The relationship between life satisfaction, personal health, quality of life, and medication adherence among adolescents living with HIV in southwestern Uganda

Abstract

Purpose

Adolescents living with HIV (ALHIV) are at high risk for adherence to antiretroviral (ART) treatment and poor health-seeking behaviors, both of which potentially compromise their quality of and satisfaction with life. This study, therefore, seeks to examine the relationship between physical health, pediatric quality of life, life satisfaction, and medication adherence among ALHIV in southwestern Uganda.

Methods

Baseline data from the Suubi+Adherence study, 2012–2018, that recruited 702 adolescents, aged 10–16 years, living with HIV in Uganda were analyzed. To account for overdispersion, negative binomial regression analyses were used to examine the impact of physical health, pediatric quality of life, and life satisfaction on self-reported medication adherence. We controlled for participants’ socio-demographic factors.

Results

Results indicated that after adjusting for socio-demographic characteristics, adolescents’ reported satisfaction with life was associated with a decrease in the reported number of days missed taking medication by 41.2% (IRR (incidence risk ratio)=0.588; p (p-value)=0.014). On the other hand, the low level of pediatric quality of life was associated with a 5% increase in the reported number of days missed taking medication (IRR=1.055, p=0.044). Personal health was not statistically significant in the model.

Conclusion

Our study findings indicated that quality of life and life satisfaction are significantly associated with antiretroviral (ART) medication among ALHIV. Hence, strengthening existing support systems and creating additional support for optimal ART adherence and treatment outcomes for ALHIV in low-resource communities might be beneficial. Moreover, with the increasing HIV prevalence rates among adolescents, effective and comprehensive efforts that are responsive to the special needs of ALHIV must be developed to ensure optimal adherence to ART medication as it leads to low vertical infection and superinfection rates.

Introduction

Adolescents living with HIV (ALHIV) are at high risk for antiretroviral (ART) adherence to treatment and poor health-seeking behaviors, which can lead to compromised life satisfaction and quality of life (Habere & Mellins 2009; Lowenthal et al. 2014; Vreeman et al. 2008). Adolescents are at a higher risk of engaging in risky behaviors, such as unprotected sex, which makes them more susceptible to HIV and other sexually transmitted infections (Lowenthal et al. 2014). In 2018, approximately 190,000 adolescents aged 10–19 years were newly infected with HIV worldwide ([The United Nations International Children’s Emergency Fund] UNICEF 2019) and it is predicted that without concerted investment, adolescents will account for approximately 183,000 new HIV infections annually by the year 2030 (UNICEF 2019). These estimates undermine the new UNAIDS efforts to eradicate HIV (WHO 2019). An estimated 25.7 million people are living with HIV in sub-Saharan Africa (SSA) ([The Joint United Nations Programme on HIV/AIDS] UNAIDS 2019). In this region, nearly 90% of adolescents live with HIV (Agwu and Fairlie 2013; Lowenthal et al. 2014; UNICEF 2019). Despite the scale-up of antiretroviral (ART) medication availability among adolescents in SSA, adherence to ART remains a challenge (Nabukeera-Barungi et al. 2015; Vreeman et al. 2008). For example, in a study among 702 ALHIV aged 10–16 years in Uganda, Bermudez and colleagues (2018) found that only 59% of the participants were virally suppressed.

Adolescents and young adults have the highest rates of non-adherence to ART in Uganda. In a survey of 30 HIV health clinics in Uganda, Nabukeera-Barungi and colleagues (2015) found that 90.4% of adolescents had reported a less than 95% rate of adherence during their most recent visit. In addition, adolescents based in rural communities were more likely to report non-adherence to ART medication. Therefore, the ALHIV population may require special considerations regarding the management of HIV/AIDS (Naswa &Marfatia 2010).

With the scale-up of ART in the early 2000s, more children living with HIV are able to transition into adolescence while managing HIV as a chronic disease (Audureau et al. 2013; Habere & Mellins 2009). Indeed, studies have shown that timely initiation and adherence to ART is associated with positive HIV-related health outcomes. Specifically, maintaining an optimal adherence rate of greater than 90% has been linked to increased viral load suppression, CD4+ cell count, reduced morbidity and mortality, and improved physical health and pediatric quality of life among patients living with HIV (Chandwani et al. 2012; Firdu et al. 2017; Habere & Mellins 2009; Vyankandondera et al. 2013).

