Association of syndemic conditions and quality of life among people living with HIV/AIDS

The syndemics theory seeks to understand the e ﬀ ect of multiple synergic problems in promoting poor health outcomes. To disentangle which and how syndemic conditions a ﬀ ect the quality of life (QoL) may be important to improve well-being of people living with HIV/AIDS (PLWHA). This study evaluates the association between syndemic conditions and QoL among PLWHA. We performed a secondary analysis using data obtained between 2014 and 2017 among PLWHA under care in Rio de Janeiro, Brazil. The outcomes were the six QoL domains (physical, psychological, level of independence, social relationships, environmental, and spirituality) measured through the World Health Organization Quality of Life in HIV infection scale, abbreviated version (WHOQOL-HIV-BREF). The independent variables were demographic and clinical characteristics, syndemic conditions (binge drinking, compulsive sexual behavior, polysubstance use, intimate partner violence, and depression), and syndemics (two or more syndemic conditions simultaneously). Bivariate analysis ( t -test and ANOVA) and linear regressions were performed for each quality-of-life domain. The analytical sample comprised 1530 participants, mostly male at birth (64%) and with median age of 43 years. The syndemic conditions most frequently observed were binge drinking (56%), IPV (13%), and depression (9%). Both individual syndemic conditions and syndemics were associated with worse QoL. In the multivariate analysis, positive screening for depression was associated with worse QoL in all domains. Polysubstance users presented worse QoL at social and environmental domains. Intimate partner violence was associated with worse QoL at environment domain while binge drinking was associated with worse scores in the physical domain. The presence of syndemics increased the likelihood of worse scores in the psychological, social, and environment domains. Our study expands the understanding of QoL in PLWHA, as it considers a holistic/integral, multifactorial, and synergistic approach to the determinants of QoL. Seeking strategies that target syndemics may be important to improve patient-centered outcomes in health.


Introduction
The World Health Organization (WHO) defines the quality of life (QoL) as "an individual's perception of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards and concerns", a concept that involves physical and psychological wellbeing, personal beliefs, social relationships, and one's relationship with his or her environment (WHO, well-being (Noronha et al., 2016), i.e., The degree to which people enjoy the possibilities of their lives and are satisfied with life (Ferrans, 1996). To evaluate HRQoL, there are generic questionnaires not related to a specific disease, such as WHOQOL-100 or WHO-QOL-BREF, and specific questionnaires, such as the WHOQOL-HIV, which evaluates QoL in people living with HIV/AIDS (PLWHA) (Ferro, 2012).
To our knowledge, this is the first study evaluating the effect of syndemics on QoL. The main hypothesis of the study is that individuals presenting multiple syndemic conditions would have worse QoL. Thus, the main objective of the present study is to evaluate the association between the presence of syndemics and QoL among PLWHA.

MethodsStudy design
A secondary analysis of the data collected in the project "Quality of life in a cohort of people living with HIV/ AIDS" was performed (Castro et al., 2018;Machado et al., 2017). Briefly, the project was a cross-sectional study conducted between 2014 and 2017. It included a convenience sample (n = 1588) of adults under care for HIV at the Evandro Chagas National Institute of Infectious Diseases, Oswaldo Cruz Foundation, Rio de Janeiro, Brazil.

Outcomes
QoL was assessed using the WHOQOL-HIV-BREF (O'Connell & Skevington, 2012;WHO, 2002). Developed by the WHO, this scale is widely used (WHO, 2012) and validated for use in Brazil (Silveira et al., 2019;Zimpel & Fleck, 2007). The WHOQOL-HIV-BREF has 31 questions with response options ranging from 1 to 5 (Likert scale) and divided into six domains: physical, psychological, level of independence, social relationships, environment, and spirituality/religion/ personal beliefs (WHO, 2002). The score for each domain is calculated according to the formulas provided by the WHO (Orley, 1996) and ranges from 0 to 20. The higher the score, the better the QoL. In the present analysis, each WHOQOL-HIV-BREF domain was considered an outcome.

