University of Cambridge > Talks.cam > MRC Biostatistics Unit Seminars > Harnessing social networks for HIV surveillance

Harnessing social networks for HIV surveillance

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Michael Sweeting.

The transmission dynamics of HIV are strongly affected by the structure of the underlying risk network. However, collecting data on sexual or injection drug use networks is notoriously difficult and costly. I will discuss how social network data may be harnessed to guide HIV surveillance, and potentially prevention, by identifying subpopulations with elevated HIV prevalence and/or risk behaviours. I will consider two sampling methods, respondent-driven sampling (RDS) and time-location sampling (TLS), which are widely used to sample at-risk populations. In these approaches, the underlying social network is typically considered as a nuisance factor, and the primary objective is to obtain design-based unbiased estimates of quantities such as HIV prevalence. In contrast, I will treat the social network as the primary outcome of interest, rather than a nuisance.

I will show how in RDS , individuals infected with HIV or other sexually transmitted infections (STIs) often recruit other infected individuals. Using Markov tree models, I will demonstrate that may be difficult to distinguish between infected individuals intentionally recruiting other infected individuals and the more likely scenario that individuals tend to recruit within specific social subpopulations with different HIV /STI prevalences. In a time-location sample of men who have sex with men (MSM) recruited from bars, I will demonstrate how specific venues structure the population in terms of quantities such as age and HIV prevalence. To address the severe problems of confounding, I will show how causal inference approaches such as marginal structural models can be used to consider whether there are venue-specific factors in HIV risk, beyond easily measured individual factors. Finally, I will consider how even very crude measures of social networks, such as the number of friends one has in a specific subpopulation, may be used as a proxy for belonging to different social subpopulations with different risk behaviors.

This talk is part of the MRC Biostatistics Unit Seminars series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2020 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity