Understanding patterns of family support and its role on viral load suppression among youth living with HIV aged 15 to 24 years in southwestern Uganda

Abstract Background Active family support helps as a buffer against adverse life events associated with antiretroviral therapy (ART) uptake and adherence. There is limited data available to explain how family support shapes and affects individual healthcare choices, decisions, experiences, and health outcomes among youth living with HIV (YLWH). We aimed to describe family support patterns and its role in viral load suppression among YLWH at a rural hospital in southwestern Uganda. Methods We performed a mixed‐method cross‐sectional study between March and September 2020, enrolling 88 eligible YLWH that received ART for at least 6 months. Our primary outcome of interest was viral load suppression, defined as a viral load detected of ≤500 copies/mL. Data analysis was performed using Statistical Package for Social Sciences version 20. Fifteen individuals were also purposively selected from the original sample and participated in an in‐depth interview that was digitally recorded. Generated transcripts were coded and categories generated manually using the inductive content analytic approach. All participants provided written consent or guardian/parent assent (those <18 years) to participate in the study. Results Forty‐nine percent of YLWH were females, the median age was 21 (IQR: 16‐22) years. About half of the participants (53%) stayed with a family member. A third (34%) of participants had not disclosed their status to any person they stayed with at home. Only 23% reported getting moderate to high family social support (Median score 2.3; IQR: 1.6‐3.2). Seventy‐eight percent of YLWH recorded viral load suppression. Viral load suppression was associated with one living with a parent, sibling, or spouse (AOR: 6.45; 95% CI: 1.16‐16.13; P = .033), having a primary caretaker with a regular income (AOR: 1.57; 95% CI: 1.09‐4.17; P = .014), and living or communicating with family at least twice a week (AOR: 4.2; 95% CI: 1.65‐7.14; P = .003). Other significant factors included youth receiving moderate to high family support (AOR: 12.11; 95% CI: 2.06‐17.09; P = .006) and those that perceived family support in the last 2 years as helpful (AOR: 1.98; 95% CI: 1.34‐3.44; P = .001). HIV stigma (AOR: 0.10; 95% CI: 0.02‐0.23; P = .007) and depression (AOR: 0.31; 95% CI: 0.06‐0.52; P = .041) decreased viral load suppression. Qualitative data showed that dysfunctional family relationships, economic insecurity, physical separation, HIV‐ and disclosure‐related stigma, past and ongoing family experiences with HIV/ART affected active family support. These factors fueled feelings of abandonment, helplessness, discrimination, and economic or emotional strife among YLWH. Conclusion Our data showed that living with a family member, having a primary caretaker with a regular income, living or communicating with family members regularly, and reporting good family support were associated with viral load suppression among YLWH in rural southwestern Uganda. Experiencing depression due to HIV and or disclosure‐related stigma was associated with increased viral load. All YLWH desire ongoing emotional, physical, and financial support from immediate family to thrive and take medications daily and timely. Future interventions should explore contextual community approaches that encourage acceptance, disclosure, and resource mobilization for YLWH who rely on family support to use ART appropriately.


