A NATIONAL EFFORT TO MODEL AIDS EPIDEMIOLOGY Report of a workshop held at Leesburg, Virginia July 25-29, 1988 I am pleased to transmit to you the report on the Workshop, "A National Effort to Model Aids Epidemiology." For the participants and the working group leaders I would like to thank you for providing the initiative for the Workshop. As you will see from the findings and recommendations and from the chapters of the report, the objectives of the Workshop were achieved in terms of an assessment and a plan. In addition, the intensive participation at the Workshop by leading scientists in mathematics, statistics, biology, epidemiology, and the behavioral and social sciences gave evidence that the essential collaboration required to understand the spread of AIDS and HIV can be achieved. Sincerely yours, final Colen A NATIONAL EFFORT TO MODEL AIDS EPIDEMIOLOGY Report of a workshop held at Leesburg, Virginia II. The uses of modeling in the development of understanding of DATABASE MANAGEMENT AND ACCESS I. An AIDS database directory, 48 II. Support for data access by modelers and epidemiologists, 51 1 Executive Summary This is a report on the AIDS Modeling and Epidemiology Workshop, convened and organized under the auspices of the Office of Science and Technology Policy (OSTP), Public Health Service (PHS), Department of Energy (DOE), and National Science Foundation (NSF) at Leesburg, Virginia, July 25-29, 1988. The purpose of the Workshop was to examine the current status of AIDS and HIV modeling, to assess the potential benefits from mathematical and statistical analysis, and make recommendations for a program of research. Ninety scientists in the fields of mathematics, statistics, biology, epidemiology, data management, and the behavorial and social sciences participated (see Appendix C). The workshop, through six working groups, conducted the following analyses: a review of existing modeling efforts, an identification of models that could be created and the data requirements for models; an examination of the biological and epidemiological information available for modeling use and an estimate of the potential for obtaining data not now in hand; a similar examination for behavioral and sociological data; an examination of the data themselves in terms of existing and ongoing data collection, and forthcoming surveys, together with descriptions of data quality where possible; and an examination of the accessibility of data and their current and planned management. In addition, a working group considered options for organization and oversight of a new modeling effort. To assure that these group studies could receive attention from experts in each topic and also understand the needs and interactions of the other groups, the groups met in pairs. Plenary sessions were used only for progress reporting. A special cross-section group with members from each of the others studied a set of new surveys that are scheduled to begin shortly. The major findings of the workshop are 1. Mathematical modeling and statistical analysis must be brought to bear fully on the AIDS epidemic. The success of mathematical techniques in other fields of application and the presence of modern computational capability argue strongly for this added approach to AIDS and HIV analysis and projection. The modeling of the gonorrhea epidemic earlier in this decade, cost-benefit analysis of rubella in England, and predictions of the spread of Hong Kong flu have exhibited that |