With the advent of highly effective antiretroviral therapy, people with HIV (PWH) in the United States are living longer, healthier lives. Despite successes in treatment, a large percentage of PWH experience challenges related to poor quality of life (QoL), including pertaining to physical health, mental or emotional health, and structural/subsistence factors related to social determinants of health (SDOH) (1)—defined as conditions in the environments where people are born, live, learn, work, play, worship, and age (2). Aspects of poor quality of life are highly prevalent among PWH and have been associated with adverse clinical outcomes (3, 4). In addition, certain SDOH, such as food insecurity and homelessness, frequently co-occur among PWH, and experiences with multiple negative SDOH are associated with deleterious outcomes (5). Given the unique challenges that PWH face, focusing on assessing and addressing QoL provides a more holistic approach for addressing overall health and well-being of PWH—beyond viral suppression (6, 7).

The U.S. National HIV/AIDS Strategy (NHAS), which provides the roadmap for ending the HIV epidemic (EHE) in the United States by 2030, now prioritizes improving QoL among PWH—particularly among certain disproportionately affected populations. A multiagency federal workgroup was convened to identify appropriate indicators for monitoring QoL among PWH. During listening sessions held to gather input from persons with lived experience, the community voiced concerns about needing “to consider other factors that influence QoL, including SDOH (e.g., unemployment, food insecurity, housing instability), and not solely focus on health-related QoL.” The workgroup also considered several data sources to monitor QoL. The CDC’s Medical Monitoring Project was chosen given its ability to report representative estimates for a variety of characteristics related to QoL, based on a geographically-diverse group of PWH, both nationally and for jurisdictions that report to MMP (8, 9). Five national indicators were added to monitor QoL annually based on MMP measures—good or better self-rated health (physical health), unmet needs for mental health services (mental or emotional health), and hunger or food insecurity, unemployment, and unstable housing or homelessness (structural or subsistence) (8).

Although QoL is monitored among PWH through NHAS, including among some priority populations, co-occurrence of different aspects of QoL is not well described among PWH, and monitoring of co-occurring QoL indicators is not routinely recommended. We utilized the existing NHAS framework to develop a novel poor QoL index that quantifies the number of indicators of poor QoL experienced by PWH. We then applied this index to MMP data, reporting the distribution of scores among PWH alongside data on individual indicators of poor QoL to identify populations who had the worst QoL and how their QoL was affected—overall and by NHAS priority populations. Finally, we examined associations between the poor QoL index scores and key clinical outcomes.

Detailed information on demographic characteristics and SDOH among PWH using MMP has been extensively described elsewhere (1). Based on the scores of the poor QoL index, over half of all PWH (55.1%) had ≥1 indicator of poor QoL; 31.5% had 1 indicator, 15.2% had 2 indicators, 6.4% had 3 indicators, 1.7% had 4 indicators, and 0.3% had 5 indicators (Figure 1). Structural/subsistence factors were strongly associated with each other (Appendix Table 2). Of all NHAS priority populations, the prevalence of having ≥3 indicators of poor QoL was highest among PWID (26.5%) and persons aged 18–24 years (12%).

Of NHAS priority populations, PWID had the highest percentage of poor or fair self-rated health (PWID: 38.2% vs. range of poor or fair self-rated health among other NHAS priority populations: 12.8%–36.4%), unemployment (32.8% vs. 14.0%–25.1%), hunger/food insecurity (40.0% vs. 15.9%–28.9%), and unstable housing/homelessness (44.2% vs. 17.5%–36.7%; Figure 2, Appendix Table 3). Relative to all adults with HIV, notably large percentages of persons aged 18–24 years also experienced unemployment (25.1% versus 15.1%), hunger/food insecurity (28.9% versus 17.8%), and unstable housing/homelessness (36.7% versus 18.7%). In addition, large percentages of Black women had poor or fair self-rated health, relative to all adults with HIV (36.4% versus 29.4%). Of all NHAS priority populations, people aged 18–24 years had the lowest prevalence of poor or fair self-rated health (12.8% vs. 21.8%–38.2%). The prevalence of unmet needs for mental health services among those with needs was lowest among Hispanic/Latino MSM (22.4%) and highest among Black MSM (27.2%).

