Assessing Hiv Behavioral Risks and Their Associated Socio- Demographic Factors Among Adolescents: Survey Findings From Hiv Risk Education Program in Nimule Peri-urban Town, South Sudan

doi:10.21203/rs.3.rs-4130001/v1

Download PDF

Research Article

Assessing Hiv Behavioral Risks and Their Associated Socio- Demographic Factors Among Adolescents: Survey Findings From Hiv Risk Education Program in Nimule Peri-urban Town, South Sudan

https://doi.org/10.21203/rs.3.rs-4130001/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

INTRODUCTION: As the world advances toward HIV epidemic control, gains made are being slowed down by an increase in new HIV infections particularly among adolescents whose risky behaviors remain a public health concern. This study aimed to assess self-reported HIV behavioral risks and their associated socio-demographic factors among adolescents in Nimule peri-urban town in South Sudan.

METHODS: We designed and undertook before and after the study with two cross-sectional surveys conducted at baseline and end line for adolescents aged 10-17 years who consented, enrolled, and participated in a peer-led HIV risk education program delivered in health clubs. Using a standardized adolescent survey questionnaire, data on ten (10) self-reported HIV risks including demographic factors of the respondents were collected between December 2020 and December 2021. A multi-level mixed-effect logistic regression analysis was performed using STATA version 16 statistical software to assess the associations between self-reported behavioral risks and associated sociodemographic factors.

RESULTS: Among the 577 adolescents enrolled, 301 (54.0%) were female and 256 (46.0%) were males. The median age was 14 (IQR: 11-16) years at baseline and 15(IQR: 12-17) years at end-line. A slight decrease from 89 (16.0%) to 88 (15.8%) was observed between baseline and end-line in the proportion of adolescents who self-reported that they have never had sex in their lifetime with a corresponding increase from 58 (64.4%) at baseline to 70 (78.6%) at the end line in the proportion of adolescents self-reporting having used condoms in their last sex. Compared with the baseline survey, the odds of self-reported HIV risks were 23% lower (aOR 0.77, 95% CI:0.73-0.80, p<0.001). Equally, higher odds of self-reported risks were observed among male adolescents (aOR 1.11, 95% CI:1.06-1.15, p<0.001), those in the age band 15-17 years (aOR 1.27, 95% CI:1.21-1.34, p<0.001), those who were unemployed (aOR 1.2, 95% CI: 1.21-1.34, p<0.001) and those cared for by non-parents (aOR 1.29, 95% CI:1.21-1.36, p<0.001)

CONCLUSION: Peer-led risk program education programs have the potential to reduce HIV risk behaviors among adolescents and this signifies the crucial role these programs can play in accelerating HIV prevention for young adolescents. We, therefore, recommend the integration of risk education into HIV prevention programs for adolescents in low- and middle-income countries like South Sudan

HIV

Risk Behaviors

Risk Education

Adolescents

South Sudan

Globally and in Sub-Saharan Africa in particular, HIV and AIDS continue to span for generations making millions of adolescents vulnerable to new HIV infections with a slower from 2.1 million in 2010 to 1.7 million in 2015, (Granich et al., 2015). With Decades of devastating armed conflict, South Sudan has a fragile health system with limited access to HIV prevention services(Jervase et al. 2018). According to UNAIDS global HIV and AIDS country statistics (2020), South Sudan estimated national HIV prevalence was estimated at 2.9% with 17,000 new infections in 2021, representing an increase from 23% in 2021 to 35%. High HIV prevalence was also reported among sexually exploited adolescents in Nimule Town(Okiria et al. 2023). Unlike other countries in the region, a limited level of awareness about HIV behavioral risks and prevention methods has been observed as a major barrier to effective prevention programs in South Sudan(Okiria et al. 2023). A high prevalence of HIV and syphilis has been identified to impede effective HIV prevention programs among adolescents in South Sudan (Culbreth et al. 2020; Aguirrebengoa et al. 2020). Other studies conducted in 10 African countries revealed disproportionately wide gaps in HIV prevention characterized by increased transactional sex, limited condom use, and uptake of HIV testing services among refugee adolescent girls and young women in Uganda have also been reported(Logie et al. 2021; Spiegel, Schilperoord, and Dahab 2014). Key interventions such as risk education programs and economic empowerment for adolescents have been shown to reduce risk behaviors among adolescents (Fasil et al. 2022; Team 2012). While risk education programs are being implemented, these programs have not been extensively evaluated in South Sudan. This study is designed to assess HIV behavioral risks and their associated socio-demographic factors among adolescents in Nimule per-urban town in South Sudan

Study design and participants

We undertook a before and after study of adolescents with two cross-sectional surveys conducted at baseline and end line before and after delivery of comprehensive HIV risk education program and 24 months of follow-up of the study participants. This study design and its conceptual framework (Fig. 1) were largely guided by the programmatic worldview philosophical paradigm derived from the work of Peirce, James, Mead, and Dewey (Cherryholmes, 1992) to bring about the desired positive change of the study on adolescent risk behaviors. In addition, we employed the theoretical models of change including the Theory of planned behavior (TPB), the Theory of reasoned Action (TRA), and the Health Belief Model (HBM) to model adolescent risk behaviors during the implementation of the risk education program. The study population included adolescents aged 10–17 years recruited from 17 residential areas including Motoyo East, Motoyo West, Kololo East, Kololo West, Rock City, Melijo, Hai Kanisa, Matara, Abila, Malakia West, Malakia East, Longia, Bio II, Jeli, Nimule Central, Olikwi and Rei within Nimule Peri-urban town. Located at the border between South Sudan and Uganda, this study area was suitably selected due to its high population density comprising mainly internally displaced persons (IDPs) and the host community. Due to its increasing commercial activities, Nimule is home to female sex workers (Okiria et al. 2023).

