The One Who Doesn’t Take ART Medication Has no Wealth at All and No Purpose on Earth – a Qualitative Assessment of How HIV-Positive Adults in Uganda Understand the Health and Wealth- Related Bene ts of ART

Uzaib Saya (  uzaibsaya@gmail.com ) Pardee RAND Graduate School Sarah MacCarthy University of Alabama at Birmingham School of Public Health Barbara Mukasa Mildmay Uganda, Mildmay Hospital and Institute of Health Sciences Peter Wabukala Mildmay Uganda, Mildmay Hospital and Institute of Health Sciences Lillian Lunkuse Mildmay Uganda, Mildmay Hospital and Institute of Health Sciences Zachary Wagner RAND Corporation Sebastian Linnemayr RAND Corporation

Increased access to ART has been shown to improve the health, quality of life, and life expectancy of people living with HIV (PLWH) [2][3][4]. However, maximizing the bene ts from ART-such as delayed HIVrelated symptoms-depends on optimal retention in care and continued adherence to treatment over time (also called ART persistence). Poor adherence, such as missing doses, could increase a person's viral load and the risk for transmitting HIV. In Uganda, less than half of PLWH achieve 85% adherence to their ART medication [5]. Sustained engagement in HIV care and adherence to ART is largely determined by longterm individual behavior, as well as issues at the household and structural levels (especially those in uenced by economic, institutional, political factors) [6][7][8][9][10].
Evidence from the HIV and public health literatures indicate that there are various demand and supplyrelated reasons for ART initiation and continued adherence-these include socio-demographic and socioeconomic characteristics, existing health status, affective factors (such as fear of stigma, depression), social support, as well as institutional and health system barriers [11]. While these factors help understand how ART adherence is shaped, it is equally important to understand how long-term factors can be leveraged to promote ART investment and sustain present-day ART adherence. One way to do this is to consider how such treatment provides bene ts in the future. Evidence from the economics literature shows that declining mortality and increased life expectancy shape future-oriented behavior and affect economic choices and human capital investments [12,13]. The availability of an HIV treatment that increases life expectancy by an average of 18 years [14] could potentially alter how individuals consider health-related risk-taking and information-seeking, and wealth-related investment decisions. Highlighting the perceived bene ts of ART adherence in the short and long-term, especially how it affects both health and wealth can be a promising approach to expand ART use. Prior quantitative literature from the HIV context has examined how information about testing and ART availability changes subsequent behavior.
Studies from Kenya and Tanzania suggest that individuals who expected an HIV positive result but tested negative were more likely to change behaviors that increased their risk for contracting HIV, indicating that HIV test results in uence behavior when providing unexpected information [15]. When ART was made available in Kenya, female PLWH reported a 70% increase in pregnancies and 35% increase in selfreported sexual behavior [16]. These estimates suggest that availability of HIV testing and treatment can change health behaviors, but exactly how and why individuals change their behaviors is not well understood. A study from Malawi found that improved ART availability decreased individuals' selfreported mortality risk as measured by their life expectancy, but also increased labor supply, and futureoriented expenses in their children in the form of their education and clothing expenses [17,18]. There is also some evidence from South Africa suggesting that increased knowledge about ART-related life expectancy gains had an effect on human capital investments [19]. Understanding how individuals perceive bene ts of ART use on other domains such as wealth can help promote expansion of, and investment in ART.
In this qualitative study, we explored the perceived bene ts of ART adherence on health and wealth at the individual, household, and structural levels. We interviewed adult PLWH at an HIV clinic in Kampala, Uganda where the waiting area had visible reminders and cues about the positive in uence of ART (Appendix B)-these cues initially served as motivation for exploring our research question more rigorously using one-on-one interviews with ART clients. The results from this study can guide policymakers and researchers alike in understanding sustained ART use and how ART affects the broader lives of PLWH.

Methods
Our qualitative study reported is informed by the Consolidated Criteria for Reporting Qualitative research (COREQ) [20] (Appendix A).