After adjusting for age, race/ethnicity, and sex, every indicator of poor QoL was significantly associated with adverse clinical outcomes, including not being retained in care, missing ≥1 HIV care appointments, missing ≥1 ART dose in the past 30 days, not having sustained viral suppression, and having ≥1 ER visit or hospitalization (Table). Having poor or fair self-rated health was strongly associated with having ≥1 ER visit (adjusted PR [aPR]: 1.70; 95% CI: 1.62–1.78) or hospitalization (aPR: 2.34; 95% CI: 2.15–2.54). Experiencing hunger/food insecurity (aPR: 1.94; 95% CI: 1.80–2.10) or unstable housing/homelessness (aPR: 1.86; 95% CI: 1.73–1.99) were strongly associated with missing ≥1 HIV care appointment. We also observed dose-response associations between higher poor QoL index scores and all adverse clinical outcomes included in the analysis, with many non-overlapping CIs (e.g., missing ≥1 HIV care appointments, missing ≥1 ART dose in the past 30 days, having ≥1 ER visits). The strongest associations (i.e., those with the greatest magnitude) were observed with missing ≥1 HIV care appointment (aPR range: 1.60–3.23), having ≥1 hospitalization (aPRs range: 1.84–3.22), and having ≥1 ER visit (aPR range: 1.45–2.18).

Using nationally representative data and the existing NHAS framework for QoL, we developed and applied a novel index to identify groups of PWH most highly affected by poor QoL. The index includes structural factors, such as unemployment, hunger/food insecurity, and housing instability, based on community input. These factors are not in other established QoL indices like the MOS or the WHOQOL-BREF (12, 13). We demonstrated the utility of using such an index alongside data on individual indicators to provide a more complete picture of people’s QoL and identify areas for improvement for QoL among PWH. Specifically, the index can first be used to identify groups of PWH who are most affected by poor QoL (i.e., those with higher scores), and the individual indicator data can then be used to identify how to prioritize valuable resources in the clinical and public health settings to improve people’s QoL. Using this method of assessment, we found that over 1 in 4 PWID experienced ≥3 indicators of poor QoL. In addition, a large percentage of people aged 18–24 years experienced structural or subsistence factors associated with poor QoL, and 1 in 3 Black women had poor or fair self-rated health. Dose-response relationships were also observed between the number of indicators of poor QoL and adverse clinical outcomes, with poorer QoL being associated with worse outcomes—thus demonstrating the utility of using a poor QoL index to identify people who may be at higher risk of experiencing health challenges.

This method of assessment could be applied more broadly to improve PWH’s lives. First, this method could be used to monitor—both nationally and locally—patterns in QoL among PWH to identify areas of improvement and directing resources for improving PWH’s QoL. Second, it could be used in the clinical setting to identify HIV patients with poor QoL and intervene in ways to improve their health—including through linkage to needed services. As NHAS continues to prioritize improving QoL among PWH, expanding recommendations to include assessment of co-occurring QoL domains aligns with a whole-person approach. Using this study’s proposed method for poor QoL assessment could help identify and prioritize interventions for PWH who may be at higher risk for experiencing adverse health outcomes. Identifying areas for improvement in meeting NHAS-related goals for improving QoL among PWH could help inform resource allocation for targeted public health action. In addition, updating national guidance for developing local epidemiologic profiles—which are used to inform local HIV prevention and treatment funding—to include co-occurrence of QoL factors through our proposed approach would help ensure local monitoring (14). Updating national guidance and improving national and local monitoring through a multi-dimensional approach could be helpful in improving the health and well-being of PWH.

At a systems level, local availability of services to address physical health, mental or emotional health, and subsistence needs—which may at least partially be affected by local monitoring of patterns—should reflect the needs of PWH. Mismatches in service availability versus need could result in adverse outcomes. A recent analysis demonstrated that national funding for the Department of Housing and Urban Development’s Housing Opportunities for People with HIV/AIDS (HOPWA) program—the only federally-funded program dedicated to addressing housing instability among PWH—was insufficient overall and may not have matched housing assistance needs among PWH in certain jurisdictions (15). One potential reason for this could be that funding is partially based on local poverty rates among the general population; however, PWH may have greater documented needs for housing assistance than the total US population (16), and this disparity could vary by jurisdiction. PWH may have greater needs related to mental health and SDOH, including subsistence needs, than the total US population (16, 17). Ensuring that federal and local funding structures are based on reporting of QoL in epidemiologic profile reports could help ensure appropriate resource allocation based on need.