Sample size estimation

We used the WHO (1991) cluster sampling strategy to estimate the proportion of adolescents who were knowledgeable about HIV prevention methods (S. K. Lwanga and S. Lemeshow, 1991). We estimated that 768 adolescents would be required to participate in the study to detect an increase in the relative level of knowledge of 40% or greater at a 5% significance level with 80% power, assuming baseline knowledge of 50% and a design effect of 2

Study variables

The outcome variable “self-reported HIV risks” was measured using ten (10) risk indicators which include; the proportion of adolescents who self-reported at baseline and end-line survey that they had ever had sex or intention to have sex in the next 30 days, ever used a condom during last sexual intercourse, ever experienced any signs and symptoms such as genital sores, discharge or burning urination in the past 30 days, ever slept outside their homes in the past four weeks, ever been sexually touched in the genitals by a person of the opposite sex in the past four weeks, ever consumed any alcoholic drink in the past 30 days, intends to drink alcohol in the next 30 days, ever smoked any tobacco products or intends to do so in the past 30 days. Those who responded at least “yes” to any one of the ten questions were regarded as having HIV risk behaviors while those who responded “no’ were regarded as having no risk. Key socio-demographic variables collected included sex, education, marital status, having a caregiver, employment, ownership of a phone, and health status rating.

Participants recruitment: Using the cluster sampling technique, all eligible adolescents aged 10–17 years from HIV-affected households were purposively recruited into the study. Here all caregivers and adolescents currently enrolled in ART were pre-identified from the Antiretroviral treatment registers in Nimule Hospital. A list of households where these adolescents and caregivers live was generated and assigned to trained Research Assistants to visit. From this list, Research Assistants randomly selected households for visits where they administered an informed and assent form (Appendix 2) to each caregiver of the adolescent. All adolescents whose caregivers provided assent were recruited into the study. This procedure was performed until the sample size (n) was reached.

Inclusion and exclusion criteria

Inclusion criteria: The study included adolescents who fit the following criteria: -

Aged 10–17 years
Currently enrolled in ART at Nimule Hospital
Living in an HIV-affected household
Currently resident in Nimule town for the past six months
Caregiver provided assent to participate in the study

Exclusion criteria: The following adolescents were excluded from the study: -

Below 9 years and above 17 years
Not residing in Nimule Town
Adolescents whose caregivers decline to assent

Study procedures

The following procedures and steps were taken to recruit the study participants: -

Step 1: Pre-identification: Here, trained Case Care Workers (CCWs) working with clinical staff pre-identified and generated a list of all households for adolescents and caregivers currently enrolled in antiretroviral treatment (ART) in Nimule Hospital. This list was later used by the CCWs to visit these households and enroll all eligible adolescents in the study

Step 2: Enrollment: At this stage, CCWs booked home visits, conducted home visits to all pre-identified households, introduced the study to caregivers, and administered the assent form (Appendix 2). All eligible adolescents aged 10–17 years were enrolled in the study and assigned an alphanumeric unique identification code (UIC). All participants' enrollment and assent were securely kept under key and lock by the study Monitoring and Evaluation officer

Step 3: Baseline survey: Using a standardized adolescent Health survey questionnaire (Appendix 3), a baseline survey was conducted from December, 1st to 30th, 2020. Adolescent sociodemographic data on age, sex, marital status, employment, and education including self-reported behavioral risks were collected. These data were analyzed and stored.

Step 4: Delivery of planned interventions and follow-up: Interventions were delivered to both adolescents and their primary caregivers within their intervention groups. While adolescents were targeted with risk education sessions, their primary caregivers were targeted with positive parenting training, financial literacy training, and cash transfer. These interventions are detailed below: -

Savings and Internal Lending (SILC) groups

These groups were formed for caregivers and out-of-school adolescents to participate in small-scale businesses, and save and lend money to group members. By doing so, out-of-school adolescents and their caregivers had increased employment and income and were able to meet their basic needs without relying on gifts. Formed voluntarily, members completed 10 days of financial literacy training and received quarterly seed capital to invest in their businesses.

