The Characteristics and HIV-Related Outcomes of People Living with Co-occurring HIV and Mental Health Conditions in the United States: A Systematic Review of Literature from 2016 to 2021

Considering advances in HIV prevention and treatment, jurisdictional efforts to end the HIV/AIDS epidemic, and reduced stigma towards people living with HIV infection and mental health conditions, the authors systematically reviewed studies published between 2016 and 2021 and identified 45 studies that met the eligibility criteria. The review found that stigma towards mental health conditions still acts as a barrier to accessing HIV treatment, which impacts treatment outcomes. Additionally, social determinants of health, such as housing instability and poverty, appear to impact mental health and, therefore, HIV-related outcomes. The review also highlighted the mutually reinforcing effects of HIV, mental health, and substance use conditions, providing valuable insights into the syndemic effects of these co-occurring conditions. Overall, the review highlights the need to address stigma and social determinants of health in HIV prevention and treatment efforts and to integrate mental health services into HIV care to improve outcomes for people living with both HIV and mental health conditions.


Introduction
The HIV epidemic remains a prominent public health issue within the United States.In 2020, 30,635 people in the United States were diagnosed with HIV, with the impact of HIV disproportionately affecting MSM and Black/African American communities [1].Although there is no cure for HIV, treatment is available to achieve VLS, resulting in increased QoL and the prevention of HIV transmission to others [2].
PLWH have disproportionately experienced syndemic factors including mood disorders, anxiety, drug use, unhealthy alcohol use, and poorer QoL relative to the general population [3][4][5].Mental health conditions are more common among PLWH due to the stress of living with a serious condition, potential loss of social support, and experiences of HIV/AIDS stigma and discrimination amongst the general population [6].It is estimated that 63% of PLWH have a co-occurring mental health condition, in comparison to 31% of people not living with an HIV diagnosis [7].Research demonstrates that people with HIV are more likely to experience multiple, co-occurring mental illnesses and substance use disorders-collectively referred to as behavioral health conditions [8].Co-occurring HIV and behavioral

Study Selection and Data Extraction
All search results were compiled and screened by one reviewer (NI) for duplicates, removing 45 results.Next, two reviewers (HK, TB) independently applied inclusion and exclusion criteria to the title and abstract of each citation.To be included in this systematic review, original peer-reviewed articles had to meet the following criteria.The applied inclusion and exclusion criteria addressed six major areas: type of research, date of publication, geographic location, study population, clinical characteristics, and patient interventions (Table 1).Following the independent review of titles and abstracts, a third reviewer (NI) compared the inclusion/exclusion recommendations and resolved disagreements.This screening process resulted in the removal of 952 results.The inclusion and exclusion criteria were then applied to the text of the remaining 61 results (Fig. 1).
The reviewers used a two-step process to assess the articles based on eligibility criteria.In the first step, they selected articles based on title and abstract, and in the second step, they screened the full text of the included articles.If the title/ abstract was insufficient, they retrieved and examined the full article before making a final decision.Conflicts and disagreements between the reviewers were discussed and resolved through consensus.Following the screening process and additions from the screened systematic reviews, 45 studies were included in the data extraction process.Reviewers (AY, ED, HK, NI, TB, TO) examined and summarized all remaining studies utilizing a standardized template (Table 2).

Results
The initial literature search resulted in a total of 1058 citations from the electronic databases.After removing 45 duplicates, a total of 1013 records were potentially eligible and screened.A full-text assessment of 61 articles was performed.This led to 45 eligible articles relevant to our systematic review that were included for final data extraction and further analysis.Figure 1 shows the flow chart of articles examined for this systematic review.The results begin with an overall summary of the studies included in the systematic review and then describe the findings

