Past NEEMA Funded Projects by Topic Area – STD

The questions explored in this project all relate to understanding the interlocking relationships between HIV prevention methods (primarily pre-exposure prophylaxis, or PrEP), and STD transmission and treatment, to understand the future course of the HIV and STD epidemics for men who have sex with men (MSM) overall and by race. Because of the complex feedback and high-dimensionality of the behaviors and subgroups of interest, and the infeasibility of conducting randomized trials to probe the questions of interest, agent-based modeling is well-suited. Agent-based models of MSM have been developed to represent the epidemics in Atlanta, GA and Baltimore, MD, and are being used to assess the impact of PrEP guidelines, the role of STI clinics, HIV/STI coinfection, and expedited partner therapy (EPT) options for STI prevention.

Men who have sex with men (MSM) experience a very high burden of new and existing HIV infections and STI. This project specifically focuses on the Adolescent Sexual Minority Male (ASMM) population to create estimates and models of the impact of HIV prevention interventions, particularly HIV pre-exposure prophylaxis (PrEP), among this population. In the first phase, literature searches and data analyses were conducted in order to publish estimates of behavioral and testing parameters for ASMM. In the second phase, agent-based models are being adapted to understand the impact of PrEP-targeting strategies in the ASMM population.

Men who have sex with men (MSM), are a group highly-impacted by HIV and STI. Accurate estimates of this population are needed in order to compute measures of disease burden, yet MSM are not systematically enumerated in the US Census. The primary goal of this project is to develop methods that synthesize existing data sources to estimate the population sizes of men who have sex with men within US states. The first phase of this effort involved producing manuscripts and online tools the provided the number MSM at the US state, metropolitan statistical area (MSA), and county levels. Subsequent work has merged these estimates with HIV and syphilis diagnoses data to yield disease rates for MSM. Given stark HIV/STI disparities between demographic subpopulations of MSM, work is ongoing to extend these results to race- and age-specific population estimates.

The goal of this project is to build on existing frameworks for estimating the incidence and prevalence of STIs in the United States, with an extension of these approaches to accommodate calibration to a broader array of data sources and a statistical estimation framework that enables quantification of uncertainty in the estimates. This project will estimate the prevalence and incidence of chlamydia, gonorrhea, trichomoniasis, and syphilis in the United States over the period 2000-2014, accounting for sampling and nonsampling error including specific sources of bias in various observation mechanisms (e.g. self-reported diagnoses). Data sources that including laboratory examination results from national surveys, surveillance case reports, self-reported diagnoses and other types of information will be used. Statistical models will be developed that will be used in the synthesis of these multiple data sources, with an initial application to chlamydia that will be followed by adaptation for the other three infections. A Bayesian estimation approach is used based on a structural model that mirrors the previous approach, and an observation model that reflects the systematic and stochastic characteristics of each of the unique types of data sources.

Chlamydia remains one of the most prevalent sexually transmitted infections (STIs) in the US due to a significant asymptomatic reservoir. The goal of this project is twofold. First, it develops a new model for chlamydia transmission in the United States. A pair-formation model has been developed to explicitly account for development and dissolution of partnerships reflecting all possible permutations of infection status, and thus allowing for a more realistic accounting for different infection risks depending on the concordance or discordance of partner infection status. This modeling approach allows for examination of the impact of partner services in more detail than in standard compartmental models. The model also includes an expanded natural history of chlamydia to identify reinfection. A complex age structure guides pair formation within and across age groups, and concurrent partners outside of some pairs is allowed. Second, a pair formation model of chlamydia is used to investigate the potential strengths and limitations of different prevention strategies, namely (i) partner services, (ii) more frequent screening of those with a higher rate of chlamydia, and (iii) point-of-care testing. The pair formation model is also be used to examine the potential impact of more frequent screening among those who have previously been tested and treated for chlamydia, a group that has an elevated risk for re-infection. Combinations of prevention strategies will also be explored to identify which strategies, both existing and novel, give the largest joint benefits. Such analyses of chlamydia prevention strategies can inform policy decisions and be used to allocate resources.