Study Sample
Between July-August 2018, we conducted semi-structured interviews with a purposive sample of 40 HIVpositive adults who were enrolled in a randomized controlled trial (RCT) called "Behavioral Economics Incentives to Support HIV Treatment Adherence" (BEST) (clinicaltrials.gov: NCT03494777) [21]. This twoyear trial is testing the e cacy of using small lottery incentives to support ART adherence for treatmentmature HIV clients.
The participants were all patients at Mildmay Uganda, a clinic in Kampala, Uganda that specializes in the provision of comprehensive HIV/AIDS prevention, care, and treatment services. They were recruited into the RCT and were all 18 years of age or older, receiving ART at the participating clinic for 2 or more years, and had demonstrated recent adherence problems within six months of being recruited based on clinical records (de ned as showing lack of viral suppression, being sent to adherence counseling, or at disease stage 3 or 4 as per WHO guidelines). Individuals were excluded if they were not able to provide informed consent, spoke neither English nor Luganda (the local language), were participating in any other adherence-related study, were inconsistently using the trial-issued Medication Event Monitoring System (MEMS) cap to monitor adherence, used third-line treatment, or came to the clinic outside regular working hours.

Recruitment
We used Mildmay's electronic medical record system to identify eligible participants. Once identi ed in the clinic database, the study team took note of the subsequent recruitment opportunity based on their next available date of appointment. The study coordinators then looked out for individuals due for a visit that day and approached all eligible participants in-person and inquired whether s/he was interested in participating in an ongoing study. Respondents were assured they would not lose their spot in the queue for any clinic services. Those individuals that initially agreed were taken to a separate study room to verify eligibility, and the survey objective and procedures were explained. Once the participant gave written informed consent to participate in the RCT, s/he was given a MEMS cap to monitor real-time adherence and instructed to store their HIV medication in a pill bottle with the MEMS-cap attached. The qualitative interview was then conducted with only the participant and interviewer present in the room, and the participant was compensated USh 20,000 (equivalent to US $5) for participation. The target sample size was 40 ART clients as this was expected to be su cient to achieve saturation.

Data Collection
A team of one male and one female trained qualitative researchers (co-authors LL and PW) conducted semi-structured interviews in English and/or the local language Luganda at the preference of the participant, with interviews typically lasting 30-40 minutes. The interviewers (one male, one female) had undergone an extensive 30-hour qualitative interview training (led by co-authors UYS and SM) and had previously conducted other qualitative and quantitative studies with PLWH, especially those with ART adherence challenges. The interview guide (Appendix C) focused on understanding the determinants of ART adherence, as well as the effects of ART adherence on health, life expectancy, and wealth. Prior to interviews with any study participants, the data collection team piloted the interview guide among themselves and other team members to ensure use of appropriate language and local context cues. Open-ended questions focused on the key facilitators and challenges in taking ART, how ART affected health currently and in the future, and whether taking ART as prescribed would also in uence wealth and life expectancy.
All interviews were audio-recorded, transcribed verbatim and translated from Luganda into English, and stored on a secure data portal. Repeat interviews were not carried out. To ensure con dentiality, we separated personal identi cation information from the response data, and respondents were only identi ed through their clinic ID. Approximately 20% of the participant interviews were re-transcribed by an additional team member for quality control and re-evaluated against the original transcript to ensure consistency. The transcripts were then entered into Dedoose software. Transcripts were not returned to participants for comment and/or correction.
The interview team met regularly with two co-authors (UYS and SM) over the course of three weeks to discuss feedback on how participants described the effects of ART adherence on health and wealth. The team also discussed any problems that came up relating to interviewing goals and techniques. Troubleshooting involved improving the style of interviewer probing especially when they encountered issues such as when respondents said there was nothing stopping them from taking medications and did not report any barriers to taking ART. Interviewers were instructed to probe further on these points since all participants were eligible for the parent RCT because of adherence-related problems. Another issue raised was that of translating certain words into the local language, Luganda -for example, the words "health" and "lifestyle" are often interchangeable. As a result, when asked about the effects of ART on one's health, many respondents provided responses focused on their life goals (e.g., job, home, family etc.) rather than discussing immediate health-related impacts.
Demographic and adherence data on 38 of the 40 enrolled clients were obtained from a follow-up baseline interview in October 2018-January 2019. In 2 cases (5% of overall sample), data were not obtained due to non-participation in the follow-up baseline interview. At this stage, the participants had either dropped out of the study, or the MEMS-recorded adherence was less than 30%, which made them ineligible to continue in the parent study. Participants were typically eligible for their baseline interview three months after the pre-baseline visit when the qualitative interviews were conducted. These quantitative data included age, sex, level of education, marital status, WHO HIV infection stage, employment status, MEMS-measured adherence, and whether participant currently had undetectable viral load (to act as a proxy for the biological response to ART). These data were used as participant descriptors and if relevant, to gauge qualitative differences across groups of participants (e.g., by sex, infection stage etc.).

