental or guardian consent and assent were obtained from adolescents < 18 years while those ≥18 years consented before enrolment into the study. Adolescents < 18 years who came to these facilities without a guardian or parent were advised to come back with a parent/guardian who could consent on their behalf. Parents/guardians and the adolescents were informed that the study included an assessment of knowledge and access to HIV services and that some would be offered an HIV test and tested if they accepted to do so. Respondents were interviewed by trained research assistants using semi-structure questionnaires and a modified HIV knowledge tool. All respondents were asked if they had ever tested for HIV and received their results. Those who had never tested where asked if they would like to be tested for HIV. Those who accepted were linked to the HTS sites within the health facilities. Based on the institutional policies, all research assistants signed a confidentiality agreement before data collection to ensure confidentiality of respondent’s results. Those who refused to test for HIV were asked for the reasons, and these were documented through open ended questions. All questionnaires were checked by field supervisors, for quality control. On a daily basis, the completed questionnaires were collected by a regional supervisor who kept them locked in an office. Data was entered into an access database, cleaned from spreadsheets. The reasons for refusal to test for HIV among those who had never tested were also coded and entered into the database. Clean data was exported to Stata statistical software version 13.0 for analysis. Data collected included; adolescents’ socio-demographic characteristics, HIV testing and receipt of results, knowledge of HIV prevention and transmission, knowledge of partner’s HIV status, engagement in high-risk sexual behaviours, history of having children, ever had sex, and use of substances or drugs. The main outcome in this study was “ever tested” for HIV which was coded as 1 for “Yes” and 0 for “No”. Independent variables included; socio-demographic characteristics like sex, age group, marital status, HIV knowledge (Yes = 1, No = 0), engaging in high-risk sexual behaviours (Yes = 1, No = 0), and use of drugs or other illicit substances (Yes = 1, No = 0). Sex was coded as (Female = 1, Male = 0), Age group coded as (15–19 years = 1 and 10-14 years = 0), Ever had sex coded as (Yes = 1, No = 0), Education level coded as (Nursery = 0, Primary = 1, ≥ secondary = 2). During testing for significant covariates; marital status was coded as; Never married = 0, Married/Cohabiting = 1 and Divorced/Separated = 2. HIV sero-status of their most recent partner was coded as (Yes = 1, No = 0), while the number of children was coded as (None = 0, 1–3 = 1). The HIV knowledge score was based on an aggregate score obtained by using a KQ-18 HIV questionnaire [29] modified to suit the cultural context for the study population. In line with other literature that considered mean and median scores for cut-off scores [30], participants in our study who scored above or equal to the median score of 11 were considered to have adequate HIV knowledge. Adolescents were considered to have engaged in high-risk sexual behaviours if they inconsistently used a condom and either had multiple sexual partners (two or more sexual partners) or having engaged in transactional sex in the last 6 months. Descriptive statistics for ever tested were presented as frequencies and percentages. A chi-square test was used to elicit associations between individual characteristics with HIV testing. Odds ratios were generated using a multivariate logistic regression model to elicit associations with HIV testing. We adjusted for a number of variables to include; sex of the adolescent, highest level of education attained, ever had sex, HIV knowledge, number of children the adolescent has ever had, knowing the HIV status of their sexual partners and high-risk sexual behaviours. We excluded marital status as an independent variable in the final model because of potential multi-collinearity with “ever had sex”. Variables with a p-value of 0.2 and below at bivariate analysis were entered into the multivariate models. Model parsimony was ensured by using the backward stepwise modelling and the likelihood ratio test between the full and restricted models. The model that yielded the highest variability in explaining the predicted variable (ever tested) was considered as the best fit. All analyses were conducted using STATA v.13 (College Station, TX).
N/A