Previous research indicates that social support, especially from family members, is associated with better adherence rates and adherence self-efficacy among adolescents (Damulira et al. 2019; Habere & Mellins 2009; Nabunya et al. 2020). In addition, support groups, counseling, healthcare workers who have positive attitudes, and the provision of food and transportation have been found to promote adherence to ART medication among adolescents (Adejumo et al. 2015; Ammon et al. 2018; Hudelson & Cluver 2015; Kunihira et al. 2010; Nabukeera-Barungi et al. 2015). However, even with advances in HIV treatment, difficulties associated with ART initiation and adherence among ALHIV persist (Habere & Mellins 2009; Nabukeera-Barungi et al. 2015). Several factors contribute to suboptimal adherence, including complex regimens and heavy pill burden, poverty, forgetfulness, undiagnosed and/or untreated mental health issues, and the inability to disclose HIV status out of fear of being ostracized by peers and impacting potential love interests (Chandwani et al. 2012; Damulira et al. 2019; Habere & Mellins 2009; Lowenthal et al. 2014). Suboptimal adherence can also lead to a lower quality of life among ALHIV. On the other hand, perception of a good quality of life is important for adherence and chronic disease management among adult populations and may have important implications for promoting medication adherence among adolescents (Fernandez-Lazaro et al. 2019).

Moreover, quality of life, physical health, and life satisfaction have a dynamic reciprocal relationship with ART adherence. For instance, adolescents are more likely to begin HIV medication feeling weak, and over time are able to maintain ART adherence, which in turn improves their overall health. However, paradoxically as their health improves, their adherence to ART medication decreases (MacCarthy et al. 2018; Nabukeera-Barungi et al. 2015). In other words, as ALHIV start to feel good and physically strong, they may forget to take their medication or decide not to keep to their prescribed treatment regimens (MacCarthy et al. 2018). In addition, most ALHIV have lost their parents to HIV-related complications, which may contribute to poor mental health functioning (MacCarthy et al. 2018). This is especially noteworthy as poor mental health is associated with non-adherence to ART medication that in turn has adverse impacts on overall health (Sin & DiMatteo 2014).

Similarly, living with HIV as a chronic illness as well as its associated comorbidities may lower the quality of life for ALHIV, which, in turn, reduces adherence to ART medication (MacCarthy et al. 2018). Given that suboptimal adherence is associated with the development of drug resistance, accelerated disease progression to AIDS, and increased mortality among adolescents, strategies to address suboptimal adherence in this population are critical (Hudelson & Cluver 2015; Kim et al. 2014; Slogrove et al. 2018; Usitalo et al. 2014).

Overall, there is a scarcity of literature that investigates the role of personal health and pediatric quality of life as predictors of ART adherence, especially among ALHIV in SSA (Blashill et al. 2013; Silva et al. 2014). In order to help fill this gap, this study examines the relationship between personal health, life satisfaction and pediatric quality of life, and self-reported adherence among ALHIV in southwestern Uganda.

Methodology

We utilized baseline data from the Suubi+Adherence study, 2012–2018 (Grant No. R01HD074949) funded by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD). The primary aim of the study was to examine the impact, and cost associated with a family economic empowerment intervention, such as matched child savings accounts, financial and health education and income-generating activities training, on adherence to ART among ALHIV in rural Uganda. In the current study, we used baseline data to examine the relationship between personal health, life satisfaction, pediatric quality of life, and self-reported adherence to HIV medication.

Study sample

Overall, a total of 702 adolescents living with HIV, aged 10–16 years at study initiation, participated in the study. To participate in the study, they had to fulfill three criteria, namely living with HIV as confirmed by a medical report, living within a family and not an institution, prescribed ART, and receiving care at one of the 39 accredited health clinics located in the Greater Masaka region. Data were collected by trained research assistants using interviewer-administered surveys, with each lasting approximately 90 minutes. A full description of the recruitment procedures has been published elsewhere (Ssewamala et al. 2019).

Ethical considerations

The study received ethical clearance from the Columbia University Institutional Review Board (#AAAK3852), Makerere University School of Public Health Review Board (Protocol #210), and the Uganda National Council of Science and Technology (UNCST, SS 2969). Caregivers provided written consent for their children to participate and children provided written assent separately, to avoid coercion. All consent forms and study-related materials were provided in both English and Luganda, the local language spoken in the study area. Prior to engaging with participants, the study staff received training on the protection of human subjects and good clinical practices to ensure the safety and well-being of study participants. All eligible children in the family and at the study clinics were recruited, provided they met the inclusion criteria.