Independent variables and respective measurement criteria
Demographic characteristics included age group (dichotomized around the median age of the sample: 43 years), gender, race, educational level, and marital status (living with a partner or not). Transmission routes were categorized as homosexual, heterosexual and other (injectable drug use, blood transfusion, and other/non-specified). Self-rated health (SRH) was assessed according to the National Health Survey (5point Likert scale ranging from very poor to very good) (Jylhä, 2009). The clinical variables included the self-reported HIV stage (asymptomatic or symptomatic/AIDS), viral load (VL) (dichotomized into undetectable (less than or equal to 50 copies per ml) and detectable (WHO, 2007)), and CD4+ T cell count (CD4) (categorized into 200 cells per mm³, 200-500 cells per mm³, and greater than or equal to 500 cells per mm³ (WHO, 2007)). The VL and CD4 closest to the project interview were used within a ± 365 days window. If there was no VL or CD4 within this window, we considered the data missing, and excluded those individuals from the analysis.
Binge drinking, CSB, polysubstance use, IPV, and positive screening for depression were considered syndemic conditions. Binge drinking (consumption of four doses of alcohol by women or five doses by men within approximately 2 h) was measured according to the National Institute on Alcohol Abuse and Alcoholism (NIAAA, 2004). CSB was measured using the Sexual Compulsivity Scale (Kalichman et al., 1994;Kalichman & Rompa, 1995), which has been validated in Brazil (Scanavino et al., 2013) and whose positive cutoff point was set at 24. Polysubstance use was assessed using the Alcohol, Smoking and Substance Involvement Screening Test developed by the World Health Organization (WHO-ASSIST) (Henrique et al., 2004;WHO, 2016). The use of two or more substances was considered polysubstance use. IPV was considered present when the answer was positive to one of the following questions: Has a sexual partner tried to or physically hurt you? And Has a partner used physical force or verbal threats to force sexual relations? Screening for depression was considered positive with a score greater than or equal to 3 on the Patient Health Questionnaire-2 (PHQ-2) (Kroenke et al., 2003).
Each of the above conditions was given a score of 0 (absent) or 1 (present), and the values were added to indicate the number of syndemic conditions presented by each individual. The subjects were then dichotomized into individuals with zero or one vs. two or more syndemic conditions, and the latter were considered the syndemic group.

Statistical analysis
The scores of each WHOQOL-HIV-BREF domain were summarized using the mean and standard deviation and compared between the categories of each of the independent variables using Student's t-test for independent samples or ANOVA.
For each outcome (domain), two linear regressions were performed. The first included each of the syndemic conditions individually (binge drinking, CSB, polysubstance use, IPV, and depression). The second contained the dichotomized syndemic variable and the variables that, given their P-value <0.2 in the bivariate analysis, were included as possible confounding factors. Before this, the presence of multicollinearity was tested by inspecting the correlation coefficients and tolerance values (variance inflation factor). For each model, the goodness of fit was measured by the F test of global significance, the R2, the independence of the residuals (Durbin-Watson test), and graphical analysis of the residuals. The data were analyzed with SPSS Statistics version 19.

Results
Of the 1588 individuals interviewed, 58 were excluded because they had no VL or CD4 measured within 365 days from the date of the interview. Thus, analytical sample included 1530 participants.
Being male was associated with higher WHOQOL-HIV-BREF scores (i.e., Better QoL) in all domains except the spiritual domain. The following variables were associated with higher scores in the psychological, social, and environment domains: homosexual transmission and "other" transmission routes, age up to 43 years, and educational level of secondary and up. Individuals who lived with a partner had higher scores in the independence, social, and environment domains. Good and very good SRH were associated with higher scores in all domains (all p < 0.001), as well as the asymptomatic stage (does not apply to the physical domain). Regarding the clinical variables, undetectable VL was associated with higher QoL scores in the psychological, independence, social, and environment domains. CD4 < 200 was associated with higher scores in the physical domain and lower scores in the spiritual domain.
Of the syndemic variables, depression was associated with lower scores (i.e., Worse QoL) in all domains, with the exception of the spiritual domain, which showed the opposite pattern. Binge drinking and CSB were associated with lower QoL scores in the physical domain. Polysubstance use and IPV were associated with lower scores in the psychological, social, and environment domains. The presence of syndemics (two or more syndemic conditions simultaneously) was associated with worse QoL scores in the psychological, social, environment, and independence domains ( Table 1).
The results of the fitted linear regressions that included the syndemic conditions separately are shown in Table 2. Binge drinking was associated with worse QoL in the physical domain; polysubstance use with worse scores in the social relationship and environment domains; IPV with worse scores in the environment domain; and depression with worse scores in all domains except the spiritual domain.
The presence of syndemics was associated with worse QoL in the psychological, social relationship, and environment domains and better QoL in the spirituality domain (Table 3).

