A dynamic network model to disentangle the roles of steady and casual partners for HIV transmission among MSM

A Review of Network Models for HIV Spread

BACKGROUND

HIV/AIDS has been a global health crisis for over 4 decades. Network models, which simulate human behavior and intervention impacts, have become an essential tool in guiding HIV prevention strategies and policies. However, no comprehensive survey of network models in HIV research has been conducted. This article fills that gap, offering a summary of past work and future directions to engage more researchers and inform policy related to eliminating HIV.

SETTING

Network models explicitly represent interactions between individuals, making them well-suited to study HIV transmission dynamics. Two primary modeling paradigms exist: a mechanistic approach from applied mathematics and a statistical approach from the social sciences. Each has distinct strengths and weaknesses, which should be understood for effective application to HIV research.

METHODS

We conducted a systematic review of network models used in HIV research, detailing the model types, populations, interventions, behaviors, datasets, and software used, while identifying potential future research directions.

RESULTS

Network models are particularly valuable for studying behaviors central to HIV transmission, such as partner selection and treatment adherence. Unlike traditional models, they focus on individual behaviors, aligning them with clinical practice. However, more accurate network data are needed for better model calibration and actionable insights.

CONCLUSIONS

This article serves as a point of reference for HIV researchers interested in applying network models and understanding their limitations. To our knowledge, this is the most comprehensive review of HIV network models to date.

Interrelationship between chemsex engagement and PrEP (pre-exposure prophylaxis) experience in men who have sex with men: a community-based cross-sectional study

BACKGROUND

Chemsex engagement is known to be associated with higher-risk sexual behaviour, HIV and sexually transmitted infection (STI). To reduce HIV infection risk, pre-exposure prophylaxis (PrEP) is increasingly used in the men who have sex with men (MSM) community. This study aims to examine the interrelationship between chemsex engagement and PrEP use in MSM.

METHODS

This is a cross-sectional study with the recruitment of MSM in the community in Hong Kong. Participating MSM were invited to complete an online survey which included PrEP (experienced or naïve) use, chemsex engagement history and HIV test results. Chemsex engagement referred to use of specific drugs (methamphetamine and/or gamma-hydroxybutyrate) before and/or during sex. Characteristics of MSM with and without history of chemsex engagement were examined in bivariable and multivariable logistic regression. In a case-control analysis with a 1:1 ratio of PrEP-naïve and PrEP-experienced MSM, the association of chemsex engagement with history of PrEP use was examined in bivariable and multivariable logistic regression.

RESULTS

Between February 2022 and January 2024, 338 eligible MSM were recruited. A higher proportion of MSM engaged in chemsex reported history of PrEP use and group sex. Prevalence of ever chemsex engagement was significantly higher in PrEP-experienced (29%) than PrEP-naïve MSM (9%) (adjusted OR=2.69, 95% CI 1.28 to 5.67), after adjustment for confounders of age, history of STI testing, multiple sex partnership and inconsistent condom use. The history of chemsex engagement in the past 6 months was associated positively with ever and recent (6 months) use of PrEP. Five out of seven PrEP-naïve MSM who had engaged in chemsex in the past 6 months indicated their intention to take PrEP.

CONCLUSIONS

Chemsex engagement was positively associated with PrEP use while most of the PrEP-naïve MSM who had recently engaged in chemsex expressed interest to start PrEP. PrEP service could be an important community platform for chemsex intervention.

Marriage and Steady Relationships with Women in Men Who Have Sex with Men in Sub-Saharan Africa: A Mixed-Method Systematic Review and Meta-analyses