Characteristics of Studies
Analysis of the 45 studies resulted in the identification of themes related to five major behavioral health categories (Fig. 2).Depression and substance use disorder were the most common conditions studied among people living with co-occurring HIV and mental health conditions, comprising 56% (25/45) and 29% (13/45) of studies reviewed, respectively.Other major behavioral health categories include anxiety at 16% (7/45), SMI and other psychiatric disorders at 16% (7/45), and trauma at 13% (6/45).Some studies conducted research across multiple behavioral health categories.Cross-sectional (22/45, 46%) and cohort studies (14/45, 31%) were the major study designs.There were two randomized controlled trials (4%), one case study (2%), and seven (16%) secondary data analyses.For methodology, 96% (43/45) of the articles used quantitative methods.The sample sizes ranged from 5 to 129,140, with the smallest sample size belonging to a case study and the largest belonging to a national cohort study.The average sample size was 7178, and the median was 315.The cohort studies and cross-sectional studies mainly recruited participants from clinical settings and collected data through patient assessments.Studies with medium to large samples were usually recruited from or were embedded in existing large cohort studies.The two randomized controlled trials collected primary data from trials.The secondary data analyses used data from medical chart/medical record reviews, existing large cohort studies (Women's Interagency HIV study, Preventing AIDS Through Health for Triples Study, etc.), HIV surveillance systems (e.g., the Medical Monitoring Project), or a large randomized clinical trial.The case study recruited participants from community clinics and community events.
The studies employed a variety of instruments to measure the dependent and independent variables.Demographic variables, which include social determinants of health (e.g., unemployment, housing, incarceration), and HIV-related behaviors (e.g., substance use, appointment adherence, care utilization) were mostly measured with self-report questions.Some studies measured substance use with standardized tests [19,20].Psychological and physical well-being variables were measured with verified scales (e.g., GAD, CES-D, PHQ-9, SF-12, NPI-Q, etc.).HIV-related outcomes (viral suppression, treatment adherence) were measured using biological markers or self-reports.Other outcome variables include patient/provider relationship, hospitalization, and emergency department/ urgent care visits, etc.

Depression
Depression was the most consistent theme identified.Over half of the studies (55%) reviewed focused on depressive symptoms or a depression diagnosis among PLWH.Almost one in four PLWH experienced moderate to severe symptoms of depression [21,22].Although certain studies found no observed associations found between depression and selfreported adherence [21] or no direct association between depression and VLS [19,23].PLWH with recent onset depression were less likely to reach VLS and had increased mortality [21,24].This suggests that clinicians should be aware of changes in the mental health of their HIV patients and how this may impact HIV-related outcomes.Risk factors including female gender, Caucasian race, injection drug use, MSM, having ≥ 1 ED visit, poor appointment adherence, an AIDS diagnosis, or a positive drug screen by Substance Abuse and Mental Illness Symptoms Screener (SAMISS) increased PLWH's odds for depression [22].It is important to note the higher potential for depressive symptoms found among females and people who inject drugs and consider this as a topic for further investigation in future studies of health outcomes among PLWH and co-occurring mental health conditions.Studies also suggest that changes in the severity of depressive symptoms may be more specifically related to social determinants of health than the patient-provider relationship for PLWH [23,[25][26][27][28][29][30].Multiple social determinants of health such as lower income level, transactional sex, housing instability, long-term survivor status, low social support (perceived or real), poor educational attainment, and food insecurity were shown to be associated with a higher prevalence of depressive symptoms [23,[25][26][27][28][29][30].Although the patient-provider relationship was beneficial for mental health outcomes in PLWH, addressing sociodemographic factors may be of greater importance [25].Women who reported more individuals who could care for them had more family support and those who reported feeling loved were less likely to report stigma [28].This highlights the importance of addressing social determinants of mental health such as social support and networks to better inform interventions to support the mental health of all PLWH, particularly WLWH.Additionally, PLWH with household incomes at or below the federal poverty level had a higher prevalence of depression compared with those living above the poverty level [29].This can create additional barriers to diagnosing and treating depression in PLWH such as poor adherence to medical appointments due to lack of transportation or inability to pay insurance copayments, all of which ultimately impact the ability of a PLWH with co-occurring depression to achieve VLS.Moreover, the literature highlighted the situation where social determinants of health interact with common comorbidities of depression, including substance use, violence, and other traumatic experiences [20,31,32].For example, Thurston et al. [20] suggested that experiencing more than one substance abuse, violence, and an HIV/AIDS diagnosis (SAVA) condition was associated with more depressive symptoms, particularly for WLWH.Going beyond the SAVA Syndemic, a qualitative study found that intersecting experiences of food insecurity and intimate partner violence co-occur with poor mental health and substance use, negatively influencing HIV prevention and treatment outcomes [32].
African American PLWH, especially females, were shown vulnerable to Substance Use, Mental Illness, and Familial Conflict non-negotiation related to poor HIV clinical outcomes [31].Although not statistically significant, African American females with a higher prevalence of mental illness were less likely than African American males to achieve VLS (67.1 vs. 65.9%,respectively) and more likely to utilize acute care services in the prior 6 months (53.9 vs. 48.5%,respectively) [31].Mental illness prevalence plays a crucial role in determining VLS.The impact could outweigh that of the presence of syndemic factors.For example, in a study examining the influence of substance use, mental illness, and familial conflict syndemic among PLWH who inject drugs, individuals with the highest prevalence of mental illness (Class 2) had 4.6 times the odds of having an unsuppressed viral load than individuals with the highest Substance Use, Mental Illness, and Familial Conflict syndemic burden (Class 4) (OR 1.48, 14.29; 95% CI) [31].
Certain healthcare utilization behaviors act as protective factors against depression.Pregnant WLWH with possible or definite depression were more likely to have perinatal care than WLWH with no depression (63.6 vs. 39.2,P < 0.05), therefore reducing the likelihood of perinatal transmission [23].This is likely because WLWH with depression benefited from more supportive services and intensive case management during the prenatal period regardless of intentional help-seeking for depressive symptoms.Resilience also emerged as a protective factor by reducing depressive symptoms among mothers living with HIV [23].Interventions focused on reducing the social determinants of mental health for PLWH with depression may focus on increasing resilience as a protective factor.Though this study focuses on pregnant WLWH, it offers an opportunity to consider how consistent engagement in healthcare-including supportive services and intensive case management-can provide support for PLWH with co-occurring depression.