Measures

The outcome of self-reported adherence was captured by the following question: “In the last 30 days, how many days did you miss at least one dose of your HIV medications?” with high numbers indicating higher rates of non-adherence to ART medication. The main independent variables were life satisfaction, physical health, and quality of life. Participants were asked to indicate how satisfied they were with their life overall on a 5-point Likert scale ranging from 1 = not at all satisfied to 5 = extremely satisfied. For physical health assessment, two items were used. Participants were asked to rate their health at present (1 = very poor to 5 = excellent), whether they had low energy (1 = never to 5 = almost always). For this analysis, these items were categorized as binary variables. For instance, “not satisfied at all” and “not satisfied” were combined into “not satisfied” and the rest of the three categories as “satisfied.” To measure the quality of life, four items were adapted from the Pediatric Quality of Life Inventory that assesses children’s physical health as it relates to their experience at school (Thompson et al. 1987). Specifically, participants rated their quality of life on a 5-point Likert scale, ranging from 5 = always to 1 = never. The theoretical range was 4–20, with lower scores indicating better quality of life.

To adjust for potential confounding, socio-demographic characteristics were included in the analyses as control variables. These included participants’ age, number of adults and children in the household, sex, availability of a medication supporter (i.e. someone to help them take or pick up their medications), and type of primary caregiver. We also recognize that other factors may influence adherence to ART medication among ALHIV, for example, stigma. The data used in this study was selected using a true randomization process and therefore, issues of endogeneity and confounding were already addressed by the randomized nature of the study.

Data analysis

All analyses were conducted using Stata Statistical Software Version 15 while adjusting for the socio-demographic characteristics of the study population. First, we conducted univariate analyses to describe the sample. Since the outcome variable was a highly skewed count variable, bounded by zero, and measured in whole numbers, it violated most of the assumptions embedded in linear regression (Long & Freese 2014). For this study, therefore we used negative binomial regression. However, for sensitivity analysis, we ran both Poisson and negative binomial regression to predict the relationship between physical health, quality of life, and life satisfaction on self-reported adherence to ART among adolescents living with HIV. Preliminary analyses using Poisson regression indicated that overdispersion was a problem, which had the effect of yielding biased standard errors in the downward direction and increased the probability of incorrect inferences (Long & Freese 2014). To address this problem, we ran a negative binomial regression that included an error term accounting for overdispersion for count outcome variables with numerous zeros, hence yielding estimates that are superior to Poisson regression. To adjust for biased standard errors, we used robust standard errors with the STATA command vce(robust) in the final model (Long & Freese 2014). We also used a variation inflation factor to check for multicollinearity of the predictors and found that all the mean values were less than five, which is the recommended standard. After accounting for missing data, the total number included in the final sample analysis was 612 participants.

Results

Description of sample characteristics

Table 1 presents sample characteristics among 702 ALHIV. Results indicated that at recruitment more than half of the adolescents were females (56.41%; n=396), aged 10 to 16 years (M (mean) =12.4 years; SD (standard deviation) =1.97), and living in households with an average of six adults (M=6; SD=2.5). There were also on average two children under the age of 18 years (M=2; SD=1.9) in participants’ households. Furthermore, the majority of respondents (86%, n=604) reported having a medication supporter, defined as someone who either assisted them with taking their medication and/or helped pick up their medication from the health facility. These people included biological parents, aunts, brothers, grandparents, and community health care workers who were sometimes self-appointed or appointed by healthcare professionals. In addition, adolescents reported having a moderate quality of life (M=9.81; SD=3.6). The majority reported feeling good about their physical health (75.07%; n=527) and being satisfied with their overall life (80.34%; n=564). On the other hand, more than half of the adolescents reported having low energy (57.83%; n=406).

Table 1.

Sample characteristics (n=702)

Variable	n (%) or mean (SD)
Self-reported adherence (0–10 days), m (SD)	0.50 (1.23)
Pediatric quality of life (4–20), m (SD)	9.8 (3.6)
Age (10–16 years), m (SD)	12.42 (1.97)
Household composition, mean (SD)
No. of adults in the household	5.74 (2.54)
No. of children in the household	2.35 (1.93)
Primary caregiver, n (%)
Biological parent	330 (47.08)
Grandparent	165 (23.54)
Other relative	206 (29.39)
Medication supporter partner, n (%)
Yes	604 (86.04)
No	98 (13.96)
Sex, n (%)
Male	306 (43.59)
Female	396 (56.41)
Personal health, n (%)
Personal physical health rating
Good health	527 (75.07)
Poor health	175 (24.93)
Personal energy, n (%)
Low energy	406 (57.83)
High energy	296 (42.17)
Life satisfaction, n (%)
Satisfied	564 (80.34)
Not satisfied	138 (19.66)

m=mean; SD=standard deviation, n=frequency count/sample size, %=percentage, and no=number

