Using Knowledge of Social Networks to Prevent Human Immunodeficiency Virus Infections: The Colorado Springs Study

Privacy and Confidentiality Considerations for Collecting HIV Risk Network Data among Men who Have Sex with Men and Implications for Constructing Valid Risk Networks

Studies aiming to construct risk networks have historically collected network members' names, demographic characteristics and relational data (i.e., type, strength, duration, frequency of interaction, and HIV-related risk behaviors between the pair). Due to difficulties in constructing risk networks stemming from partner anonymity and the use of nicknames, some studies also collect network members' screen names, phone numbers, physical attributes, and scars/tattoos to assist with entity resolution. In-depth interviews with 20 men who have sex with men and transgender women in Kentucky assessed privacy/confidentiality concerns as well as accuracy/recall issues associated with providing these details. Most preferred providing alters' nicknames/first names only or a first name with the first letter of the last name. Many perceived screen names, phone numbers, and scars/tattoos to be too personal/identifying. Willingness to provide more detailed information varied by relationship type/strength, which could influence the validity of the resulting network.

HIV and Sexually Transmitted Infection Epidemic Potential of Networks of Men Who Have Sex With Men in Two Cities

BACKGROUND

The speed with which a pathogen circulates in a sexual network is a function of network connectivity. Cross-sectional connectivity is a function of network features like momentary degree and assortative mixing. Temporal connectivity is driven by partner acquisition rates. The forward-reachable path (FRP) has been proposed as a summary measure of these two aspects of transmission potential. We use empirical data from San Francisco and Atlanta to estimate the generative parameters of the FRP and compare results to the HIV/sexually transmitted infection epidemics in each city.

METHODS

We used temporal exponential random graph models to estimate the generative parameters for each city's dynamic sexual network from survey data. We then simulated stochastic dynamic networks from the fitted models and calculated the FRP for each realization, overall, and stratified by partnership type and demographics.

RESULTS

The overall mean and median paths were higher in San Francisco than in Atlanta. The overall paths for each city were greater than the sum of the paths in each individual partnership network. In the casual partnership network, the mean path was highest in the youngest age group and lowest in the oldest age group, despite the fact that the youngest group had the lowest mean momentary degree and past-year partner counts.

CONCLUSIONS

The FRP by age group revealed the additional utility of the measure beyond the temporal and cross-sectional network connectivity measures. Other nonnetwork factors are still necessary to infer total epidemic potential for any specific pathogen.

Sex, drugs, and race: how behaviors differentially contribute to the sexually transmitted infection risk network structure

OBJECTIVES

We examined how risk behaviors differentially connect a population at high risk for sexually transmitted infections.

METHODS

Starting from observed networks representing the full risk network and the risk network among respondents only, we constructed a series of edge-deleted counterfactual networks that selectively remove sex ties, drug ties, and ties involving both sex and drugs and a comparison random set. With these edge-deleted networks, we have demonstrated how each tie type differentially contributes to the connectivity of the observed networks on a series of standard network connectivity measures (component and bicomponent size, distance, and transitivity ratio) and the observed network racial segregation.

RESULTS

Sex ties are unique from the other tie types in the network, providing wider reach in the network in relatively nonredundant ways. In this population, sex ties are more likely to bridge races than are other tie types.

CONCLUSIONS

Interventions derived from only 1 mode of transmission at a time (e.g., condom promotion or needle exchange) would have different potential for curtailing sexually transmitted infection spread through the population than would attempts that simultaneously address all risk-relevant behaviors.