Theoretical Framework
To thematically categorize data in terms of participants' attitudes and expectations of future outcomes from ART adherence, we relied on health behavior change frameworks that incorporated behavioral learning theory, such as the theory of planned behavior which examines individual beliefs, perceptions, and ease of behavior change [22,23]. In the case of our study sample, we sought to understand how PLWH perceived the bene ts of ART adherence in the short and long-term, especially how factors at the social ecological levels shaped how they evaluated their own pill-taking behavior [24]. The socialecological model has been extensively studied and used to understand how factors with various domains determine health behavior such as ART adherence [25,26].
In this study, we hoped to better understand the extent to which factors at the individual, household, and structural levels in uenced how PLWH perceived the bene ts of ART. The individual level identi ed intrapersonal in uences including the experiences and attitudes towards the long-term impacts of ART adherence, while the household level examined interpersonal factors incorporating social dynamics with family and friends, while the structural factors included the larger political and cultural context and includes beliefs such as stigma and fatalism, and beliefs about disclosure to family and friends that may in uence individuals' ability to assess the effects of ART on health and wealth.

Analysis
We used a combination of inductive and deductive content analysis to categorize data based on emergent themes as well as previously structured hypotheses [27,28]. We repeatedly read the 40 transcripts to become familiar with the data managed in Dedoose, and coded the data based on recurring key issues and themes. We developed a structured coding framework based on a close assessment of all transcripts that included themes as well as content descriptions, inclusion/exclusion criteria, and sample quotes. Additional codes were created based on reading the transcripts. The coders double-coded eight interviews separately to reach a total of 130 excerpts, after which 30 were randomly picked using a random number generator, and each coder blind-coded them. This resulted in 93% agreement, after which one coder (UYS) continued coding the remaining interviews. These coders met biweekly thereafter to identify any emerging themes and discuss any questions or concerns. Once all coding was completed, one coder (UYS) read the excerpts per code application and extracted selected quotes per theme, and then Page 7/25 reviewed all coded excerpts and wrote a summary of results. We grouped themes at the levels of the social ecological model and examined the effects of ART use on respondents' health and wealth. As a nal step, we extracted quotations to illustrate common themes or responses among ART clients. Each quotation was labeled with the client's sex and WHO HIV stage. Table 1 describes the sample's demographic and health characteristics using survey and clinic data. The median age of participants was 32 years (range 18-56 years) and 50% of the participants were male, 68% were employed, and 55% had completed secondary education or more. More men in the sample had completed secondary education (67%) relative to female respondents (45%). The mean monthly income of participants was USD $43.50. Men in the sample had a higher monthly mean income at $47 compared to their female counterparts ($40) even though more female respondents (75%) reported being employed (largely driven by self-employment) than their male counterparts (61%). Most participants (69%) had a Stage 1 HIV classi cation (CD4 > 350 cells/µL) compared to 13% and 15% with Stage 2 and Stage 3 or 4 classi cations (CD4 < 350 cells/µL), respectively. A little less than half of the sample (45%) was virally suppressed (de ned as having less than 200 copies/mL) based on the most recent viral load conducted at the clinic prior to the interview. Notes: Data for sex and age obtained from clinic database, so N=40, but others may not add up to same number due to missing survey data.