Financial literacy training

This intervention targeted primary caregivers and adolescents out of school adolescents 34 savings groups each consisting of about 6–12 members. Each CCW supervised about 6–7 groups. Using a standardized financial literacy training curriculum developed by UNDP, 400 primary caregivers, and out-of-school adolescents completed 10 days of training and developed a business plan. After completing the training, trainees voluntarily formed their Savings and Internal Lending Groups (SILCs) with each group supported by a case care worker. Group members nominate their leaders (Chairperson, Secretary, and Treasurer) and adapt their constitution and group norms. Group members actively engaged in small businesses and met weekly to collect mandatory savings and lend money to group members. During group meetings, case workers would invite healthcare providers to deliver HIV health education, and treatment literacy, refill ARVs, and provide HIV testing services and referrals to facilities. Data on household employment, income, spending, and food availability in the households were collected to assess adolescent household economic resilience at baseline and end line.

Household cash transfer

This intervention was to support caregivers and out-of-school adolescents to access financial credit facilities to improve their businesses. After completing the financial literacy training and forming groups, each caregiver and adolescent received a quarterly disbursement of $50 cash conditioned on starting a small-scale business and being a member of a registered savings and lending group within the study area.

Cash disbursements were done via the mobile money platform ( M-Gurush) where each caregiver received a free telephone SIM card. All individual transactions such as withdrawals and deposits were monitored quarterly. By injecting cash into adolescent’s households, caregivers and out-of-school adolescents would become finically resilient by creating self-employment opportunities by engaging in small-scale businesses such as making tea, bakery, and selling fruits and vegetables to increase their household income and meeting basic needs of adolescents such as education, food, and health care.

Positive parenting groups

This intervention was delivered for both primary caregivers and adolescents. Using a standardized curriculum adopted from Parenting for Life Long Health (PLH); a South African-based parenting organization with a footprint in delivering parenting programs globally to reduce violence against adolescents by fostering positive relationships between caregivers and adolescents. The curriculum aligned consisted of 14 modules delivered by trained CCWs. The session delivered included but was not limited to sexuality education, dealing with adolescents’ risk behaviors such as sexual relationships, alcohol, and substance use, how to discipline adolescents in a non-violent way, and Antiretroviral viral drug use. The sessions attended by both caregivers and adolescents were playful involving singing, dancing, storytelling, and physical exercises. This made adolescents and caregivers actively engage in the sessions and increase their coping capacity with psychological distress. A total of four hundred primary caregivers and 400 adolescents completed the training. Other adolescents who didn’t attend the sessions were then reached by their caregivers and adolescents who completed the sessions

Adolescent risk education. Adolescents recruited into peer-led health clubs competed in HIV risk education sessions. Adapted from the South Sudan Ministry of Health comprehensive sexuality education curriculum for adolescents which comprises of 4 modules and 31 sessions were delivered by trained Peer educators for 3 months. The facilitators delivered 3 sessions per week. Each session consists of a demonstration, activity, and games lasting for about 20–25 minutes. Key messages in the sessions included risky sex, consistent and correct condom use, identifying signs and symptoms of sexually transmitted infections and how to access screening and treatment services, alcohol, and substance use, and menstrual hygiene promotion. The delivery of sessions took into consideration cultural norms such as observing privacy when discussing sexuality education and delivering separate sessions for boys and girls to minimize bullying tendencies between boys and girls. An HIV risk screening tool (Appendix 2) was used to screen and identify teenagers who were at high risk of contracting HIV and STIs, such as those who reported having had sex, being pregnant, or showing signs and symptoms of STIs. These adolescents were then referred to health facilities so they could receive HIV prevention and treatment services.

Supportive home visits and referrals

In addition to the risk education delivered within the health clubs, CCWs conducted monthly home visits aimed at supporting caregivers to provide HIV risk education to their adolescents and supporting them in acquiring HIV and STI testing and treatment. During these visits, CCWs support adolescent referrals to access HIV/STIs testing, remind adolescents and caregivers to honor their facility appointments, and conduct ARV pills count to identify and provide treatment literacy.

Working with the implementing partners and health facilities, specific services were mapped a comprehensive bi-directional referral protocol was developed, and a referral form to facilitate referrals. In addition, these home visits were aimed at monitoring adherence to treatment and viral load suppression for adolescents on ART. Working closely with the ART clinic in Nimule Hospital, CCWs using treatment literacy materials developed by the South Sudan Ministry of Health conducted assessed adherence to treatment using the ART treatment adherence assessment form, identified specific treatment adherence challenges such as stock out of ARVs, missing taking ARVs and provide treatment literacy to ensure compliance to the recommendation of their health care provider. In addition, CCWs also identify high viral load adolescents including their caregivers, and refer them for viral load testing and complete enhanced adherence support counseling to maintain low viral load. In addition, CCWs collected viral load results after every 3 months. This intervention was aimed at increasing retention of ART (2nd. 95) and viral suppression (3rd. 95) for adolescents and their caregivers on ART.

Step 5: End-line survey: An end-line survey was conducted in December 2022 after the delivery of planned interventions where all adolescents who completed the study completed the adolescent Health survey questionnaire at the end-line survey.

Data collection

We used a structured, pre-tested adolescent health survey English script questionnaire (Appendix 3) to collect data at both baseline (December 2020) and end-line (December 2022). This questionnaire was validated by comparing it with a similar questionnaire used in a South African study to assess the knowledge, attitudes, and practices of young people toward HIV prevention(Shamu et al. 2020).