Our proposed method of assessment could be used in the clinical setting to identify and intervene to improve the health of HIV patients with poor QoL. The Infectious Diseases Society of America currently recommends initial assessment and ongoing evaluation of social and structural factors such as unstable housing, food insecurity, and access to transportation, as well as annual mental health evaluations (18), but the frequency of assessment is unspecified or is inconsistent between guidelines (18–20). Lack of initial and routine assessment of such factors could result in missed opportunities for timely linkage to needed services and could negatively impact patient care and outcomes. Within the clinical setting, integrating routine evaluations of physical health with mental and emotional health and subsistence factors reflects a patient-centered, whole person approach. Incorporating the use of the poor QoL index—via a self-administered questionnaire in the patient portal or a staff-administered intake assessment—could provide an efficient method for assessment. Linking responses to ICD codes (e.g., ICD-10 “Z” codes for SDOH-related factors) could facilitate documenting patient responses to the index, helping providers easily identify and track patients’ issues that may impact their health, and identify patients with the greatest need for intervention (21). Such an approach could provide opportunities for building patient-provider relationships and linkage to services in a timely fashion by meeting patients where they are, ensuring patient-centered care. This could be especially helpful for patients disproportionately affected by HIV who may be less likely to be engaged in care and more likely to have unmet needs for ancillary services or adverse outcomes, such as PWID (22). Specifically, using the poor QoL index to prioritize data-to-care efforts could be helpful in prioritizing limited public health resources in addressing the needs of patients at highest risk of having adverse outcomes.

Connecting people to appropriate ancillary services is essential for addressing certain SDOH and supporting PWH’s health and QoL, as unmet needs for such services were found to have dose-response relationships with negative outcomes (23). However, services related to addressing complex physical health, mental health, or subsistence needs may not be available onsite or through referral at every facility. For instance, a recent analysis of HIV care facilities attended by MMP respondents demonstrated that over 1 in 6 facilities did not offer mental health services, 1 in 4 did not offer housing assistance, and 1 in 3 did not offer food bank or meal delivery services onsite or through referral (24). Connecting people to needed services may require additional time and logistics for referrals or submitting appropriate documentation for receiving subsistence benefits. Having a patient navigator or case manager available could help manage these logistics. However, nearly 1 in 6 PWH reported needing but not receiving HIV case management services, which could limit access to needed ancillary services (1). Broadening access to patient navigation or case management services aligns with a patient-centered care approach, as recommended through NHAS, and ensures seamless follow through once needs are identified. Such an approach also mirrors the structure used by Ryan White HIV/AIDS Program (RWHAP)-funded clinics, which are more likely than other HIV clinics to co-locate ancillary services with HIV care (25, 26). For non-RWHAP-funded clinics without co-located services, strengthening local partnerships may help connect patients to services not available onsite.

This analysis was subject to limitations. Self-reported QoL indicator data may have been subject to social desirability bias. All measures were assumed to be co-occurring if respondents had experiences with >1 factor, which may not have been the case if they did not occur at the same point in time. The QoL index components were purposely chosen based on the existing NHAS QoL framework; future studies could evaluate the benefit of adding additional index components. The index weights each component equally; future work could consider weighting components differently (e.g., based on associations with clinical outcomes) to prioritize data to care lists or resources for patients with the greatest need. Associations between individual indicators and outcomes were not adjusted for other indicators; relationships between indicators are complex, and may co-occur, affect each other, and be time-varying. Future studies could consider examining causal relationships with outcomes. This study was cross-sectional; therefore, associations may not imply causality. Future longitudinal studies could be conducted to evaluate downstream effects of poor QoL on health outcomes. The COVID-19 pandemic affected aspects of QoL over time; annual estimates of individual QoL indicators have been previously described (1). In addition, response rates during the 2019 and 2020 MMP data cycles, which could have affected overall patterns in QoL in NHAS priority populations if response rates were differential with respect to variables of interest. Although MMP’s response rate was suboptimal, estimates were adjusted for nonresponse and post-stratified to NHSS population totals by age, race/ethnicity, and sex.

Using the existing NHAS framework for QoL, this was the first national analysis to demonstrate the utility of using a novel index alongside the distribution of individual indicators of QoL to identify populations most disproportionately affected by poor QoL and how to prioritize valuable resources for improvement. Unlike many other established measures, this index incorporates factors related to unemployment, housing instability and food insecurity based on input from the community of PWH. When we applied this method of assessment to a representative sample of PWH in the United States, we found that a substantial percentage experienced multiple, co-occurring indicators of poor QoL, and certain populations highly affected by HIV have specific needs for improving QoL. Poor QoL was associated with a variety of adverse outcomes, with a dose-response relationship between the number of poor indicators and outcomes. This assessment method could be used to monitor national and local patterns in QoL among PWH and to identify and provide services in the clinical setting for PWH with poor QoL to improve their well-being and health. Expanding national reporting and clinical recommendations to include this multi-dimensional method of assessment of QoL could improve interventions targeted to improve people’s QoL, and therefore, meet NHAS and EHE goals related to improving QoL and health and well-being of all PWH in the United States.