Scholarlink

Background <p>Syndemic theory posits that social and structural inequities, such as stigma and discrimination, enable conditions like HIV to cluster and interact, worsening health outcomes. While widely used in HIV research, most studies rely on additive indices and rarely test the structure or interconnections among syndemic factors. We compare two analytic approaches, exploratory factor analysis, and network analysis, to assess the structure of syndemic conditions influencing PrEP uptake among Latino men who have sex with men (“MSM). Our model includes childhood sexual abuse, depression, physical abuse, internalized homophobia, polydrug use, sexual orientation stigma, and racial discrimination. Few studies have explored how these factors interact in relation to Pre-Exposure Prophylaxis (PrEP), and none have compared methods to evaluate whether results differ. Our goal is to test whether factor and network analyses produce similar or divergent insights about PrEP-related syndemics, a key public health issue affecting LMSM.</p> Methods <p>This secondary analysis used data from 130 Latino men who have sex with men (LMSM) in Miami-Dade County, FL, recruited through community organizations as part of an HIV prevention study. Eligibility included age ≥ 18, Latino identity, and recent same-sex activity. Syndemic variables included depression, anxiety, internalized homophobia, sexual orientation stigma, polydrug use, childhood sexual abuse, physical abuse, and racial discrimination. Most were binary; stigma and homophobia scales were dichotomized at the median. Network analysis estimated partial polychoric correlations between variables using graphical LASSO with EBIC model selection. Nodes represent syndemic factors; edges represent regularized associations. Results were compared to an exploratory factor analysis (EFA) with oblique rotation to assess latent structure.</p> Results <p>Both network analysis and exploratory factor analysis (EFA) revealed similar patterns of interconnected syndemic factors. Network analysis identified significant associations between depression and sexual orientation stigma (b = 0.18, 95% CI [0.01, 0.35]), depression and anxiety (b = 0.62, 95% CI [0.48, 0.78]), and internalized homonegativity and discrimination (b = 0.22, 95% CI [0.01, 0.42]). Depression and anxiety were the most central nodes. EFA also grouped depression, anxiety, and internalized homonegativity on the same latent factor. The expanded model, including discrimination, showed better fit (lower BIC) and sampling adequacy (KMO = 0.80 vs. 0.60). Both methods identified overlapping clusters, with EFA capturing latent structure and network analysis highlighting direct relationships.</p> Conclusions <p>These results suggest that depression and anxiety are key syndemic indicators, not only due to their strong statistical associations and centrality within the network but also because they may drive or exacerbate other syndemic conditions. Targeting these mental health factors in interventions could reduce broader syndemic burden, potentially improving PrEP uptake and overall well-being among HIV-negative LMSM.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Examining factors and network analysis to validate and expand the HIV risk syndemic theory

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月11日

查看全文
  • Ariana L. Johnson,
  • WayWay M. Hlaing,
  • Raymond Balise,
  • Adam Carrico,
  • Mariano Kanamori

摘要

Background

Syndemic theory posits that social and structural inequities, such as stigma and discrimination, enable conditions like HIV to cluster and interact, worsening health outcomes. While widely used in HIV research, most studies rely on additive indices and rarely test the structure or interconnections among syndemic factors. We compare two analytic approaches, exploratory factor analysis, and network analysis, to assess the structure of syndemic conditions influencing PrEP uptake among Latino men who have sex with men (“MSM). Our model includes childhood sexual abuse, depression, physical abuse, internalized homophobia, polydrug use, sexual orientation stigma, and racial discrimination. Few studies have explored how these factors interact in relation to Pre-Exposure Prophylaxis (PrEP), and none have compared methods to evaluate whether results differ. Our goal is to test whether factor and network analyses produce similar or divergent insights about PrEP-related syndemics, a key public health issue affecting LMSM.

Methods

This secondary analysis used data from 130 Latino men who have sex with men (LMSM) in Miami-Dade County, FL, recruited through community organizations as part of an HIV prevention study. Eligibility included age ≥ 18, Latino identity, and recent same-sex activity. Syndemic variables included depression, anxiety, internalized homophobia, sexual orientation stigma, polydrug use, childhood sexual abuse, physical abuse, and racial discrimination. Most were binary; stigma and homophobia scales were dichotomized at the median. Network analysis estimated partial polychoric correlations between variables using graphical LASSO with EBIC model selection. Nodes represent syndemic factors; edges represent regularized associations. Results were compared to an exploratory factor analysis (EFA) with oblique rotation to assess latent structure.

Results

Both network analysis and exploratory factor analysis (EFA) revealed similar patterns of interconnected syndemic factors. Network analysis identified significant associations between depression and sexual orientation stigma (b = 0.18, 95% CI [0.01, 0.35]), depression and anxiety (b = 0.62, 95% CI [0.48, 0.78]), and internalized homonegativity and discrimination (b = 0.22, 95% CI [0.01, 0.42]). Depression and anxiety were the most central nodes. EFA also grouped depression, anxiety, and internalized homonegativity on the same latent factor. The expanded model, including discrimination, showed better fit (lower BIC) and sampling adequacy (KMO = 0.80 vs. 0.60). Both methods identified overlapping clusters, with EFA capturing latent structure and network analysis highlighting direct relationships.

Conclusions

These results suggest that depression and anxiety are key syndemic indicators, not only due to their strong statistical associations and centrality within the network but also because they may drive or exacerbate other syndemic conditions. Targeting these mental health factors in interventions could reduce broader syndemic burden, potentially improving PrEP uptake and overall well-being among HIV-negative LMSM.