High HIV prevalence in Sub-Saharan African (SSA) in men who have sex with men (MSM) leads to greater risk for their wives and other steady female partners because of prolonged exposure. To provide insights into the context possibly contributing to the risk of HIV transmission from MSM to women, our mixed-method synthesis about MSM' marriage and steady relationships with cisgender women aimed to: (i) assess the extent of engagement in steady relationships with women and in risky behaviors with these women across SSA's four regions; (ii) explore the underlying dynamics within these relationships by gathering qualitative information. We used quantitative and qualitative data specifically pertaining to related to marriage or other steady relationships with women from a systematic review on men who have sex with both men and women (MSMW) in SSA (PROSPERO-CRD42021237836). Meta-analyses were performed on quantitative data for each region. Pooled proportions were calculated using random-effects models. Qualitative data were analyzed using thematic synthesis. Data were selected from 125 studies. For Southern, East and West Africa, the estimated pooled proportions of married MSM were 4% (95% CI 2-8%; n = 10 studies; 4183 MSM), 8% (6-11%; 19; 7070), and 7% (6-9%; 13; 4705). In Southern Africa, 29% (11-56%; 5; 1667) of MSM had steady female partners. In East Africa, 34% (14-61%; 5; 2003) were currently or previously married to women. Motives to marry women included a desire to have children, and to conform to heteronormative social norms and family pressure. Marriage was seen as a means to discontinue homosexual behaviors or, conversely, a way to secretly continue same-sex behaviors more freely. Procreative intentions and a desire for secrecy often deterred MSM from using HIV prevention methods with their wives. For some MSM, steady relationships with women provided them with mutual support. However, these relationships could also lead to stressful and conflict-ridden situations, potentially resulting in psychosocial and HIV-related risks for the MSM as well as their male and steady female partners. Steady relationships with women are common in MSM in SSA. Sexuality concealment strategies with steady female partners depend on the circumstances that lead MSM to enter into these relationships, and have various implications on sexual behaviors with both male and female partners. Community-based support, HIV research, prevention, and care programs should be adapted to MSM's different life situations to reduce direct HIV transmission risk to steady female partners.

Approximate inference for longitudinal mechanistic HIV contact network

Network models are increasingly used to study infectious disease spread. Exponential Random Graph models have a history in this area, with scalable inference methods now available. An alternative approach uses mechanistic network models. Mechanistic network models directly capture individual behaviors, making them suitable for studying sexually transmitted diseases. Combining mechanistic models with Approximate Bayesian Computation allows flexible modeling using domain-specific interaction rules among agents, avoiding network model oversimplifications. These models are ideal for longitudinal settings as they explicitly incorporate network evolution over time. We implemented a discrete-time version of a previously published continuous-time model of evolving contact networks for men who have sex with men and proposed an ABC-based approximate inference scheme for it. As expected, we found that a two-wave longitudinal study design improves the accuracy of inference compared to a cross-sectional design. However, the gains in precision in collecting data twice, up to 18%, depend on the spacing of the two waves and are sensitive to the choice of summary statistics. In addition to methodological developments, our results inform the design of future longitudinal network studies in sexually transmitted diseases, specifically in terms of what data to collect from participants and when to do so.

Understanding dynamics and overlapping epidemiologies of HIV, HSV-2, chlamydia, gonorrhea, and syphilis in sexual networks of men who have sex with men

Introduction

We aimed to investigate the overlapping epidemiologies of human immunodeficiency virus (HIV), herpes simplex virus type 2 (HSV-2), chlamydia, gonorrhea, and syphilis in sexual networks of men who have sex with men (MSM), and to explore to what extent the epidemiology of one sexually transmitted infection (STI) relates to or differs from that of another STI.

Methods

An individual-based Monte Carlo simulation model was employed to simulate the concurrent transmission of STIs within diverse sexual networks of MSM. The model simulated sexual partnering, birth, death, and STI transmission within each specific sexual network. The model parameters were chosen based on the current knowledge and understanding of the natural history, transmission, and epidemiology of each considered STI. Associations were measured using the Spearman's rank correlation coefficient (SRCC) and maximal information coefficient (MIC).

Results

A total of 500 sexual networks were simulated by varying the mean and variance of the number of partners for both short-term and all partnerships, degree correlation, and clustering coefficient. HSV-2 had the highest current infection prevalence across the simulations, followed by HIV, chlamydia, syphilis, and gonorrhea. Threshold and saturation effects emerged in the relationship between STIs across the simulated networks, and all STIs demonstrated moderate to strong associations. The strongest current infection prevalence association was between HIV and gonorrhea, with an SRCC of 0.84 (95% CI: 0.80-0.87) and an MIC of 0.81 (95% CI: 0.74-0.88). The weakest association was between HSV-2 and syphilis, with an SRCC of 0.54 (95% CI: 0.48-0.59) and an MIC of 0.57 (95% CI, 0.49-0.65). Gonorrhea exhibited the strongest associations with the other STIs while syphilis had the weakest associations. Across the simulated networks, proportions of the population with zero, one, two, three, four, and five concurrent STI infections were 48.6, 37.7, 11.1, 2.4, 0.3, and < 0.1%, respectively. For lifetime exposure to these infections, these proportions were 13.6, 21.0, 22.9, 24.3, 13.4, and 4.8%, respectively.