| INTRODUCTION
Young people (15-24 years old) constitute the most significant population in the world and, at the same time, represent an age group with one of the highest new HIV infections. 1 Young people are also at a greater risk of AIDS-related deaths, discrimination, marginalization, exclusion, poor antiretroviral therapy (ART) adherence, and the lowest utilization of health care. 1,2 According to several studies, young people are faced with persisting barriers that negatively impact their access to services such as family or spousal consent requirements, family, economic and structural factors, social support, difficulties in transitioning from pediatric care to self-management, criminalization against vital young populations, age, inadequate health systems, early and forced marriages, and a lack of appropriate sexual education. 1,[3][4][5] In Uganda, individuals who are 10 to 24 years of age comprise 33% of the whole population and they account for the highest number of the country's HIV/AIDS cases. 6 Most youth in Uganda are financially and emotionally dependent on their families, and studies have shown that the main barriers affecting their ART uptake and adherence include unreliable social support, change in guardianship, poverty, HIV-and disclosure-related stigma, school attendance limiting their privacy, loss to follow-up, drug side effects, and substance abuse, among others. [7][8][9] Social support has been documented to improve medication adherence through emotional support (psychological and informational) and instrumental support (physical and economic), helping to overcome many physical, structural, and financial barriers to access care in time. 10 Support from close family members could also make individuals feel a sense of security and belonging, facilitating them to overcome significant physical hurdles such as food insecurity, housework, child care, and transport challenges to access ART. 11 Healthy family support is therefore vital for every individual's wellbeing, acting as a source of stability, happiness, empathy, encouragement, connection, and a platform to express how one feels, celebrate good experiences, and talk about challenging times. 12 This active family support also provides members with the much-needed sense of identity that could ultimately help enhance the quality of life and adherence by providing a necessary buffer against adverse life events.
However, negative family social support, socio-cultural perception of HIV disease, stigma, poor relationships from unsupportive family members, non-communication, mistrust, resentment, abandonment, or loose family ties can affect the pattern of health-seeking, pill-taking behavior, coping mechanism against HIV-related stigma, and wellbeing of an individual as a whole. 10,13,14 Several studies have attempted to explore the effect of social support on medication adherence. 10,15 However, limited data explain how support from family members explicitly shapes and affects individual healthcare decisions, experiences, and health outcomes among youth living with HIV (YLWH). This study aims to describe patterns of family support and how it affects viral load suppression among youth aged 15 to 24 years in southwestern Uganda.

| Study design and setting
We conducted a cross-sectional study to describe the patterns of family support and its effect on viral load suppression among youth aged 15 to 24 years in southwestern Uganda. The study was conducted at the Kinoni Health Center IV, a publically funded and operated health center in Rwampara, a rural, resource-limited district located in southwestern Uganda. The health center serves over 100 000 patients annually from across 20 villages. It provides general outpatient care, maternal and child health care, inpatient care, general surgery, laboratory diagnostics, and HIV care services to both children and adults.

| Study population
The study was conducted among YLWH between 15 and 24 years of age, both male and female, registered at the ART clinic of Kinoni Health Centre IV.

| Sampling procedure and recruitment
We selected all study participants who had attended the HIV/ART clinic for the last 6 months as per the facility records assisted by the nurse-in-charge for the required age group. We enrolled youth A subset of 15 YLWH was purposively selected from the total enrolled individuals for qualitative interviews based on study participant's family relationship dynamics, viral load outcome, ART enrolment duration, experience, HIV disclosure status, social support characteristics, and variations in the types and quality of social support provided by family members. In addition, participants were invited to come to the clinic or private research space alone. The interviews aimed at gathering in-depth information from specific participants with characteristics that would help explain their experiences and the role of family support on their ART uptake and utilization.

| Data collection
Participants completed a structured interviewer-led questionnaire with data on known explanatory factors that affect ART uptake and adherence: socio-demographics, health and depression, 16 HIV serostatus disclosure, food insecurity, 11 alcohol use, 17 HIV stigma, 18 and social support. 19 The questionnaire also contained sections on the last viral load recorded over the previous 6 months, relationship and communication with members at home, income, disclosure status, presence of an HIV-positive family member, the number of people at home, and family support. Our primary outcome of interest was viral load suppression, defined as a viral load of less than or equal to 500 copies/mL. We expressed the quality of family support as emotional (psychological and informational) and instrumental (physical and economic) social support obtained from family members using a standardized score ranging from 1 to 4, 19 with 4 indicating high levels of social support.
An in-depth interview was administered to 15 purposively selected YLWH, which explored family and primary caretaker relationships, ART and disclosure experiences, pill-taking behavior, food insecurity, and type and variations of family support. The interview guide was developed using the Health Utilization Model (HUM). 20 All qualitative interviews were conducted within 1 week of participant enrolment by two trained RAs and were digitally recorded with the participant's permission in the native local language (Runyankole) in a comfortable and private location within the Health Center premises or a mutually agreed personal space. Interviews lasted between 45 minutes to 1 hour. The recorded interviews were transcribed from the local language directly to English by a well-trained RA.