The relationship between physical health, quality of life, life satisfaction and self-reported adherence to ART among adolescents living with HIV

Overall, adolescents reported that the number of days they missed their medication within the previous 30 days ranged from 0–10 days (M=0.50; SD=1.23). In the multivariate analysis, we examined the association between pediatric quality of life, physical health, life satisfaction, and self-reported ART adherence while controlling for other socio-demographic variables. Findings indicated that the following factors were associated with an increase in the reported number of days that adolescents missed medication: pediatric quality of life, age of respondents, having a grandparent as the primary caregiver, and being male.

Specifically, a decline in the pediatric quality of life score indicated by higher scores increased adolescents’ likelihood of reporting missed days of at least one dose of HIV medication by 5.5% (IRR=1.055; p=0.044). Similarly, for every additional one-year increase in age, adolescents’ had a 12.3% increase in the reported number of missed days of HIV medication (IRR=1.123; p=0.015). Having grandparents as primary caregivers compared to biological parents increased the likelihood of reporting missed days of at least one dose of HIV medication by 56% (IRR=1.56; p=0.046). Finally, being a male adolescent increased the likelihood of reporting more missed days of taking HIV medication by 59% (IRR=1.590; p=0.017) compared to female adolescents. On the other hand, life satisfaction was associated with a decrease in the reported number of days participants missed their medication. Specifically, being satisfied with life decreased the likelihood of reporting missed days of at least one dose of HIV medication by 41.2%, after holding other variables constant (IRR=0.588; p=0.014). Physical health rating and personal energy were not statistically significant in the model.

Discussion

This study examined the relationship between personal health, quality of life, life satisfaction, and self-reported medication adherence among a sample of ALHIV in a low-resource setting in southwestern Uganda. We found that factors such as the age of participants, having a grandparent as the primary caregiver, being male and low levels of pediatric quality of life were associated with an increase in the reported number of days adolescents missed their medication. The majority of ALHIV reported optimal adherence, in other words, they did not miss any doses of their HIV medication. Among those that reported non-adherence, at least one day of their prescribed ART medication had been missed in the past 30 days. We believe several factors can explain the high levels of adherence in the study sample. First, with the rollout of free ART in Uganda among all age groups, a significant number of children and adolescents were enrolled in ART (WHO 2019). Second, given that all participants in our study were young, with an average age of 12.4 years, and living with families, it is possible that caregivers or family members supported their adherence. This result is consistent with previous findings that showed that children depended on their caregivers to meet their adherence requirements (Damulira et al. 2019; Nabukeera-Barungi et al. 2015; Nabunya et al. 2020).

However, as adolescents grow and transition into young adulthood, they experience challenges related to establishing intimate relationships, being independent of caregivers, finding employment, and disclosing their status to others, some of which are predictors of suboptimal adherence among adolescents in the literature (Habere & Mellins 2009). Indeed, in our study, older adolescents were likely to report more days of missed medication. An increase in age is associated with autonomy, greater reliance on the judgment of peers and less dependence on parents or caregivers to pick up and ensure taking of ART medication, all of which contribute to suboptimal adherence (Habere & Mellins 2009). During this period, adolescents feel an increased need to fit in with their peers and start romantic relationships with people who may be unaware of their HIV status (Habere & Mellins 2009). As a result, they might be afraid to take their medication in the presence of their peers and/or significant others (Damulira et al. 2019; Habere & Mellins 2009).

Families, including extended families, are critical for medication adherence and are regarded as adherence promoting agents (Adejumo et al. 2015; Ammon et al. 2018; Biressaw et al. 2013; Hudelson & Cluver 2015; Musiime et al. 2012; Vreeman et al. 2008). In SSA, including Uganda, grandparents play an important role in taking care of children, including those infected and affected by HIV/AIDS (Biressaw et al. 2013). In addition to HIV and AIDS, extreme poverty and other preventable diseases, which limit their capacity to support and care for orphaned children, also burden many communities and families. Studies have demonstrated that the large number of young people impacted by HIV has overwhelmed even this hitherto social safety net (Curley et al. 2010). This has implications for both the quality of care and support provided to all adolescents, including ALHIV. Indeed, participants who reported living with grandparents as their primary caregivers were 56% more likely to report an increase in the reported number of days they missed medication in our study. Similarly, after the death of parents, ALHIV continue to struggle with the emotional pain of loss and are likely to move from caregiver to caregiver, which significantly affects their overall quality of life and subsequent adherence to ART medication (MacCarthy et al. 2018). Moreover, despite their pivotal role in child upbringing, grandparents may lack knowledge of HIV medication regimens and the ability to effectively support adolescents’ adherence to ART medication (Galea et al. 2018).