Conclusion

STI epidemiologies demonstrate substantial overlap and associations, alongside nuanced differences that shape a unique pattern for each STI. Gonorrhea exhibits an "intermediate STI epidemiology," reflected by the highest average correlation coefficient with other STIs.

Impact of voluntary testing on infectious disease epidemiology: A game theoretic approach

The World Health Organization recommends test-and-treat interventions to curb and even eliminate epidemics of HIV, viral hepatitis, and sexually transmitted infections (e.g., chlamydia, gonorrhea, syphilis and trichomoniasis). Epidemic models show these goals are achievable, provided the participation of individuals in test-and-treat interventions is sufficiently high. We combine epidemic models and game theoretic models to describe individual's decisions to get tested for infectious diseases within certain epidemiological contexts, and, implicitly, their voluntary participation to test-and-treat interventions. We develop three hybrid models, to discuss interventions against HIV, HCV, and sexually transmitted infections, and the potential behavioral response from the target population. Our findings are similar across diseases. Particularly, individuals use three distinct behavioral patterns relative to testing, based on their perceived costs for testing, besides the payoff for discovering their disease status. Firstly, if the cost of testing is too high, then individuals refrain from voluntary testing and get tested only if they are symptomatic. Secondly, if the cost is moderate, some individuals will test voluntarily, starting treatment if needed. Hence, the spread of the disease declines and the disease epidemiology is mitigated. Thirdly, the most beneficial testing behavior takes place as individuals perceive a per-test payoff that surpasses a certain threshold, every time they get tested. Consequently, individuals achieve high voluntary testing rates, which may result in the elimination of the epidemic, albeit on temporary basis. Trials and studies have attained different levels of participation and testing rates. To increase testing rates, they should provide each eligible individual with a payoff, above a given threshold, each time the individual tests voluntarily.

Estimating contact network properties by integrating multiple data sources associated with infectious diseases

To effectively mitigate the spread of communicable diseases, it is necessary to understand the interactions that enable disease transmission among individuals in a population; we refer to the set of these interactions as a contact network. The structure of the contact network can have profound effects on both the spread of infectious diseases and the effectiveness of control programs. Therefore, understanding the contact network permits more efficient use of resources. Measuring the structure of the network, however, is a challenging problem. We present a Bayesian approach to integrate multiple data sources associated with the transmission of infectious diseases to more precisely and accurately estimate important properties of the contact network. An important aspect of the approach is the use of the congruence class models for networks. We conduct simulation studies modeling pathogens resembling SARS-CoV-2 and HIV to assess the method; subsequently, we apply our approach to HIV data from the University of California San Diego Primary Infection Resource Consortium. Based on simulation studies, we demonstrate that the integration of epidemiological and viral genetic data with risk behavior survey data can lead to large decreases in mean squared error (MSE) in contact network estimates compared to estimates based strictly on risk behavior information. This decrease in MSE is present even in settings where the risk behavior surveys contain measurement error. Through these simulations, we also highlight certain settings where the approach does not improve MSE.

Staged HIV transmission and treatment in a dynamic model with long-term partnerships

The transmission dynamics of HIV are closely tied to the duration and overlap of sexual partnerships. We develop an autonomous population model that can account for the possibilities of an infection from either a casual sexual partner or a long-term partner who was either infected at the start of the partnership or has been newly infected since the onset of the partnership. The impact of the long-term partnerships on the rate of infection is captured by calculating the expected values of the rate of infection from these extended contacts. The model includes three stages of infectiousness: acute, chronic, and virally suppressed. We calculate HIV incidence and the fraction of new infections attributed to casual contacts and long-term partnerships allowing for variability in condom usage, the effect of achieving and maintaining viral suppression, and early intervention by beginning HAART during the acute phase of infection. We present our results using data on MSM HIV transmission from the CDC in the U.S. While the acute stage is the most infectious, the majority of the new infections will be transmitted by long-term partners in the chronic stage when condom use is infrequent as is common in long-term relationships. Time series analysis of the solution, as well as parameter sensitivity analysis, are used to determine effective intervention strategies.