| Data analysis
We considered all the 88 YLWH who completed all study procedures.
We described demographic and clinical data for the enrolled participants using standard descriptive statistics. We assessed participant correlates of poor viral load suppression of viral load ≤500 copies/mL, computed for each participant. All data analysis was performed using the Statistical Package for Social Sciences (SPSS) version 20. We estimated the P values with Chi-squared tests utilizing a level of statistical significance of ≤.05. Continuous variables were summarized using medians and interquartile ranges. We used univariate logistic regression to assess unadjusted associations between covariates and viral load suppression, and these were expressed using crude odds ratio (OR) and 95% confidence intervals (CIs). We tested the variables for collinearity. Those with a P value of less than or equal to .10 in unadjusted analyses were included in a multivariable logistic regression analysis, adding one at a time to control for confounders.
Interviews were transcribed in English and coded manually. Coding was jointly done by ECA and RN. Together with others, disagreements in coding were resolved to ensure consistency. We reviewed the coded data to identify repeated patterns and sorted them to derive categories using the inductive content analytic approach. 21 We aimed to construct categories describing individual healthcare experiences, relationship dynamics, involvement and perspectives of family to support healthcare decisions, ART uptake and utilization, as well as barriers and challenges that affect their well-being. Themes were then generated from the categories identified and presented with illustrative quotes from the participants' interviews to explain how these relationships and support-or lack thereof-shape their healthcare decisions and access to and utilization of ART care.  This study was initiated in March 2020 and ended in September 2020. A total of 88 study participants were enrolled and interviewed.

| Ethics and approval
The median age was 21 (IQR: 16-24) years, and almost half of this population was aged 23 to 24 years, with 68% having attained at least primary education ( Table 1). Half of the interviewed youth (51%) were male. Fifty-seven percent resided more than 5 km away from the health center where they accessed their routine ART care. Sixty-eight percent of these youth reported some form of employment. About   Table 2). Participants who were assessed with no depression recorded higher levels of viral load suppression (94% vs 57%, P = .002). Participants who recorded no to low HIV stigma levels also had better viral load suppression (94% vs 32%, P = .011). The majority of the participants who stayed with their parent, sibling, or spouse registered a higher viral load of ≤500 copies/mL per category than those who stayed with friends, alone, or other relations such as employer (87% vs 68%, P = .031). Better viral load suppression was observed for participants who had disclosed to any family member (86% vs 63%, P = .013). Participants reporting moderate to high family support registered higher viral load suppression rates than those with low to no support (87% vs 65%, P = .009). Higher rates of viral load of ≤500 copies/mL were also reported among participants who perceived family support as helpful (87% vs 65%, P = .001).
Bivariate analyses identified several factors that were associated with viral suppression (

| Qualitative findings
Our qualitative data showed that youth desired ongoing social and family support to cope with their daily emotional, physical, and economic needs. Most participants depended on family for food, shelter, housework, encouragement, and financial support to refill and take medications daily and timely as needed. Our data also showed the fol-

| Dysfunctional relationships affecting regular communication and contact
Most participants reported dynamic relationships at the individual, family, and community levels that changed over time, affecting their emotional and physical well-being. Many participants lived and depended on their relatives, spouses, or friends for food, money, housework, shelter, advice, or emotional support. Others stayed alone because of the death or separation of parents, family conflicts, and their perceived need to be independent and start their own families.
Participants often expressed a great desire for continuous physical and emotional support from their close friends, parents, and siblings to constantly communicate challenges, seek advice, encouragement, and cope with disclosure effects. The support rendered also included helping them financially or physically to pick and take their medicines on time. Owing to the inability to get the desired support from immediate family due to mistrust, dysfunctional relationship, and communication, some participants chose to date or get married early to partners perceived to financially and emotionally support them.
According to some participants, these complex and dysfunctional relationship arrangements often led to separation, fueling feelings of abandonment, emotional strife, helplessness, self-neglect, low self-esteem, anxiety, and depression. Some participants, for example, who stayed with partners who did not know their HIV sero-status reported experiences of bearing abuse, mistreatment, and loneliness within these relationships to remain and get financial and physical provisions.
These challenges often affected pill-taking behavior, as individuals struggled to pick up their refills on time. One of the participants with a high viral load, who also reported a low family support score, said:

| Economic insecurity and physical separation affecting resource mobilization
Families helped participants to mobilize basic needs such as food, transport, personal needs, and shelter. Individuals who were entirely dependent on economically secure families reported no worries concerning feeding and upkeep. They also reported experiencing emotional support, as they did not consider themselves a burden to other family members and consistently picked and used their medication as advised by their healthcare providers. However, some participants reported food scarcity, which affected their pill-taking schedules, and economic strife, which led them to missing medicine refill dates due to  3.6 | Fear of disappointment, judgment, and discrimination from family affecting active support   Previous studies have documented the importance of family dynamics in the mental health of these young adults, indicating that the familial context in which a person with HIV on ART resides is interconnected with their health outcomes. 22 It has also been shown that providing people with information regarding the myths and the usefulness of ART improves their adherence. 23 Other studies have shown a high rate of ART adherence among the youth with HIV from cohesive families. 12 Therefore, the youth mainly rely on their family's social support to enable them to follow the treatment plan offered and cope emotionally and financially with having to take their medication correctly and on time. 24 Good relationships and social support also improve early linkage to health services, positive living, HIV status disclosure, and self-acceptance of the HIV status. 25  our data also showed that the lack of proper food or proper and timely food preparation affected ART use, as individuals tried to avoid the side effects of ART.
Some studies have observed that males, especially the adolescent youth, are less likely to test and or disclose their HIV status to others, including close family members, because of the fear of being discriminated against. 7,27 These young people also avoided disclosure to those outside their homes because of perceived stigma and discrimination. 28 On the other end, disclosing one's HIV status and receiving acceptance and social support from close relations were associated with improved long-term quality of life among the youth. 15 Our data showed that individuals who had acquired HIV at birth experienced automatic exposure to family members over time and obtained the necessary family support to continue disclosing to significant others. Participants who disclosed earlier had the needed continuous HIV care over time, especially among those whose family relationships were good. Incoherent family relationship, on the other hand, facilitated emotional strife and fear of victim-blaming following disclosure, and this affected pill-taking behavior, regular refills, and adherence among participants. As previously reported, 29 the lack of physical and emotional support from close families of the HIVinfected youth significantly affected their mental and physical wellbeing.

| Strengths and limitations
Our study had several strengths. First, ours was one of the few studies, to the best of our knowledge, that used a mixed-method approach to document the patterns of family support and its primary role in viral load suppression among YLWH. The location being in rural south- HIV and ART were more aware of the usefulness of ART and seemed to provide the support individuals needed for ongoing medication uptake and use. The absence of good family support facilitated feelings of abandonment, economic or emotional stress, anxiety, helplessness, desperation, and low perceived usefulness for ARV drugs.
A contextual understanding of community needs and factors that provide an enabling environment to suppress viral load among YLWH is needed to maximize their mental well-being and ART clinical treatment outcomes. In addition, future studies should explore group family HIV/ART counseling and a community awareness approach to encourage acceptance, resource mobilization, and disclosure for groups who greatly rely and thrive on active family support.

ACKNOWLEDGMENT
We are grateful to the District Health Officer, Rwampara District, the in-charge, and the staff of Kinoni Health Center for accepting to work with us and the respondents who freely shared their experiences with us.

CONFLICT OF INTEREST
All authors declare no conflict of interest. All the authors read and approved the final version of the manuscript.

TRANSPARENCY STATEMENT
I, Dr. Esther Cathyln Atukunda affirms that there has not been any important aspects of the data that have been left out intentionally. I also affirm that the data herein included in this manuscript is accurate and transparent, and that I had full access to all of the data in the study. I take complete responsibility for the integrity of the data and the accuracy of the data analysis.

DATA AVAILABILITY STATEMENT
All data has been included in this manuscript. Any additional data will be available on request.