Sample Characteristics
a Individuals in Uganda typically obtain vocational education after primary or secondary school education as post-primary or post-secondary training, but always prior to any university training.
b Income estimation is based on the sample after excluding 2 outliers due to their disproportionate likely due to data entry error. USD estimates calculated based on exchange rate of 1 USD=3700 Ugandan Shillings in January 2019 Notes: Data for sex and age obtained from clinic database, so N=40, but others may not add up to same number due to missing survey data.
a Individuals in Uganda typically obtain vocational education after primary or secondary school education as post-primary or post-secondary training, but always prior to any university training.
b Income estimation is based on the sample after excluding 2 outliers due to their disproportionate likely due to data entry error. USD estimates calculated based on exchange rate of 1 USD=3700 Ugandan Shillings in January 2019

Qualitative Interview Findings
We identi ed several factors pertaining to the perceived effects of ART on health and wealth at individual, household, and structural levels that are described in further detail below. Table 2 summarizes these themes along with the relative frequency with which they were mentioned by participants in our sample. *** a The words "health" and "lifestyle" are interchangeable in the local context per the data collection team. As a result, when asked about the effects of ART on one's health, many respondents provided responses focused on their life goals (e.g. job, home, family etc.) rather than discussing immediate health-related impacts. For the purposes of this analysis, we have combined those themes.
b Relative frequencies are denoted by: * discussed by < 25% of respondents (or n<10), ** discussed by 25-50% of respondents (or n=10-20), *** discussed by > 50% of respondents (or n>20) Individual level factors Respondents described how they perceived of individual-level bene ts to their health (via physical improvements to their health and well-being, being less susceptible to disease, and improved long-term personal habits) and wealth (increased motivation to work and earn more money).

Increased observable physical improvements and lowered susceptibility to illnesses
Respondents were encouraged by the positive rami cations of taking medication, such as having more energy, gaining weight, and feeling stronger overall. One respondent cited feeling stronger after taking medicine, while another noted that without taking ART, she would feel unwell and weak and credited her ART medication to helping her gain energy and be healthy.
"My situation is now good…right now my body is okay, I am strong and well because when people look at me, they cannot believe that I even take medicine, even my wife." -Male, Stage 1 "I was badly off, used to be so tiny and had rashes, I even asked the doctor that will I ever get the medication and gain energy again. I was like 25 kilos but now am in 70s, I thank so much this hospital" -Female, Stage 2 Respondents who tended to be at a higher disease stage also described being less susceptible to infection such as u and other illnesses once they were adhering to their ART. One respondent described how she had not gotten sick from any illness ever since starting her ART and went as far as to say that without ART, she would not be alive.
"It has helped me a lot because if it's not for taking this medication then I wouldn't be alive. I no longer get diseases that disturb me like back then, it's now 10 years I have never gotten sick of malaria ever since I started taking the medication and I no longer get rashes. It has really helped a lot" -Female, Stage 2 Some respondents rationalized how ART can help them; they described factors such as a change in antibodies, lowered CD4 count, or higher viral load. A few of them also described how not taking ART would result in "waking the virus" and HIV would no longer be suppressed.
"It's possible because those who made this medicine rst researched and found out that if a person uses this medicine and it suppresses the virus that came in the body, so if it sleeps and then antibodies continue to increase and do their work. That means a person can live a long life, because now the virus is suppressed so it's not doing any effect and every time it wakes up the antibodies have the power to ght" -Female, Stage 2 "this one (who misses his dose of ART) might die without realizing it. He might just (have) small u and we hear that he has died but he caused it himself because the doctor tells you have to take the medicine… If he (is) not using it well, the antibodies won't be moving well in other words the body is weak but this one that takes the body is not weak." -Female, Stage 1