Dynamic analysis of an HIV/AIDS treatment model incorporating MSM

Acquired immunodeficiency syndrome (AIDS) has a serious impact on human health and life safety. In order to study its related factors, this paper establishes an HIV/AIDS model with treatment individuals based on heterosexual contact and male-to-male sexual contact. Using the method of next generation matrix, the threshold [Formula: see text] of the model is given. When [Formula: see text], it proves the global stability of the disease-free equilibrium. When [Formula: see text], it studies the dynamics of the boundary equilibrium and the endemic equilibrium under different conditions. Finally, through numerical simulations, the correctness of the theoretical results is verified. The key parameters affecting the spread of HIV are found through parameter sensitivity analysis, which provides a theoretical basis for effective control of the spread of HIV.

Inter-partner Agreement of Condom Use and Sexual Positioning in Male Couples

Accurate, reliable self-reports of sexual behavior are a crucial component of valid HIV risk-estimation and behavioral intervention evaluation, yet this data's reliability remains understudied. The goal of this study was to describe interpartner agreement on recent receptive anal intercourse (AI) and condomless receptive AI frequencies, among a sample of male couples. We quantified interpartner agreement on self-reported receptive AI and condomless receptive AI (absolute and relative to AI frequency), and position and condom use during a couple's most recent AI, using cross-sectional data from male partners (US, 2016-2017; N = 718 individuals). Proportional and statistical agreement (intraclass correlation coefficients (ICC), kappa (k)) were assessed. Observed agreement for receptive AI frequency was 49.44% (ICC (95% CI): 0.82(0.79, 0.84)) and for relative receptive frequency, 59.05% (ICC: 0.96 (0.85, 0.96). Agreement on condomless receptive AI was 90.21% (ICC: 0.78 (0.75, 0.82), and for relative condomless receptive AI, 91.15% (ICC: 0.80 (0.77, 0.83). Most recent AI position agreement was 89.42% (k (95% CI): 0.84 (0.80, 0.88)), and condom use, 98.89% (k: 0.82 (0.87, 0.98)). Observed agreement was higher among those who reported consistent positioning and condom use. Further research on self-reported sexual behavior data is needed to improve research validity and intervention effectiveness.

Epidemic models with discrete state structures

The state of an infectious disease can represent the degree of infectivity of infected individuals, or susceptibility of susceptible individuals, or immunity of recovered individuals, or a combination of these measures. When the disease progression is long such as for HIV, individuals often experience switches among different states. We derive an epidemic model in which infected individuals have a discrete set of states of infectivity and can switch among different states. The model also incorporates a general incidence form in which new infections are distributed among different disease states. We discuss the importance of the transmission-transfer network for infectious diseases. Under the assumption that the transmission-transfer network is strongly connected, we establish that the basic reproduction numberR 0 is a sharp threshold parameter: ifR 0 ≤ 1 , the disease-free equilibrium is globally asymptotically stable and the disease always dies out; ifR 0 > 1 , the disease-free equilibrium is unstable, the system is uniformly persistent and initial outbreaks lead to persistent disease infection. For a restricted class of incidence functions, we prove that there is a unique endemic equilibrium and it is globally asymptotically stable whenR 0 > 1 . Furthermore, we discuss the impact of different state structures onR 0 , on the distribution of the disease states at the unique endemic equilibrium, and on disease control and preventions. Implications to the COVID-19 pandemic are also discussed.

The Rhythm of Risk: Sexual Behaviour, PrEP Use and HIV Risk Perception Between 1999 and 2018 Among Men Who Have Sex with Men in Amsterdam, The Netherlands

HIV risk perception plays a crucial role in the uptake of preventive strategies. We investigated how risk perception and its determinants changed between 1999 and 2018 in an open, prospective cohort of 1323 HIV-negative men who have sex with men (MSM). Risk perception, defined as the perceived likelihood of acquiring HIV in the past 6 months, changed over time: being relatively lower in 2008–2011, higher in 2012–2016, and again lower in 2017–2018. Irrespective of calendar year, condomless anal intercourse (AI) with casual partners and high numbers of partners were associated with higher risk perception. In 2017–2018, condomless receptive AI with a partner living with HIV was no longer associated with risk perception, while PrEP use and condomless AI with a steady partner were associated with lower risk perception. We showed that risk perception has fluctuated among MSM in the past 20 years. The Undetectable equals Untransmittable statement and PrEP coincided with lower perceived risk.