Discussion
In this cross-sectional study that evaluated QoL and syndemics among 1530 PLWHA under treatment, the syndemic conditions most frequently observed were binge drinking (56%), IPV (13%), and depression (9%). According to our hypothesis, both individual syndemic conditions and syndemics were associated with worse QoL. Positive screening for depression was associated with worse scores in all domains; polysubstance use, IPV, and binge drinking were associated with worse scores in specific domains. The presence of syndemics led to worse scores in the psychological, social, and environment domains.
When considering the syndemic conditions separately, the environmental domain was the most affected, and worse QoL in this domain was reported among individuals experiencing IPV, multiple-drug use, and depression. In previous studies conducted in Brazil, this domain was also one of the most affected (da Silva et al., 2013;Hipolito et al., 2017;Quadros Coelho et al., 2015;dos Santos et al., 2007). Notably, the environmental domain includes questions about safety, physical environment, financial resources, and opportunities for acquiring new information and skills. Our results from the multivariate analysis showed that individuals not living with a partner and having lower educational attainment presented worse QoL in the environmental domain. It is possible that such characteristics are related to lower-income. Other authors have showed that the environmental domain is influenced by socioeconomic factors, such as low income, and lower educational level (Hidru et al., 2016;Hipolito et al., 2017;Quadros Coelho et al., 2015;Reis et al., 2012). A study including PLWHA from Burkina Faso, a country in which 46.5% of people live below the poverty line, found the lowest scores in the environment domain (Bakiono et al., 2015), and the same was found in Ethiopia (Deribew et al., 2013(Deribew et al., , 2009). In addition, another study showed that half of the PLWHA experienced income reduction after an HIV diagnosis, which was associated with lower QoL (Mahalakshmy et al., 2011).
Polysubstance use and depression were negatively associated with the score on the social relationship domain, which includes questions about social inclusion, personal relationships, sexual activity, and social support. In the present study, not living with a partner, age 43 or older, and symptomatic disease stage were also associated. These findings may be associated with situations of stigma and discrimination faced by PLWHA (Bellini et al., 2015;Mahalakshmy et al., 2011;Suleiman et al., 2020). Furthermore, loneliness and lack of support from relatives were associated with a lower global QoL score in the study conducted in Burkina Faso (Bakiono et al., 2015). Older age has also been associated with fewer social interactions and a smaller social network (Domènech-Abella et al., 2017). Feeling alone, social stigma and being discriminated against are associated with depression and substance abuse (Mahalakshmy et al., 2011). Family counseling (Mahalakshmy et al., 2011) and the strengthening of the social support network (Suleiman et al., 2020) may be important strategies to improve the care of PLWHA.
The psychological domain covers negative and positive feelings, concentration, self-esteem, and self-image. The presence of depression was the only syndemic condition associated with worse QoL in this domain, being the domain that best negatively correlates with the Beck Depression Inventory (Beck et al., 1997;Silveira et al., 2019). In a study conducted in Sweden, a country with one of the best HIV treatment outcomes, hopelessness, a component associated with depression, negative selfimage and social stigma were associated with lower QoL (Zeluf-Andersson et al., 2019). Although HIV symptoms are decreased by cART, stigma, discrimination and uncertainties remain, affecting PLWHA psychologically (Schönnesson, 2002). In addition, depression, anxiety, and other psychological factors are associated with adherence to HIV treatment (Betancur et al., 2017), so providing adequate screening and treatment for PLWHA might improve their QoL.
The limitations of this study include the non-probabilistic nature of the sample, which prevents the generalization of data to other populations of PLWHA. Some participants were excluded because they did not have VL or CD4 information near enough to the date of the interview, and there is a strong possibility that this lack of tests is not random. However, a small proportion of participants were in this situation, mitigating any selection bias. Given the cross-sectional design, it is not possible to infer causality, and it is not possible to discard reverse causality. For example, worse QoL in the environment domain could be the cause of depression and not the reverse. Although we used validated instruments, the data are self-reported, and there may have been measurement and social desirability biases. Lastly, the instruments that evaluate mental health are only screening tools, and a more detailed evaluation would be necessary to diagnose psychiatric disorders. This limitation may have led us to overestimate the prevalence of depression and substance abuse in this sample.
Despite these limitations, there is evidence that a better QoL may influence patient adherence to care and thus viral suppression achievement (Bhatta & Liabsuetrakul, 2017;Degroote et al., 2014), which are necessary conditions for survival among PLWHA. Our results show that syndemic conditions mostly relate to psychological, social, and environment domains of QoL. This is important regarding patient-centered outcomes and should be considered when designing health care and public health interventions targeting PLWHA.
Ethical aspects and consent to participate in the study The project was approved by the Research Ethics Committee of Evandro Chagas National Institute of Infectious Diseases, Oswaldo Cruz Foundation (INI/ FIOCRUZ) (CAAE 17844113.2.0000.5262). All participants were explained about the study verbally, read and signed an informed consent form before answering the research questionnaire. The procedures performed were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Availability of data and materials
Raw data in the format of .csv file may be obtained with Dr. Castro upon reasonable request.

Disclosure statement
No potential conflict of interest was reported by the author(s).