Substance Use
Substance use was identified in approximately 29% of the studies (13/45).Findings showed that PLWH disproportionately experience unhealthy alcohol use, drug use, anxiety, mood disorders, and have a worse overall QoL relative to the general population [3][4][5].Specifically, a history of injection drug use or hazardous alcohol use was associated with poorer physical QoL (difference = − 3.6, 95% CI − 6.6, − 0.6; P < 0.05); overall substance use was linked to poorer mental QoL.The influence of substance use differs by age, which may be due to the better physical QoL but worse mental QoL in younger PLWH [4].More work is needed to identify age-specific interventions on substance use to improve QoL for PLWH.
Mental illness and substance use have been reported as co-occurring and mutually reinforcing among PLWH [20,31,32].Co-occurring substance use and mood disorders led to poor engagement in cognitive-behavioral therapy/psychotherapy, reducing treatment effectiveness [8].Active substance use, along with mental illness and familial conflict, all contributed to poor HIV-related health outcomes among disadvantaged African American PLWH [31].Ultimately, the co-occurring and mutually reinforcing interactions between food insecurity, poor mental health, and substance use contributed to worse health outcomes by increasing HIV risk behaviors and undermining engagement in HIV care and treatment [32].Comorbid substance uses and poor mental health outcomes were associated with lower rates of VLS and increased rates of hospitalization [33][34][35].For example, in a study examining the VLS rate among psychiatric inpatients who use drugs, VLS among psychiatric inpatients was 52%, compared to the Center for Disease Control (CDC) national suppression rate of 58% [9].
The frequency of alcohol use may also be strongly correlated with poor retention in HIV care, particularly among WLWH [33][34][35].Alcohol use was also found to be common among WLWH with polysubstance use such as marijuana, and cocaine [2,3,33,35].For example, African American WLWH who are hazardous/heavy drinkers were more likely to engage in polysubstance use (marijuana, cocaine, and heroin) compared to African American WLWH who are moderate/non-drinkers (OR 2.70, 4.39, 4.43, respectively; 95% CI) [3].Engaging and treating African American, Black, and Latino PLWH who are polysubstance users along with comorbid risk factors is essential to ending the HIV epidemic.

Anxiety
Anxiety disorders were studied among PLWH in 16% of the studies reviewed.Findings showed that PLWH were disproportionately affected by anxiety [5,38].Moreover, PLWH with anxiety were more likely to experience co-occurring physical and behavioral health conditions, including mood disorders, unhealthy alcohol use, substance use, and poorer QoL [4,5,33,39].Cis-gender women were found to be disproportionately affected by the syndemic of co-occurring HIV, anxiety, and other conditions [33,39].
Finally, WLWH and moderate to severe anxiety were found to experience significantly poorer adherence to HIV medication (panic-related anxiety [OR 0.71, 95% CI 0.54-0.93])[40] and decreased odds of attending primary care visits (OR 0.69, P = 0.03) [33].Overall, the literature proposed that incorporating routine screening for anxiety in HIV clinical settings and developing interventions for anxiety among PLWH may help reduce HIV transmission and improve health outcomes, patient retention in care, and HIVdisease management [33,38,40,41].