Quantifying the Benefits of Link-Tracing Designs for Partnership Network Studies

Difficult-to-reach populations are frequently sampled through various link-tracing based designs, which rely on interpersonal networks to identify members of the population. This article examines the substantive returns to one such multiple-link tracing design in the Colorado Springs "Project 90" HIV risk networks study. Cross-links were respondents who were targeted for enrollment because of being named as partners by at least two other respondents in the sample. We compare cross-links to other respondents on sociodemographic characteristics and network properties using bivariate and multivariate adjusted statistics. We evaluate their contributions to observed network structure by creating a set of counterfactual networks deleting the information they provided. Results suggest that the link-tracing techniques led to identifying populations that would have otherwise been missed and that their absence would have underestimated potential HIV risk by distorting epidemiologically relevant measures within the network.

Adaptive web sampling

A flexible class of adaptive sampling designs is introduced for sampling in network and spatial settings. In the designs, selections are made sequentially with a mixture distribution based on an active set that changes as the sampling progresses, using network or spatial relationships as well as sample values. The new designs have certain advantages compared with previously existing adaptive and link-tracing designs, including control over sample sizes and of the proportion of effort allocated to adaptive selections. Efficient inference involves averaging over sample paths consistent with the minimal sufficient statistic. A Markov chain resampling method makes the inference computationally feasible. The designs are evaluated in network and spatial settings using two empirical populations: a hidden human population at high risk for HIV/AIDS and an unevenly distributed bird population.

Social and geographic distance in HIV risk

OBJECTIVE

The objective of this study was to examine the relationship between social distance (measured as the geodesic, or shortest distance, between 2 people in a connected network) and geographic distance (measured as the actual distance between them in kilometers [km]).

STUDY

We used data from a study of 595 persons at risk for HIV and their sexual and drug-using partners (total N = 8920 unique individuals) conducted in Colorado Springs, Colorado, from 1988 to 1992--a longitudinal cohort study that ascertained sociodemographic, clinical, behavioral, and network information about participants. We used place of residence as the geographic marker and calculated distance between people grouped by various characteristics of interest.

RESULTS

Fifty-two percent of all dyads were separated by a distance of 4 km or less. The closest pairs were persons who both shared needles and had sexual contact (mean = 3.2 km), and HIV-positive persons and their contacts (mean = 2.9). The most distant pairs were prostitutes and their paying partners (mean = 6.1 km). In a connected subset of 348 respondents, almost half the persons were between 3 and 6 steps from each other in the social network and were separated by a distance of 2 to 8 km. Using block group centroids, the mean distance between all persons in Colorado Springs was 12.4 km compared with a mean distance of 5.4 km between all dyads in this study (P <0.0001). The subgroup of HIV-positive people and their contacts was drawn in real space on a map of Colorado Springs and revealed tight clustering of this group in the downtown area.

CONCLUSION

The association of social and geographic distance in an urban group of people at risk for HIV provides demonstration of the importance of geographic clustering in the potential transmission of HIV. The proximity of persons connected within a network, but not necessarily known to each other, suggests that a high probability of partner selection from within the group may be an important factor in maintenance of HIV endemicity.

HIV transmission and the cost-effectiveness of methadone maintenance

OBJECTIVES

This study determined the cost-effectiveness of expanding methadone maintenance treatment for heroin addiction, particularly its effect on the HIV epidemic.

METHODS

We developed a dynamic epidemic model to study the effects of increased methadone maintenance capacity on health care costs and survival, measured as quality-adjusted life-years (QALYs). We considered communities with HIV prevalence among injection drug users of 5% and 40%.

RESULTS

Additional methadone maintenance capacity costs $8200 per QALY gained in the high-prevalence community and $10,900 per QALY gained in the low-prevalence community. More than half of the benefits are gained by individuals who do not inject drugs. Even if the benefits realized by treated and untreated injection drug users are ignored, methadone maintenance expansion costs between $14,100 and $15,200 per QALY gained. Additional capacity remains cost-effective even if it is twice as expensive and half as effective as current methadone maintenance slots.

CONCLUSIONS

Expansion of methadone maintenance is cost-effective on the basis of commonly accepted criteria for medical interventions. Barriers to methadone maintenance deny injection drug users access to a cost-effective intervention that generates significant health benefits for the general population.