Improving pairwise approximations for network models with susceptible-infected-susceptible dynamics

Network models of disease spread play an important role in elucidating the impact of long-lasting infectious contacts on the dynamics of epidemics. Moment-closure approximation is a common method of generating low-dimensional deterministic models of epidemics on networks, which has found particular success for diseases with susceptible-infected-recovered (SIR) dynamics. However, the effect of network structure is arguably more important for sexually transmitted infections, where epidemiologically relevant contacts are comparatively rare and longstanding, and which are in general modelled via the susceptible-infected-susceptible (SIS)-paradigm. In this paper, we introduce an improvement to the standard pairwise approximation for network models with SIS-dynamics for two different network structures: the isolated open triple (three connected individuals in a line) and the k-regular network. This improvement is achieved by tracking the rate of change of errors between triple values and their standard pairwise approximation. For the isolated open triple, this improved pairwise model is exact, while for k-regular networks a closure is made at the level of triples to obtain a closed set of equations. This improved pairwise approximation provides an insight into the errors introduced by the standard pairwise approximation, and more closely matches both higher-order moment-closure approximations and explicit stochastic simulations with only a modest increase in dimensionality to the standard pairwise approximation.

A multilayer temporal network model for STD spreading accounting for permanent and casual partners

Sexually transmitted diseases (STD) modeling has used contact networks to study the spreading of pathogens. Recent findings have stressed the increasing role of casual partners, often enabled by online dating applications. We study the Susceptible-Infected-Susceptible (SIS) epidemic model –appropriate for STDs– over a two-layer network aimed to account for the effect of casual partners in the spreading of STDs. In this novel model, individuals have a set of steady partnerships (links in layer 1). At certain rates, every individual can switch between active and inactive states and, while active, it establishes casual partnerships with some probability with active neighbors in layer 2 (whose links can be thought as potential casual partnerships). Individuals that are not engaged in casual partnerships are classified as inactive, and the transitions between active and inactive states are independent of their infectious state. We use mean-field equations as well as stochastic simulations to derive the epidemic threshold, which decreases substantially with the addition of the second layer. Interestingly, for a given expected number of casual partnerships, which depends on the probabilities of being active, this threshold turns out to depend on the duration of casual partnerships: the longer they are, the lower the threshold.

Introducing pre-exposure prophylaxis to prevent HIV acquisition among men who have sex with men in Sweden: insights from a mathematical pair formation model

OBJECTIVES

Since 2017, the Public Health Agency of Sweden recommends that pre-exposure prophylaxis (PrEP) for HIV should be offered to high-risk individuals, in particular to men who have sex with men (MSM). The objective of this study is to develop a mathematical model investigating the effect of introducing PrEP to MSM in Sweden.

DESIGN

A pair formation model, including steady and casual sex partners, is developed to study the impact of introducing PrEP. Two groups are included in the model: sexually high active MSM and sexually low active MSM. Three mixing assumptions between the groups are considered.

SETTING

A gay-friendly MSM HIV/sexually transmitted infection testing clinic in Stockholm, Sweden. This clinic started offering PrEP to MSM in October 2018.

PARTICIPANTS

The model is calibrated according to detailed sexual behaviour data gathered in 2015 among 403 MSM.

RESULTS

By targeting sexually high active MSM, a PrEP coverage of 3.5% of the MSM population (10% of all high actives) would result in the long-term HIV prevalence to drop considerably (close to 0%). While targeting only low actives would require a PrEP coverage of 35% for a similar reduction. The main effect of PrEP is the reduced susceptibility, whereas the increased HIV testing rate (every third month) among PrEP users plays a lesser role.

CONCLUSIONS

To create a multifaceted picture of the effects of interventions against HIV, we need models that include the different stages of HIV infection and real-world data on detailed sexual behaviour to calibrate the mathematical models. Our findings conclude that targeting HIV high-risk individuals, within HIV risk populations such as MSM, with PrEP programmes could greatly decrease the long-term HIV prevalence in Sweden. Therefore, risk stratification of individuals is of importance in PrEP implementation programmes, to ensure optimising the effect and cost-effectiveness of such programmes.

The basic reproduction number, R0, in structured populations

In this paper, we provide a straightforward approach to defining and deriving the key epidemiological quantity, the basic reproduction number, R₀, for Markovian epidemics in structured populations. The methodology derived is applicable to, and demonstrated on, both SIR and SIS epidemics and allows for population as well as epidemic dynamics. The approach taken is to consider the epidemic process as a multitype process by identifying and classifying the different types of infectious units along with the infections from, and the transitions between, infectious units. For the household model, we show that our expression for R₀ agrees with earlier work despite the alternative nature of the construction of the mean reproductive matrix, and hence, the basic reproduction number.