Serious Mental Illness and Other Mental Health Conditions
Themes related to SMI and other psychiatric conditions were studied among PLWH in 16% of the studies reviewed.Overall, study purpose and findings varied significantly, demonstrating a need for further research related to SMI among PLWH.HIV infection interacts with mental health symptoms among PLWH living with co-occurring SMI.A qualitative study on the causal pathways between HIV infection and SMI [42] found that HIV diagnosis often preceded depressive symptoms for people with unipolar depression (n = 11) and symptoms of mania and psychosis often preceded HIV for people with schizophrenia/schizoaffective and bipolar disorder (n = 15).
Further, SMI negatively affects HIV treatment outcomes.Five studies looked at HIV treatment and/or health outcomes among PLWH and co-occurring SMI, with four looking at behavioral health associations with HIV treatment adherence and/or VLS [2,3,9,43], and one evaluating ART prescription patterns [44].Two of these studies focused on PLWH who were receiving inpatient mental health treatment [9,44].An evaluation of ART prescription patterns for 506 PLWH receiving care within psychiatric inpatient settings indicated that ART was prescribed upon discharge for 39% of the study population, starting at 28% the first year and was generally increasing over time (χ2 6 = 14.05;P = 0.03) [44].Findings from community-based treatment adherence interventions showed that PLWH with co-occurring psychotic and bipolar disorders had worse VLS compared to those with co-occurring non-psychotic depressive disorders [43].

Post-traumatic Stress Disorder and Trauma
Themes related to trauma were studied among PLWH in 13% of the studies reviewed.Findings showed that PLWH had a higher prevalence of PTSD [3,45] and were more likely to experience adverse/traumatic events compared to populations without HIV [5,27].For example, Hutton et al. reported that African American WLWH were three times more likely to experience lifetime PTSD, regardless of alcohol use ([heavy drinking OR/CI 6.27 (3.56, 11.0); P ≤ 0.001] [non-drinking OR/CI 4.48 (2.71, 7.40); P ≤ 0.001]) [3].HIV self-care behaviors, such as taking ART medication, could be a trigger for traumatizing lived experiences such as victimization, internalized HIV stigma, and events that may have contributed to their current HIV status [40].
The impact of PTSD and trauma on health outcomes and treatment adherence varied by study.Brown et al. reported a significant association between sexual assault and depressive symptoms (P = 0.002) and ART non-adherence (P = 0.006) [27].However, Young-Wolff et al. examined the association between adverse childhood experiences, depression and anxiety symptoms, substance use, and HIV-related outcomes among PLWH at risk for unhealthy alcohol use and found that adverse childhood experiences specifically were not associated with depression, substance use, or HIV-related outcomes [5].Even when PLWH were adhered to ART, PTSD was still significantly associated with immune dysregulation (P = 0.03) [45].
Programs using trauma-informed interventions may improve ART adherence [27] and negative health outcomes associated with past or current experiences of trauma [5], especially among middle-aged populations, men, and women.Findings also support a tailored cognitive behavioral treatment approach in helping to improve HIV medication adherence and decrease PTSD symptoms [46] (41).Interventions should be tailored to address the stressors, challenges, and resiliencies of specific populations, specifically black women living with HIV [3,46].

Discussion
This review describes the demographic and clinical characteristics and patterns of engagement in the HIV and behavioral health care continuum among people living with co-occurring HIV and mental illness in the United States.This systematic review highlights an abundance of research focusing on depression and substance use in relation to health outcomes in PLWH.The extent to which the subcategories of anxiety, trauma, and serious mental illness have been studied is similar in terms of their overall proportion of the literature, but they make up a substantially lesser volume than both depression and substance use.Additionally, most of these studies focus on a single mental health condition.Findings from the study are consistent with previous research [47,48], indicating a significant negative impact of mental health comorbidities on HIV outcomes.

Overall Findings and Implications
Depression was the most prevalent comorbidity detected in the literature that interacts with HIV status.It both contributed to and was affected by HIV outcomes.Depression was associated with decreased HIV VLS [31].Meanwhile, HIV risk behaviors and poor adherence predicted higher rates of depression among PLWH [22].Our findings also suggested associations between multiple social determinants of health and depression in PLWH, which has implications for potential interventions targeting social determinants of health to improve depressive symptoms and HIV outcomes in PLWH.Similar results were found with anxiety, which is consistent with the general association between depression and anxiety [49].However, anxiety and its specific role in HIV care and management remain understudied among PLWHA, indicating a further need for research in this area and work on the identification of anxiety as well as the development of interventions for anxiety among PLWHA.
Another highlight of the findings is that substance use often co-occurs with mental illness in a mutually reinforcing pattern in PLWH [20,31,32], which then contributes to decreased viral suppression, increased hospitalization, and decreased quality of life [33][34][35].Particularly, excessive alcohol use was associated with poor retention in the case of WLWH [33][34][35].Screening and treatment of polysubstance use disorders in PLWH and alcohol use interventions targeting WLWH can potentially help detect co-occurring mental illness in PLWH, to break the cycle of substance use and mental illness that leads to poor HIV outcomes.
Findings are less consistent regarding SMI and other psychiatric conditions.A qualitative study showed PLWH with non-psychotic depressive disorders had better VLS than those with psychotic and bipolar disorders [43].Whereas a study on PLWH receiving inpatient psychiatric care found no association between psychiatric disorders and VLS [9].Further research is warranted to determine whether SMI and other psychiatric conditions are correlated with HIV outcomes in PLWH and how they interact with other cooccurring factors in influencing HIV outcomes.
Experience of PTSD was affected by HIV status, while trauma predicted worse HIV outcomes through ART non-adherence or immune dysfunction [5,45].Existing trauma-informed interventions were shown to be effective in improving ART adherence and health outcomes associated with ACEs [5].The dissemination and implementation of trauma-informed practices have potential benefits on HIV outcomes for PLWH.Veterans propose an interesting dilemma as they may show a high rate of PTSD and may require further consideration.Future studies should investigate the trauma-specific stressors, challenges, and resilience of specific PLWH populations who experienced trauma and PTSD [3,46].
Moreover, the study found African American cisgender women disproportionately affected by the syndemics of HIV, substance use, anxiety, PTSD, and other mental health disorders.Thus, they had poorer HIV care adherence or VLS compared with other groups.Further studies should highlight the specific gender and racial factors in achieving optimal HIV outcomes among people living with co-occurring HIV and mental health conditions.

Limitations
This research only includes articles from six chosen databases and has the potential to exclude research that may have been published with other databases.Most of the reviewed studies (46%) were cross-sectional studies, which could only suggest a correlation between mental health conditions and HIV outcomes with the direction of the effects undetermined.The literature mainly focused on the cisgender adult population and populations with higher risks, such as men who have sex with men.Other demographic groups such as the transgender, bisexual, non-binary, and teenager populations were under-represented, and this represents an area in need of further research.Although some literature defined the intersection of race and ethnicity with HIV/AIDS and mental health conditions as a focal point in their research, this was largely an understudied social determinant of health among the publications analyzed in this review.Additionally, the chosen articles measured the outcome variables with different instruments, which might render direct comparisons of the results problematic.This study also did not include grey literature which can be a useful resource in systematic reviews and the inclusion of grey literature in future reviews may add additional insight into this topic.Largescale studies with similar outcome measurements will be needed for direct comparisons and more solid conclusions.

Conclusion
This systematic review offers a comprehensive discussion on the co-occurring and mutually reinforcing behavioral health conditions of PLWH to provide meaningful insights into the syndemic effects of HIV, mental health, and substance use conditions.Specifically, this systematic review reconfirms that the majority focus of the literature on co-occurring HIV and mental and behavioral health conditions are focused on depression research, followed by substance use.There is a need for further research on anxiety, trauma, and other serious mental illness.Within the work on serious mental illness, there is a need to separate and define serious mental illness conditions (i/e bipolar disorder and schizophrenia).Future research should focus on a few key priority areas.Future studies should include strategies for investigating multiple mental health conditions rather than a single mental health condition in isolation.This strategy has already been employed to some extent by current studies included in this review but is generally not the major focus of the research.Next, there is a need to focus on the impact of social determinants of health on both mental health and health-related outcomes in PLWH.Finally, researchers should attempt to focus on studies that better understand and potentially identify causal mechanisms.This includes further investigations that can explore targeted interventions based on these demographic and clinical characteristics.Longitudinal cohort studies and clinical trials are needed to explore the causal relationships between mental health conditions and HIV outcomes.
Overall, effective HIV care should effectively evaluate and screen for mental health disorders and integrate interventions with strategies tailored to specific mental health conditions.Because efforts to end the HIV epidemic at the state and federal level emphasize the connection to care in PLWH [15,16], it is essential to include effective mental health evaluation and care to improve the long-term quality of life among PLWH.More pointedly, the success in ending the HIV epidemic in the U.S. potentially depends on how effectively public health can break down the synergic effects of HIV infection, substance use, and mental health conditions.Based on this systematic review, future policies should allocate more resources towards integrated HIV and mental and behavioral health interventions that are tailored to meet the unique needs of PLWH.
Fig. 1 PRISMA 2020 flow diagram for new systematic reviews

Fig. 2
Fig. 2 Studies by behavioral health diagnosis

Table 1
Inclusion and exclusion criteria

Table 2
Characteristics of studies