This project aims to model the transmission dynamics of gonorrhea in heterosexual women and men and MSM populations in the United States. A multi-year effort to develop a suite of complementary models has been launched to capture the essential features of gonorrhea epidemiology in the US population and to assess interventions for reducing the burden of disease. The models are used to explore the impact and economics of interventions to reduce gonorrhea prevalence and burden as well as the impact of interventions in relation to patterns and trends in antimicrobial resistance. A metapopulation model of transmission at the national level helps understand current trends in gonorrhea epidemiology, including disparities in infection burden in certain population groups. The model uses a deterministic compartmental structure, and captures a range of different important stratifications including age, sex, race/ethnicity and sexual behavior. A re-parameterized meta-population model captures disparities in terms of the geographical distribution of cases and describes current trends in gonorrhea incidence in regions with different epidemic characteristics, as defined by the relative distribution of cases across different ethnic/racial and sexual orientation groups. An agent-based model of transmission in MSM captures the unique epidemiologic features observed in this population, investigates the importance of different sites of infection, and compares strategies to reduce the burden of antimicrobial resistance. A systematic review of existing gonorrhea transmission models helps guide model development and identify gaps and areas for future research.

Gonorrhea infection has a low prevalence in the general population of the United States, but disproportionate effects are seen in subgroups of the population. This project aims to explore the consequences of the unequal burden of gonorrhea in the United States. The project is being undertaken in several phases. The first builds on the general framework we are developing for estimating incidence and prevalence of STIs, based on a Bayesian estimation approach that leverages an array of different data sources and accounts for bias and measurement error in the data generation processes. This project will extend the general approach to allow estimation of outcomes in specific racial/ethnic groups. These epidemiologic estimates will be incorporated in a Markov cohort model to translate estimated disparities in incidence and prevalence of gonorrhea into estimates of cost and burden of disease by racial/ethnic groups, with the latter quantified in quality-adjusted life years lost. A subsequent phase of this project will examine the potential health and economic consequences of different intervention strategies on racial/ethnic disparities.

This project will develop and apply an agent-based transmission model of gonorrhea in men who have sex with men (MSM). This model will capture the unique epidemiological characteristics, as well as diagnostic and treatment challenges (including antimicrobial resistance emergence), associated with gonorrhea in MSM. The model will be used to investigate the importance of different sites of infection for gonorrhea transmission in MSM including how pharyngeal and rectal infections may be fueling the epidemic through asymptomatic transmission, and the potential impact of site-specific screening strategies on the ability to control the gonorrhea transmission. These analyses will form a foundation for future analyses examining the role of public health control strategies for reducing the emergence and spread of antimicrobial resistant gonorrhea.

The clinical and public health community are facing the challenge of increasingly antibiotic resistance in gonorrhea. In the search for strategies to control resistance, one proposed approach is the use of rapid diagnostics that would allow clinicians to tailor treatment to the antibiotic susceptibilities of individual infections, rather than treating empirically based on data from population-wide surveillance – in fact, efforts to develop such diagnostics are underway. However, the hypothesis motivating this strategy – that resistance emerges under antibiotic selection, so tailored therapy should prolong the utility of anti-gonococcal agents and control resistance better that the current strategy of empiric treatment – has not been evaluated. The lack of studies on this topic has been due to lack of data on several key parameters, including current extent of resistance on the population, likelihood of emergence of resistance on treatment, and fitness cost of resistance mutations. Using data from our related work to provide these parameters, the impact of tailored therapy on the slowing the emergence and spread of antibiotic resistance in gonorrhea will be assessed, with consideration of resistance to the extended spectrum cephalosporins, macrolides, and fluoroquinolones.

This project evaluates effectiveness of school-based screening for chlamydia within a variety of community settings. Our analyses are exploring the potential impact of these screening programs by modeling a range of intervention strategies in urban settings characterized by varying profiles and pre-existing levels of chlamydia screening in public high schools. Three urban school-based chlamydia screening programs will help to provide a strong empirical basis for modeling specific programs and associated outcomes. Work is currently ongoing with these program sites to review program implementation, data availability and data access. The model will incorporate the data and profile of each of participating site, and will accommodate replication of existing programs as well as generalization to enable exploration of potential outcomes under a wide range of different circumstances reflecting variation in epidemiology and existing health services coverage.

Syphilis is resurgent in the United States. The most striking epidemiological feature of the current epidemic is the disproportionate representation of men who have sex with men (MSM) among cases, with incidence also varying by geography and ethnicity. Cases of congenital syphilis have been increasing in recent years, as have rates in women, indicative of a changing epidemic. In particular, these trends suggest that the previously primarily MSM-focused epidemic has expanded into heterosexual populations. Frequent screening of at-risk individuals remains the best available tool for syphilis control, but current public health efforts have not had the desired effect in reducing the disease burden. A risk-structured transmission model will be developed to characterize the epidemiology of syphilis in the United States and evaluate the potential impact of different approaches to syphilis screening on epidemic dynamics and the health of affected individuals. The model will incorporate both MSM and heterosexual populations. Given the importance that core group composition and geography will have on epidemic characteristics, and the resultant implications for sustaining syphilis control in these different contexts, this model will be fit to outbreaks in different jurisdictions.