Additionally, low pediatric quality of life was associated with an increase in days of missed dosage. This finding is consistent with previous studies that documented that better quality of life has a reciprocal relationship with ART adherence among adolescents (MacCarthy et al. 2018; Nabukeera-Barungi et al. 2015). For ALHIV, access to food before and after medication, not being sick, shelter, clothes, and their relationships with peers could strongly influence their quality of life. This is important because many ALHIV in SSA live in low-resource settings with limited access to fulfillment of their basic needs (Curley et al. 2010). Thus, quality of life could be both a facilitator and a potential barrier to ART adherence. Consistent with previous studies (Kumarasamy et al. 2005), both the low quality of life and poor satisfaction with life have been associated with ART non-adherence among ALHIV. In the Ugandan context, it is very likely that many ALHIV do not have access to means to satisfy basic needs and experience high levels of stigma, both of which lead to low quality of life and subsequently to ART medication non-adherence (Damulira et al. 2019; Mutabazi-Mwesigire et al. 2014; Nabukeera-Barungi et al. 2015; Tuller et al. 2010).

Finally, participants in our study who reported being satisfied with life reported lower numbers of days that they missed medication. Satisfaction likely comes with optimism. If someone has something to look forward to, they are more likely to behave and act differently. Similarly, individuals who have come to terms with their HIV diagnosis are more likely to be satisfied with their lives and adhere to their ART medication (Nabukeera-Barungi et al. 2007; Sanjobo et al. 2008;). On the other hand, if individuals cannot come to terms with their illness, it can diminish their satisfaction with life and lead to suboptimal adherence (Dahab et al. 2008; Vyankandondera et al. 2013).

The primary limitation to our study findings was the reliance on self-reported adherence that can be subject to social desirability bias. However, studies are continuing to use self-reports as a measure of adherence alongside other novel techniques including viral load, pill counts, and technology-based methods (Kabore et al. 2015). In addition, in many low-resource settings, self-reported ART adherence is the cheapest option to gather data on ART adherence. These factors led to our decision to generate data from adolescents, self-reports of ART adherence. Finally, the cross-sectional nature of the data limited our ability to make causal inferences.

Conclusions

Despite these limitations, our findings indicate that quality of life and life satisfaction are significantly associated with ART adherence among ALHIV. Thus, strengthening existing and creating new support systems to promote optimal ART adherence and treatment outcomes for ALHIV in low-resource communities might be beneficial. Moreover, with increasing HIV rates among adolescents, effective and comprehensive efforts that are responsive to the special needs of ALHIV must be developed to ensure the best possible adherence to ART medication, low vertical infection, and decreased superinfection rates.

Table 2.

Negative binomial models on socio-demographic characteristics, physical health, quality of life, and life satisfaction predicting self-reported adherence to ART among adolescents living with HIV (n=612)

Predictors	Incidence rate ratio (Irr)	P-value
	Exp(β)
Pediatric quality of life (4–20), m (SD)	1.055	0.044*
Age (10–16 years), m (SD)	1.123	0.015*
Household composition, m (SD)
No. of adults in the household	1.016	0.786
No. of children in household	0.895	0.209
Personal physical health rating (ref. poor)
Good	1.027	0.897
Life satisfaction (ref. not satisfied)
Satisfied	0.588	0.014*
Personal energy (ref. high energy)
Low energy	1.083	0.658
Primary caregiver (ref. parents)
Grandparent	1.560	0.046*
Other relatives	0.903	0.644
Medication support partner (ref. no)
Yes	0.958	0.868
Sex (ref. female)
Male	1.590	0.017*
Pseudo R2	0.024
Model chi square (df)	32(11)	0.001***

Note:

The reference category is in parenthesis for categorical variables.

Negative binomial count variable regression. In the last 30 days, how many days did you miss at least one dose of your HIV medications?

^*p<.05

^***p<.001

two-tailed tests, Exp(β)=Exponentiated coefficient/Beta, m=mean; SD=standard deviation, n=frequency count/sample size, %= percentage, ref=reference category, and no=number