Network Epidemiology: A Handbook for Survey Design and Data Collection

Network Epidemiology: A Handbook for Survey Design and Data Collection

Network Epidemiology: A Handbook for Survey Design and Data Collection

Network Epidemiology: A Handbook for Survey Design and Data Collection

Synopsis

Over the past two decades, the epidemic of HIV/AIDS has challenged the public health community to fundamentally rethink the framework for preventing infectious diseases. While much progress has been made on the biomedical front in treatments for HIV infection, prevention still relies onbehaviour change. This book documents and explains the remarkable breakthroughs in behavioural research design that have emerged to confront this new challenge: the study of partnership networks. Traditionally, public health research focused on the "knowledge, attitudes, and practices (KAP)" of individuals, an approach designed for understanding health-related behaviour like seat-belt wearing and cigarette smoking. For HIV and other sexually transmitted infections, however, there are atleast two people involved in transmission. This may not seem like a big difference, but in fact it changes everything. First, it means that your risk depends on your partners -- and on their partners, and their partners: it depends on your position in the network of partnerships. Consider, forexample, the rise of infections among monogamous women. Second, it means that individuals are not free to simply change their behaviour -- condom use, or abstinence, needs to be negotiated with a partner. both the epidemiology of risk and constraints to behaviour are therefore a function of thepartnership network. And our ability to design effective prevention strategies depends on our ability to measure and summarize that network. Using the traditional research designs, you would not see this network at all -- you would only see the unconnected nodes. They key to solving this problemlies in Network Analysis, before now a relatively obscure subfield in Sociology. For empirical studies of networks to become feasible, however, many problems had to be solved. This book documents the rapid progress that has been made. It brings together eight pioneering studies that have sought to map the networks that spread infection around the world. Each chapter reviews thequestions that drove the study, the changes in methodology that were needed to implement the network survey, the mistakes and successes encountered, and the central findings that the network design made possible. An introduction provides an overview of network survey design, a glossary provides asummary of network terminology, and example questionnaires from each study provide a template for further research. This is a unique and valuable resource for the international public health research community.

Excerpt

This book is the outcome of a conference on “Partnerships and the Spread of HIV and Other Infections” sponsored by the International Union for the Scientific Study of Population (IUSSP) in Chiang Mai, Thailand in February 2000. The purpose of the conference was to synthesize a decade's worth of new empirical research that has used network analytic methods to understand the population dynamics of HIV, and make it accessible to the larger research community. Interest in social network analysis has grown rapidly among applied researchers in the population sciences—starting with epidemiologists working on the HIV/AIDS epidemic in the mid 1980s and moving quickly into many other areas in demography. Methods for network analysis have been under development during the past 50 years (Wasserman and Faust 1994), and the theory is rooted in the classic works of anthropology (e.g. Levi-Strauss 1969), but it is only in the last decade or so that this work is beginning to find a broader audience among applied researchers. A good example is the academic and popular attention now given to “small world” diffusion models (Watts 1999 ; Barabasi 2002 ; Watts 2003), the variations on the “six degrees of separation” game (see, e.g. www.cs.virginia.edu/oracle), and the range of computer viruses that regularly make their way into our email inboxes. As a result of this new attention, the pace of progress in the field of network analysis has increased, and the volume and range of new work coming out in this area is now quite remarkable.

Conducting empirical studies of networks, however, remains quite a challenge. It requires many changes in research design, and there is currently no source in the published literature that an interested researcher could turn to for a systematic introduction to these issues. The Chiang Mai conference was set up to produce such a handbook, and this is the result. The conference was explicitly organized to ensure that the presentations covered the range of issues relevant for network research: from the impact it has on data collection instruments and sampling, to the changes it requires in statistical methodology, and finally, to what we have learned from network studies in perhaps the most active research context to use these tools: the epidemic spread of HIV.

The central presentations at the conference were made by six research teams with long-standing empirical projects involving network survey research. These are not the only projects that have been undertaken in the past 10 years, but they include some of the earliest and most ambitious. They were selected to provide a good introduction to the range of survey strategies available, and to highlight the way that partnership networks span physical space, social space, and time. The projects were located in many regions of the world: multiple locations in the United States, rural and urban areas in Thailand, and several countries in sub-Saharan Africa. Each session of the

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.