Adoption of positive long-term habits
Few respondents also reported that taking ART regularly allowed them to adopt habits such as improved nutrition and exercise which would help them sustain their own physical and mental health in the longterm. One respondent cited the example of eating on time, improving personal hygiene, and participating in regular exercise, while another also cited how ART resulted in her going to more routine medical appointments. Others touted its bene ts to their spiritual well-being (i.e., ART helped them focus on their religion) and mental health (i.e., ART helped them not worry about factors outside their control and gain more con dence in their actions). "(I am) taking care of myself, I eat in time however small the food is. I also try not to overstress and over think a lot except a few things that may be hard to take in." -Female, Stage 1 "It has helped me to gain con dence and accept myself for who I am" -Female, Stage 1 2. Wealth 2.1 Increased nancial earnings and accompanying savings from being able to work more regularly Respondents described increases in their own nancial earnings and accompanying savings from working more regularly and without interruptions. In the short term, one respondent for example described how she had more time to conduct personal matters and not come to the clinic repeatedly if she got sick, and in the long term, she could also work longer and earn more money.
"It has helped me because I am able to stay alive and work hard for my future." -Female, Stage 3 "Before I started coming here, I used to come to this place all the time, but when I started taking my medicine I can even spend three months without coming back here so I would get more time to attend to my personal issues. This helps me to increase my nances because I work at my own pace." -Female, Stage 1 2.2 Increased personal motivation to work harder and earn more income Respondents also described rami cations such as motivation and personal drive to work harder (and earn more money). One respondent made the direct connection between taking her medication regularly and working harder in her profession while another made the connection to living longer and having more time to work for her future.
"I think if you're taking your medication well, it encourages you to work hard. Like for my case, right now am doing a course in cosmetology and it keeps me healthy and good looking because if you don't take your medication, the effects of poor skin will show. And there is no way you will be working on someone and yet you yourself can't look as good as you're convincing them they will be. So I think it encourages us to work harder and it strengthens us, keeps us moving"-Female, Stage 1 "Of course, when you're taking medication, you will have to live longer and that means you will have more time to work for that future. The more you take your medicine well, your body is stronger, and you can work harder."-Female, Stage 1 One respondent described other forms of non-monetary "wealth" such as being able to survive for her family and provide for her daughter's education while another respondent described increased motivation to spread religious messages to others.
"(I) am not rich, but I work and get some money; but the little I get, I thank God. But even this wealth is there because if you got the virus when you had a child of ve years and you educate her till when she completes...... isn't that wealth? Yet some time (ago) you would have died because you did not take well the medication (and) now you see that if I (am) certainly here and got checked when the kid was ve years, but now she is educated and even gave birth, so that's wealth also….my wealth is my children--the fact (is I) am with them and I thank God for that because there is no wealth greater than a child." -Female, Stage 1 "it will help me with more energy to spread the gospel to people. If I don't take drugs, I will be sick, and on Sundays it will be my day to go and I won't be able to make it."-Male, Stage 1 Another respondent drew this contrast more starkly by describing the futility of not taking one's medication, comparing the situation to not having any wealth in the short-term at all.
"The one who doesn't take the medication has no wealth at all and he has no purpose on earth… The one who doesn't take the medication might be dying soon." -Female, Stage 1

Household level factors
Respondents made the connection between ART-related bene ts to health and wealth at the household level by describing how ART helped them continue to do routine things and plan for their futures, while also deriving bene ts from social support and reduced engagement in risky behavior in the process.
1. Health 1.1 Increased ability to do routine things (e.g. school, work, raise children) and plan for future Individuals said that after taking ART regularly, they could conduct normal activities such as going to work or school, maintaining a job, and even raising a family.
"Basically at school, mentally I am very t. Before my medication, I used to be a person that, I don't know, maybe fear of my condition but now I am okay" -Female, Stage 1 The ability to plan for families (and have children who are not HIV positive) was a recurring theme, especially among female respondents.
"It has helped me because I am now happy, because I did not expect to give birth to a child who is negative, but he is now negative. It gives me encouragement." -Female, Stage 1 "This boyfriend I have rst of all he wants kids, and there is some counselor who told that if you're on medication you might not infect your boyfriend who is not infected, so in my future I might get a person who is not infected and I don't infect him and even our kids might be normal. And even the future of a job is good because you be taking your medication, when your health too, so you can't be red at work." -Female, Stage Unknown

Reduced unprotected sex and substance use
Male respondents also mentioned how taking ART mitigated the negative in uence of family and friends.
Respondents reported reduced engagement in risky behavior such as unprotected sex and substance abuse. Interestingly, none of the female respondents cited these factors.
"What might stop me from living long is maybe going back to something like adultery; I used to take alcohol but its good when you are taking medicine. I also used to take cigarettes and I left it when they told me it wasn't good for the medication, they told me too to eat well and also (be more mindful of) God because I allowed Him to enter my life. I thank God that He has keep live for long."-Male, Stage 3 "You do not have to go out with girls without protection because even if you take medicine you can still get infected with other diseases that can kill you. Sometimes you take medicine and smoke and (use) alcohol, it affects your life. There is also a bad reaction when you take the medication and take alcohol, it is a bad feeling" -Male, Stage 1 One respondent suggested that reduced engagement in risky behavior translated to savings because he avoided risky activities that he would have otherwise engaged in had he not been taking his medication.
"It helps me because a lot of boys I live with as for them, he can spend his money and goes for women but as for me I do not usually do that because I am keeping my money because I know my life depends on the medication because the medication does not interact with alcohol." -Male, Stage 2