The questionnaire was pre-tested for consistency and validity with minor adjustments made before data collection. Pre-testing of the questionnaire was done using a sample of 100 adolescents who were not part of the study. Minor edits on the wording of the questions and flow were made on the questionnaire before data collection. During training, Data collectors were trained to correctly translate the questions into local languages, including Arabic, Madi, Acholi, and Dinka. Data were collected by trained data collectors. During training, emphasis was put on role-playing interviews using the local languages (Arabic, Madi, Acholi, and Dinka) widely spoken in Nimule Town. During data collection, data collectors booked appointments with adolescent caregivers and agreed on visiting times. Each data collector was assigned a target of 4–5 questionnaires per day, based on pre-testing where each questionnaire took approximately 40–50 minutes to complete. The Monitoring and Evaluation and Research Officer conducted supportive supervision and randomly selected at least 10% of completed questionnaires from each data collector to review for consistency and completeness. Inconsistent and incomplete questionnaires were returned to the respective data collectors and corrected before they were entered into the Excel master survey database. These quality checks heightened the reliability of the data.

Data analysis

Both the baseline and end-line data sets initially captured in the Excel database were exported into STATA version 16 for data cleaning and final analysis. We merged the baseline with the end-line dataset where encoding and de-stringing of character variables were then conducted for categorical and numerical variables respectively to ensure they were compatible with the analysis software.

Some variables recorded with attributes of categorical variables with few observations were combined to avoid being dropped off by STATA during analysis. Both the baseline and end-line data were analyzed and results were compared to assess the self-reported HIV risk behaviors and their associated socio-demographic factors between the two points

Test for normality

The graphical approach (histogram), and numerical approach (Shaphiro Wilk test) were conducted to test for the normal distribution of variables. The age variable was not normally distributed and thus we reported its median (IQR)

Test for multicollinearity

We conducted a multicollinearity test among all the independent variables using the correlation matrix and variance inflation factor method, and for any two variables that were collinear (r > 0.3 or VIF > = 10) one of such variables, especially the one with higher p-values at bivariate analysis was excluded from the analysis.

Univariate analysis

We conducted univariate analysis for all the socio-demographic variables and individual outcome variable indicators to check their distribution at baseline and end line. Results were presented in tabular form

Bivariate analysis for paired data

This was conducted between observation of the outcomes collected at baseline and those at the end line to test whether they were significantly different. McNamar’s test for paired data was used to report the proportions of observations and their respective p-values.

Bivariate analysis for non-paired data

We used the chi-square test to assess the significance of each of the participant categorical social demographic variables and the outcome variable.

Any variable that showed statistical insignificance (p < 0.25) was not proceeded with to the multivariable level of analysis. Binary logistic regression method was applied between each of the independent and outcome variables generating the odds ratios and confidence intervals as measures of associations.

Multivariate analysis

We proposed that all individual variables that showed p < 0.25 at bivariate analysis be included in multivariable level analysis. A stepwise model-building approach using a binary logistic regression analysis technique was used to arrive at the final model, the Akaike Information Criterion (AIC) was used to select the best-fit model, where the model with the lowest AIC was preferred. We reported adjusted odds ratios and corresponding p values. P values of < 0.05 were regarded to be statistically significant. We used multiple imputations to mitigate bias as a result of missing data resulting from selection bias during recruitment, where participants who agreed to participate may be different in characteristics from those who declined to participate.

Ethical considerations

This study received ethical clearance from the South Sudan Ministry of Health Research and Ethical and Review Board (MOH/RERB/24/2020) and the Strathmore University Institutional Ethics and Review Committee (SU-IERC1287/22) as indicated in Appendix 1a and Ib respectively. All procedures were put in place to uphold respect for autonomy, informed assent, beneficence, and confidentiality. Study participants were de-identified using unique identifier codes (UICs). A risk-benefit analysis matrix was undertaken to address any distress among study participants as a result of the interview process.

Socio-demographic characteristics of respondents: Of the 768 participants enrolled, 211 (27.5%) immediately relocated to Ugandan as refugees fleeing ethnic conflict before the baseline survey was conducted thereby reducing the total number of participants who completed the baseline survey and retained in the study to 557 (73%). At the end line survey, all 557 participants responded to the survey questionnaire giving a retention and response rate of 100% for this study. Of the 557 participants who completed the end-line survey, the majority (54%) were female, none of whom had attained tertiary education at baseline. The median age of the participants was 14 (IQR: 11–16) years at baseline. (Table 1).

