On behalf of the Centers for Disease Control and Prevention and the Agency for Toxic Substances and Disease Registry, it is my pleasure to welcome you to Atlanta and to this fifth symposium on statistical methods. Now, the director of the Centers for Disease Control and Prevention, who also is the administrator of ATSDR, Dr. Sacher, could not be here today because of other commitments, but he does send his welcome and best wishes for a most successful symposium. CDC and ATSDR as you probably know are both public health agencies. That is, we look at population of the community as the patient, identifying important community health problems, and seek to develop community interventions to address those problems. And CDC and ATSDR address a broad range of public health problems. Some of our traditional areas of interest are, of course, infectious disease, chronic disease, environmental health, occupational health, violence, injury and many others. We accomplish our mission through public health surveillance, through outbreak investigations, through programme evaluations, through research, through health communications which is increasingly receiving more emphasis. That includes issuing advisories, guidelines, recommendations, publishing the morbidity and mortality weekly report. We are involved in public health capacity building which means that we provide technical advice and training and financial assistance -not only to state and local health departments, but many other entities as well such as community based organizations, academic centres and others. Now you will note that the first four of these activities -public health surveillance, outbreak investigations, programme surveillance, research -all involve the use of statistics, and therefore, to address the broad spectrum of health concerns facing the nations, CDC has to rely also on a variety of scientific disciplines and that includes epidemiology, laboratory science, behavioural and social sciences and decision sciences. And again, statistics as you know is a vital element of these scientific disciplines. So in order for CDC and ATSDR to continue to perform effectively, our staff in both the agencies must maintain a very high level of excellence in statistical methodology, and to develop and to enhance our statistical expertise we rely in part on statistical methods symposiums like this one. Now as I mentioned, this is the fifth symposium on statistical methods. Some of the previous symposiums have focused on clustering of health events, on surveillance and most recently on meta-analysis and related issues. The topic of this symposium, small area statistics, was chosen after very careful consideration of some of the current issues facing our epidemiologist and statisticians. In this time of heightened interest by the media and the public, CDC must more than ever respond rapidly to disease outbreaks and new health threats including environmental threats and this response takes place both at the national and international level. As many of you know we are heavily involved at an international level. Now some of our data sets on some outbreak investigations, for example, are rather small, but many of them are very large data sets and they are from very diverse to geographic origins.