Improved social support and motivation from peers and providers
Respondents also cited additional bene ts from ART-one respondent described the positive social support she received from her friends while another cited the support from colleagues whom he met at the clinic who encouraged him to continue taking his medication. Some also suggested the bene ts of improved social ties because of taking ART regularly-respondents extended this to widening their friends circle and getting lucrative business opportunities.
"It (taking ART regularly) has helped me to get a lot of things, seeing new things, living up to my youth age, getting new friends like that" -Male, Stage 3 "If you're sick like coughing or you're down bedridden, nobody will make business with you. So you take the medicine to stay strong, because nobody will make money if they are bedridden" -Male, Stage 1

Structural factors
Respondents described structural factors stemming from the local political, institutional, and cultural context that in uenced how they perceived ART bene ts; these included comfort around disclosure, lowered stigma, the idea of fatalism, and general forward-looking behavior such as having a sense of civic responsibility.

Motivates or enables disclosure to close friends and family
Some reported that taking ART and then getting better is a key driver in being able to disclose HIV status to loved ones. One respondent indicated that taking ART regularly and the promise of subsequent "healing" motivated him to disclose his status to his partner. Another respondent described how taking his medication helped with the stigma he faced prior to taking his medication.
"I have a girlfriend, but she doesn't even know that am infected. I have a time when I want to tell her, and I have never touched her like having sex but inside I ask myself "what if she gets to know", so that makes me take medication hurriedly so that they next time they check me I might be healed. My dream is to heal." -Male, Stage 1 1.2 Improved hope and greater aspirations for future and lowered stigma Other respondents cited additional effects of ART such as renewed hope and aspirations for the future, especially since lack of illness lowered social stigma associated with HIV and being on ART made them realize they could live longer.
"….even their social life will be easy in (a) community because the community will not stigmatize them as it will be a warm society as they live socially like any other person as they have accepted their status and lived on with it." -Female, Stage 1 "At rst, I could not believe (when I was told about my status) because they told me when I was 8 years that I was positive, and I thought I was going to die like my mother. I am now 19 years; around 11 years have passed by. I did not know that I would reach this far, but if I have reached here, then I know I can go further ahead. I thought I was the only one but I met a man and he told me all his kids were like me and he started taking the medication when he was 10 years old, but he is now around 40 years. That gave me courage and strength and I told myself I am not going to die, I have a life ahead of me." -Male, Stage 1 Another respondent described how taking his medication helped with the social stigma he faced prior to taking his medication.
"Let me say like that time when they told me I had to take the medication, I saw that people were going to start laughing at me and I even wanted to kill myself but then they told me to take the medication hence I will be better than the other normal people. So they told me that don't kill yourself because (you're) like a normal person now and no one suspect that you're infected or not." -Male, Stage 2

Greater fatalism (especially in thinking about inevitability of death)
Respondents also appeared to describe how being on ART minimized (and not enhanced) their sense or fear of death-some of them attributed this to fatalism and the inevitability of death, tying it closely to their religious beliefs.
"In my own thinking they say we came from God and it's where we shall go back but for me, I think even AIDS won't kill me, I will die of something else. There are God's plans because you can't say that I won't die, if it's about AIDS me I think I live long but I die of something else."-Female, Stage 2 "I think God is the one that makes for us a calendar and everyone has their own calendar.

Discussion
We conducted a qualitative study among clinic enrolled Ugandan PLWH and examined the kinds of bene ts they experienced in terms of their health and wealth. Our ndings outline the context PLWH face when thinking about the broad bene ts of ART to their lives, especially their attitudes and expectations of future outcomes resulting from ART-this topic remains largely unexplored in the qualitative literature.
Such factors can provide a promising lens to better understand the bene ts of ART that go beyond the short-term health related bene ts. In our study, such bene ts included those at the individual, household, and structural levels, and largely focused on their ability to live longer and be physically and mentally healthy, while also ful lling responsibilities pertaining to themselves (in terms of positive long-term habits and motivation to work harder), others (such as improved relations with family and friends), and society (in terms of improved civic responsibility). Our study ndings can be useful as testimonials for promoting ART expansion and investment in ART, as well as for future programming meant to target these areas as they can help understand how or why ART clients perceive certain bene ts, especially those that are accrued in the long-term.
At the individual level, respondents described the health-related bene ts of ART in terms of physical health improvements and lack of illness, as well as adoption of positive long-term habits over their longer lifespan. Respondents felt they had more control of their health and well-being because of adhering to ART. Wealth-related bene ts included increased nancial earnings and increased personal motivation to work harder. Prior research has quantitatively examined how increases in life expectancy stemming from ART can encourage forward-looking behavior because individuals are aware of their lowered mortality risk and subsequently make different life decisions [12,18]. These studies have sought to examine how PLWH respond to taking ART and change their patterns of risk-taking, information-seeking, and monetary investments in children (especially in education and clothing) [16,18,29]. The qualitative ndings from our study substantiate and contextualize some of these changes reported elsewhere; respondents on ART believe they will have a longer life with HIV and they can see direct bene ts to themselves in terms of both their health and personal wealth.
At the household level, respondents highlighted bene ts of ART on health and wealth, such as increased ability to do routine things, plan for their future and have a higher income potential. Female respondents especially described how ART helped them to better plan for their future and raise children who are not HIV positive, while also earning money to meet the needs of their families. This nding corroborates evidence from other studies describing that adherence to ART changes the context of childbearing for PLWH, speci cally since having children indicates that PLWH are leading a healthy adult life [30]. A metaanalysis found that the prevalence for fertility desires was 42% among PLWH, and this was higher among male PLWH [31]. Our qualitative ndings in this clinic-enrolled sample of PLWH show that females were more likely to describe these ART-related bene ts. This could indicate that counseling and reproductive health care be provided to female PLWH who generally would feel a greater sense of responsibility to their families. Male respondents on the other hand described the importance of ART in helping to reduce risky behavior such as unprotected sex and substance abuse. Such behaviors are well established risk factors for poor HIV care outcomes, especially in terms of facilitating testing uptake, engagement and retention in care, and ART use and adherence [32]. Evidence from Kenya and Uganda has shown that alcohol use especially among men was associated with decreased ART use, mostly due to poor engagement and retention in care [33]. Our qualitative ndings among male respondents suggest that future ART support interventions may bene t from being more tailored to drinkers or even those with low or medium level alcohol use. As care models of ART delivery branch out of the clinic, adherence support interventions should take advantage of such targeted messaging to incentivize PLWH who could possibly bene t from changes to their bene ts and enhance the bene ts they derive from ART.
At the structural level, many respondents described the link of ART with comfort in HIV status disclosure, lowered stigma, and increased civic responsibility. Our qualitative ndings are in line with related evidence which describes how ART can reduce internalized stigma in Uganda and even hasten disclosure due to improved coping mechanisms and social support [34]. All respondents highlighted the link with employment and ability to increase their earnings if they continued taking their ART. Some of these ndings are consistent with studies highlighting the important role of employment in ART adherence.
This literature describes the supportive role of employment among PLWH because it is associated with improved adherence; these studies normally describe the factors that determine adherence since nancial constraints are a key barrier to ART adherence. A systematic review of ART adherence and employment status described how employed PLWH were 27% more likely to adhere to ART than their unemployed counterparts [35,36]. Our ndings corroborate other studies also reporting that bene ts described by our study respondents (e.g. improved wealth, employment etc.) are associated with movement along the HIV care continuum [37,38]. Respondents who cite these bene ts will be more likely to engage in continued care and improved adherence.
Our study was subject to several limitations; while there were some observable differences in themes across male and female respondents, we were not able to assess if bene ts varied based on access, availability, or quality of education, jobs, or other factors related to the bene ts described by respondents. This difference may be possible to assess with a larger qualitative sample. Our ndings may also differ with a study sample that was not enrolled in a study seeking to improve adherence among low adherersthese individuals may already believe that ART improves health. The ordering of the questions (starting with health, describing effects on life expectancy, and then nally wealth) may have followed a natural order, but perhaps responses would be slightly different if the order were switched. Each participant was also interviewed only once, and had they been interviewed at the end of the 2-year parent study, we could assess if their perspectives had changed, and collected data on whether they experienced rst-hand any of the bene ts they described. While not exactly a limitation, the study ndings point to the discrepancy between participants perceiving the bene ts of ART to their health and wealth, yet not adhering to their medication (which is why they are enrolled in a treatment adherence RCT). This could be due to a desirability bias where they may not internalize the messages about ART, or they may be subject to shortterm structural and behavioral barriers (such as behavioral biases) that prevent them from fully realizing the bene ts they mentioned in the qualitative ndings.
Strengths of this study include the fact that we included individuals on long-term ART with recent adherence problems in a clinical setting and investigated the bene ts of ART in two distinct domains within their lives: health and wealth. This subject has previously not been studied extensively in literature using qualitative methods and can help understand how PLWH Perceive these bene ts