Table 1

Socio-demographic characteristics of respondents at baseline and end line (n = 557)
Indicator	Baseline	End line
	n (%)	n (%)
Gender
Female	301(54.0)	301(54.0)
Male	256(46.0)	256(46.0)
Age Group
10–14 years	328(58.9)	266(47.8)
15–18 years	229(41.1)	291(52.2)
Median age (IQR) Years	14(11–16)	15(12–17)
Education
None	7(1.3)	7(1.3)
Primary	427(76.7)	411(73.8)
Secondary	123(22.1)	116(20.8)
Tertiary	0(0.0)	23(4.1)
Reading proficiency
Can’t read at all	99(17.8)	99(17.8)
Can partly read	174(31.2)	174(31.2)
Read the whole sentence	284(51.0)	284(51.0)
Owns a phone
Yes	121(21.7)	121(21.7)
No	436(78.3)	436(78.3)
Current Marital status
Never been married	557(100.0)	557(100.0)
Employment status
Employed	430 (77.2)	430 (77.2)
Unemployed	127 (22.8)	127 (22.8)
Health rating
Poor/Fair	242(43.3)	232(41.6)
Good	184(32.9)	308(55.2)
Excellent	131(23.8)	17(3.1)
Household role
Cooking	155(28%)	155(28%)
Babysitting	44(8.0%)	44(8%)
Fetching water	261(47.0%)	260(47%)
Others	97(17%)	98(18%)
Primary caregiver
Father	237(42.5%)	237(42.5%)
Mother	225(40.4%)	225(40.4%)
Brother/sister	39(7.0%)	39(7.0%)
Uncle/aunt	56(10.1%)	56(10.1%)

HIV behavioral risks among respondents

Across the ten proxy indicators measured, respondents self-reported reductions in their HIV risks and risk intentions, between baseline and end line, across five proxy indicators There was a slight decrease from 25 (3.8%) at baseline to 20 (3.6%) at end-line in the proportion of respondents who self-reported that they had ever slept outside their homes in the past four weeks. There was an increase from 58 (64.4%) at baseline to 70 (78.6%) at the end line in the proportion of adolescents self-reporting having used condoms in their last sex. While 71 (9.7%) respondents self-reported the intention to have sex in the next 30 days at the baseline survey, none self-reported such intention at the end line. There were decreases from baseline to the end line in the proportion of respondents self-reporting their intention to consume any alcoholic drink in the next 30 days from 9 (1.6%) to 2 (0.4%) and smoke any tobacco products in the next 30 days from 10 (1.8%) to 0.

Unexpectedly, the proportion of respondents who self-reported ever having had sex decreased from 89 (16%) at baseline to 88 (15.8%) at the end line. This suggests that one respondent either fabricated their response or misunderstood the question at one of the survey time points. Disappointingly, the proportion of respondents self-reporting that they had ever consumed any alcoholic drink increased from 74 (13.6%) at baseline to 76 (13.6%) at the end line, indicating that some respondents consumed alcohol for the first time during the span of the intervention program. The proportion of respondents self-reporting having ever smoked tobacco products, having been touched in the genital by a person of the opposite sex, and having experienced any signs and symptoms of STIs in the past 30 days were all unchanged between baseline and end line. (Table 2).

Table 2

Association between HIV risks and risk intentions indicators at baseline and end-line survey (n = 557)
Indicator	Baseline	End line	McNemar’s P-value
Indicator	n (%)	n (%)
Experienced STI signs and symptoms in the past 30 days
No	539(96.8)	539(96.8)
Yes	18(3.2)	18(3.2)	1.00
Slept outside home any time in past 4 weeks
No	532(96.2)	537(96.4)
Yes	25(3.8)	20(3.6)	0.50
Touched in the genitals by a person of the opposite sex
No	523(94.0)	523(94.0)
Yes	34(6.0)	34(6.0)	1.00
Ever had sex
No	468(84.0)	469(84.2)
Yes	89(16.0)	88(15.8)	< 0.001
Used condom during last sex
No	31(35.6)	18(21.4)	1.00
Yes	58(64.4)	70(78.6)
Intend to have sex in the next 30 days
No	486(90.3)	557(100.0
Yes	71(9.7)	0(0.0)	< 0.001
Ever consumed any drinks that contain alcohol
No	483(86.7)	481(86.4)
Yes	74(13.3)	76(13.6)	< 0.001
Intend to drink alcohol in the next 30 days
No	548(98.4)	555(99.6)
Yes	9(1.6)	2(0.4)	0.02
Ever smoked tobacco products
No	519(93.2)	519(93.2)
Yes	38(6.8)	38(6.8)	< 0.001
Intend to smoke tobacco products in the next 30 days
No	547(98.2)	557(100.0)
Yes	10(1.8)	0(0.0)	0.002

Multilevel model analysis of factors associated with self-reported HIV risks and risk intentions

The odds of self-reported HIV risks and risk intentions adjusted for baseline and end-line surveys were 23% lower at the end line compared to baseline (aOR 0.77, 95% CI:0.73–0.80, p < 0.001). Male participants had 11% higher odds of self-reporting HIV risks and risk intentions compared to their female counterparts (aOR 1.11, 95% CI:1.06–1.15, p < 0.001). Participants aged 15–17 years had 27% higher odds of self-reported HIV risks and risk intentions compared to those aged 10–14 years (aOR 1.27, 95% CI:1.21–1.34, p < 0.001). The odds of reporting HIV risks and risk intentions were about 20% higher among the unemployed participants compared to the employed (aOR 1.2, 95% CI: 1.21–1.34, p < 0.001), and about 29% higher among participants who were being taken care of by non-parents compared to those with parent caregivers. (aOR 1.29, 95% CI:1.21–1.36, p < 0.001). (Table 3).

Table 3

Multilevel model analysis of factors associated with HIV behavioural risks (n = 557)
Indicator	No Risk	At Risk	P-value	AOR [95% Cl)
	n (%)	n (%)
Survey
Baseline	374(42.4)	183 (78.9)	1	1
End line	508(57.6)	49(21.1)	< 0.001	0.77(0.73–0.80)***
Sex
Female	515(58.4)	87 (37.5)
Male	367(41.6)	145(53.5)	< 0.001	1.11 (1.06–1.15)***
Age category
10–14 years	538(61.0)	56(24.1)	1	1
15–17 years	344(39.0)	176(75.9)	< 0.001	1.27 (1.21–1.34)***
Owns a phone
Yes	153(17.4)	89(38.4)		1
No	729(82.7)	143(61.6)	0.87	0.87 (0.94–1.05)
Employment status
Employed	723(82.0)	137(59.1)		1
Unemployed	159(18.0)	95(40.9)	< 0.001	1.20 (1.15–1.26)***
Caregiver
Parent	780(88.4)	144(62.1)		1
Non-parent	102(11.6)	88(37.9)	< 0.001	1.29 (1.21–1.36)***
***Significance at p < 0.001

This study revealed five key findings. Firstly, study participants self-reported lower odds of HIV risks and risk intentions across multiple indicators at the end line compared to a corresponding baseline survey. The indicators showing significant reduction at the end-line survey included sleeping outside the home in the past four weeks, using a condom in the last sex, intention to have sex, intention to consume any alcoholic drinks, and intention to smoke any tobacco products in the next 30 days. This observed decrease may be attributed to the impact of the peer-led risk education delivered as well as other supportive services. Similar studies using health education (“Effects of a Sexuality Education Programme on Young People’s STI/HIV Knowledge, Attitudes and Risk Behaviour in Northeast Nigeria - Wamanyi Yohanna, Olaoluwa Samson Agbaje, Osmond Chukwuemeka Ene, Perpetua Chinyere Ofili, Prince Iheanachor Christian Umoke, 2023,” n.d.), positive parenting (Hertzog et al. 2022)], and household economic strengthening – all of which have also revealed a positive impact on adolescent risk reduction. In the present study, the package of interventions was strengthened by targeting both the adolescents and their primary caregivers. In the study conceptual framework, the come variable (reduction in self-reported risk and risk intention) was linked to selected interventions delivered to both caregivers and adolescents to bring about positive change. In addition, the development of adolescent case care plans involving both the adolescents and caregivers ensured a more holistic approach to bring about the change adolescents and their caregivers want to see happen(Williams et al. 2021). These plans covered various aspects such as HIV prevention methods, risk identification, HIV testing, retention on treatment, and viral load suppression. We further enhanced the risk education provided to adolescents by training caregivers in positive parenting programs. This program gives caregivers core competencies in dealing with adolescents’ risk behaviors more positively.

The program also empowered caregivers to competently discuss sexuality education with their adolescents. In addition, the provision of cash transfers and financial literacy training provided caregivers the financial capacity to provide their adolescents with their basic needs such as education, health, and food. These basic needs have been identified as factors predisposing adolescents to risky behaviors such as exchanging sex for gifts(Boyer et al. 2017). Secondly and unexpectedly, the socio-demographic variable of gender (male), was significantly associated with higher odds of self-reported HIV risks and risk intentions. By contrast, other studies in the context of the HIV epidemic showed females at increased risk and vulnerability to HIV infections (Mabaso et al. 2021). We associate this disparity with male adolescents feeling free to self-report their risks and risk intentions to the study interviewers. In addition, male study participants may have been more susceptible to losing the knowledge transmitted through the HIV risk education delivered immediately after enrolment into the study, or they may have less access than female study participants to additional information on mitigation of HIV risks and risk intentions. Our findings of increased male risk differ from other studies conducted in Sub-Saharan Africa, where adolescent girls working in bars were found to have higher HIV risk behaviors (Dambach et al. 2020). Similar studies in South Sudan and Zambia also found an increased risk of HIV among females (Okiria et al. 2023; Hakim et al. 2020; Sserwanja et al. 2022). This unusual finding has shaped our recommendations (below) for how interventions could be designed to target both males and females.

Thirdly, the higher odds of self-reported HIV risks and risk intentions by adolescents in the age band 15–17 years as compared to younger respondents aged between 10–14 years was expected, and could be associated with the generally lower literacy level and awareness of HIV infection risk factors such as having sex, alcohol, and s substance use in South Sudan. These adolescents in the age band 15–17 may have acquired greater motivation to experiment with their sexuality by taking more risks as compared to younger adolescents. Higher risky behaviors such as sex without using a condom and substance use were found among similar age groups in studies in Bangladesh and Nigeria (Bossonario et al. 2022; Hasan et al. 2013; Bolarinwa 2019).

Fourthly, the expected higher odds of self-reported risks and risk intentions among the unemployed respondents could be associated with them having less money and therefore, reduced negotiating power and resources to procure and consume alcohol, and tobacco products. Conversely, the unemployed likely lacked access to HIV information delivered at the workplace. Unemployment has been identified to increase the odds of higher HIV risk behaviors among adolescents in Kenya(Juma et al. 2013). Conversely, household socioeconomic empowerment has been observed to reduce risk behaviors and improve health outcomes for adolescents(Adu-Mireku 2003; Cavazos-Rehg et al. 2020; Luseno et al. 2019)

Fifthly, the higher odds of HIV risks and risk intentions observed among respondents living under the care of a non-parent, compared to living under the care of a parent, was expected and could be associated with several factors. Adolescents living under the care of non-parents were more likely to experience communication barriers to freely discussing intimate subjects of sexuality and HIV risk education. Adolescents living without their parents have been shown to have a greater susceptibility to engaging in risky behaviors, such as sex in exchange for gifts and accompanied by alcohol and substance use(Bermudez et al. 2019; Culbreth et al. 2020; Folayan, Sam-Agudu, and Harrison 2022).

Study limitations

Our findings should be considered in light of the limitations associated with this study. Our findings should be strictly considered in the light of adolescents aged 10–17 years old and residing in Nimule town. Other limitations include the influence of the 211 participants who were lost to follow-up during the study. In addition, our findings might have been influenced by inconsistency in self-reporting risk and risk intentions by the adolescents. Equally, the use of broad strata within our categorical variables may have led to some residual confounding. Finally, selection bias in recruiting participants who consented to participate might be different from decline.

Study strengths

While this study was bound to the above limitations, it is worth highlighting its uniqueness and strengths. Delivering adolescent age-appropriate peer-led risk education programs and also delivering intervention for their primary caregivers with economic strengthening and positive parenting training ensured that both the adolescents and their caregivers were empowered with information. Although 73% of the study participants were lost to follow-up at the recruitment phase, the study managed to retain 100% of those who completed the baseline survey through the end-line survey.

The peer-led risk program education program has the potential to reduce HIV risk behaviors among adolescents in peri-urban dwellings. This underscores the crucial role this program can play in accelerating HIV and STIs prevention outcomes among adolescents. We therefore recommend the integration of risk education programs in existing and future HIV prevention programs for adolescents

ACKNOWLEDGEMENTS

This study acknowledges supervisors, partners, and collaborators including the DAAD Africa regional office in Nairobi; Accelerating Achievement for Africa’s Adolescents (Accelerate) Hub led by Prof. Lucie Cluver, the UNDP Accelerator Lab team in South Sudan led by Jacquelin Poni; Dr. Alex Dimitri of the South Sudan Ministry of Health; UNFPA; Centre for Research and Development in Population Health (CRDPH), South Sudan; Agency for Research and Development Initiative (ARDI), South Sudan; Mgurush Mobile Money Transfer agency, South Suda; Parenting for Lifelong Health ( PLH); University of Cape Town; Clowns Without Borders South Africa (CWBSA); International Programme for AIDS Care and Treatment Programmes; IntraHealth International; Nimule Hospital. Many thanks are accorded to Louise Gordon, Accelerate Hub at the University of Oxford, for editing the manuscript. Many thanks to my supervisors Prof. Ambrose Agweyu and Prof. Gilbert Kokwaro who provided technical guidance in writing all sections of this manuscript. Finally, the technical support of Nixon Ayena in data analysis is very much acknowledged.

Consent for publication: All participant information presented in this research paper doesn’t contain any personal identifiers of study participants and as such its publication doesn’t in any way compromise the personal anonymity of study participants

Funding: The authors hereby declare that this research was funded by the United Kingdom Global Challenges Research Funds (ES/S008101/1), the United Nations Development Program Accelerator Lab (00117753 ACC LAB), and the German Academic Exchange Service (57499642). This funding strictly covered the implementation of the study and didn’t include any publication-related costs.

Availability of data and materials: The data set “DATA SET_SSD_2022” used to generate the statistical values presented in this paper has been uploaded and can be accessed as a file in the supporting information section.

Conflict of Interests Declaration: The authors hereby declare no competing commercial or financial conflict of interest that could have affected the integrity of this study