Conclusions
When recounting the contributions of ART in their lives, PLWH do not simply consider the short-term health bene ts, but also the longer-term implications on other domains. This is evident from the ndings of this qualitative study. Our study ndings can be used to better promote the expansion of ART and help donors understand the broader bene ts of ART. We provide contextualized evidence on broader social and economic evidence pertaining to ART. As described by participants, improvements in health via ART can affect future health and income by making individuals more productive (e.g., being able to attend school and get the requisite skills, attain better social networks etc.). Interventions can be designed with these bene ts in mind for individuals who don't speci cally recognize such bene ts (e.g., treatment initiators), and can speci cally use short and long-term individual-level factors to drive improvements in uptake (e.g., by encouraging healthier habits, or helping to increase household income).
Findings at the household and structural levels can be useful to clinic staff such as providers or counselors who are encouraging PLWH to sustain ART use. Evidence from other parts of SSA such as in South Africa suggests the need to incorporate this kind of evidence for treatment adherence-in that context, individuals discontinued ART because they were uncertain about its value and overall bene t in their lives [39]. Having bene ts -such as counseling for their future or resources on ways to get involved helping their communities -directly integrated into the design of interventions can be useful especially in a context when PLWH face so many competing interests to increase medication adherence. These bene ts can ultimately help providers and policymakers improve ART-related outcomes. Further research is warranted to determine how these bene ts differ across groups of PLWH, especially those with varying levels of access to ART, retention, or linkage to care in HIV care programs, or other sociodemographic factors. Additional research can also examine whether these ART bene ts can result in changes in behavior such as increased human capital investments.

Declarations
Ethics approval and consent to participate We obtained ethics approval from the RAND Corporation's Human Subjects Protection Committee (#2016-0956), the Mildmay Uganda Research Ethics Committee Institutional Review Board (#02013-2018), and the Uganda National Council for Science and Technology (#2394). Written informed consent to participate in the study was obtained from all participants prior to the start of data collection once the survey objectives and procedures were explained at study enrollment. All methods were performed in accordance with the relevant guidelines and regulations set forth by the various ethics approval agencies.

Consent for publication
Not applicable Availability of data and materials Datasets generated and/or analyzed during the current study are not publicly available due the restrictions statement in our study consent forms.

Competing interests
The authors declare that they have no competing interests.

Funding
The parent study was funded through resources provided by the National Institute of Mental Health (Grant: R01MH110350, PI: Linnemayr). Additional support for this study was provided by the Pardee Dissertation Award for Global Human Progress between 2018-2021 (Pardee RAND Graduate School). The content of this paper is solely the responsibility of the authors and does not necessarily represent the o cial views of the NIH.

Author Contributions
US led the analysis and interpretation of the interview data and led the writing of the manuscript along with mentorship and guidance from SM. PW and LL supported study implementation and data collection, and BM provided eld-based supervision and overall study monitoring in Kampala. ZW provide guidance on analysis. SL conceptualized and implemented the underlying parent study and supported the writing of the manuscript. All authors read and approved the nal manuscript.