Ninrew CS. Sudan unlikely to attain Zero HIV Infections by 2030 - official warns [Internet]. Eye Radio. 2022 [cited 2023 Jun 21]. Available from: https://www.eyeradio.org/s-sudan-unlikely-to-attain-zero-hiv-infections-by-2030-official-warns/.
South Sudan Ministry of Health. World Population Day [Internet]. 2013. Available from: https://eyeradio.org/south-sudan-adolescent-pregnancy-highest-globally/.
HOULIHAN CF, LARKE NL, WATSON-JONES D, SMITH-MCCUNE KK, SHIBOSKI S, GRAVITT PE, et al. HPV infection and increased risk of HIV acquisition. A systematic review and meta-analysis. AIDS. 2012;26(17). 10.1097/QAD.0b013e328358d908.
Culbreth R, Swahn MH, Salazar LF, Ametewee LA, Kasirye R. Risk Factors Associated with HIV, Sexually Transmitted Infections (STI), and HIV/STI Co-infection Among Youth Living in the Slums of Kampala, Uganda. AIDS Behav. 2020;24(4):1023–31.
Aguirrebengoa OA, Garcia MV, Sanchez MR, D´Elia G, Méndez BC, Arrancudiaga MA, et al. Risk factors associated with sexually transmitted infections and HIV among adolescents in a reference clinic in Madrid. PLoS ONE. 2020;15(3):e0228998.
Hill LM, Maseko B, Chagomerana M, Hosseinipour MC, Bekker LG, Pettifor A, et al. HIV risk, risk perception, and PrEP interest among adolescent girls and young women in Lilongwe, Malawi: operationalizing the PrEP cascade. J Int AIDS Soc. 2020;23(S3):e25502.
Okiria AG, Achut V, McKeever E, Bolo A, Katoro J, Arkangelo GC et al. High HIV and syphilis prevalence among female sex workers and sexually exploited adolescents in Nimule town at the border of South Sudan and Uganda. PLOS ONE [Internet]. 2023 [cited 2023 May 26];18(1). Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9847914/.
Mbatha ZP. Investigating the effectiveness of orphans and vulnerable children (OVC) programs in schools: a case of Ntuzuma G-section in Durban. [Internet] [Thesis]. 2014 [cited 2023 Feb 17]. Available from: https://researchspace.ukzn.ac.za/handle/10413/13668.
Ward P, Hurrell A, Visram A, Riemenschneider N, O’Brien C, MacAuslan I et al. Cash transfer program for orphans and vulnerable children (CT-OVC), Kenya operational and impact evaluation, 2007–9. 2010.
Bermudez LG, Yu G, Lu L, Falb K, Eoomkham J, Abdella G, et al. HIV Risk Among Displaced Adolescent Girls in Ethiopia: the Role of Gender Attitudes and Self-Esteem. Prev Sci. 2019;20(1):137–46.
Bossonario PA, Ferreira MRL, Andrade RL, de Sousa P, de Bonfim KDL, Saita RO. Risk factors for HIV infection among adolescents and the youth: a systematic review. Rev Latino-Am Enfermagem. 2022;30:e3697.
Fasil N, Worku A, Oljira L, Tadesse AW, Berhane Y. Association between sexual and reproductive health education in the peer group and comprehensive knowledge of HIV among adolescent girls in rural eastern Ethiopia: a community-based cross-sectional study. BMJ open. 2022;12(10):e063292.
School-based sex education among adolescents worldwide. Interventions for the prevention of STIs and unintended pregnancies | British Journal of Child Health [Internet]. [cited 2022 Dec 8]. Available from: https://www.magonlinelibrary.com/doi/abs/10.12968/chhe.2022.3.5.229.
Hasan AH, Hassan R, Khan ZR, Nuzhat E, Arefin U. Influence of socio-demographic factors on awareness of HIV/AIDS among Bangladeshi garment workers. SpringerPlus. 2013;2(1):174.
Bolarinwa O. Knowledge of sexually transmitted infections and socio-demographic factors affecting high-risk sex among unmarried youths in Nigeria. Int J Glob Health. 2019;2(1):2.
Bolarinwa O. Knowledge of Sexually Transmitted Infections and Socio-Demographic Factors Affecting High-Risk Sex among Unmarried Youths in Nigeria. 2019;1.
Dambach P, Mahenge B, Mashasi I, Muya A, Barnhart DA, Bärnighausen TW, et al. Socio-demographic characteristics and risk factors for HIV transmission in female bar workers in sub-Saharan Africa: a systematic literature review. BMC Public Health. 2020;20(1):697.
Juma M, Alaii J, Bartholomew LK, Askew I, Van den Born B. Understanding orphan and non-orphan adolescents’ sexual risks in the context of poverty: a qualitative study in Nyanza Province, Kenya. BMC Int Health Hum Rights. 2013;13(1):32.
Thurman TR, Nice J, Luckett B, Visser M. Can family-centered programming mitigate HIV risk factors among orphaned and vulnerable adolescents? Results from a pilot study in South Africa. AIDS Care. 2018;30(9):1135–43.

Appendix 2 is not available with this version.

No competing interests reported.

Download PDF

Editorial decision: Revision requested
07 Nov, 2024
Reviews received at journal
09 Jul, 2024
Reviews received at journal
05 Jul, 2024
Reviewers agreed at journal
02 Jul, 2024
Reviewers agreed at journal
01 Jul, 2024
Reviewers agreed at journal
27 Jun, 2024
Reviewers invited by journal
20 Jun, 2024
Editor invited by journal
22 Mar, 2024
Submission checks completed at journal
21 Mar, 2024
Editor assigned by journal
21 Mar, 2024
First submitted to journal
19 Mar, 2024

You are reading this latest preprint version

Assessing Hiv Behavioral Risks and Their Associated Socio- Demographic Factors Among Adolescents: Survey Findings From Hiv Risk Education Program in Nimule Peri-urban Town, South Sudan

Status:

Version 1

Abstract

Figures

INTRODUCTION

MATERIALS AND METHODS

Inclusion and exclusion criteria

Study procedures

RESULTS

HIV behavioral risks among respondents

Multilevel model analysis of factors associated with self-reported HIV risks and risk intentions

DISCUSSION

Study limitations

Study strengths

CONCLUSIONS

Declarations

ACKNOWLEDGEMENTS

References

Appendix 2

Additional Declarations

Supplementary Files

Status:

Version 1