Second, primary care needs to have ongoing connections with local public health departments that can share data on local patterns of morbidity and mortality that can help shape clinical interventions. Given that most practices in the United States are not explicitly linked with populations, a useful heuristic to create connections among primary care, public health, and communities is the concept of a “practice footprint.” Most practices serve a panel of patients drawn from nearby neighborhoods; although usually corresponding poorly with any traditionally defined community, the panel nevertheless reflects the practice's coverage—its footprint—in a set of one or more neighborhoods. Thus, in an era when many local public health departments track risk factors, diseases, and preventive services with sophisticated geographic information systems (specifying, for example, which dwellings contain nonimmunized children), such information can be used to define the intersection of a practice's community footprint with a set of epidemiologic risks, thereby guiding action. For instance, a practice with many patients from a ZIP code with a high incidence of cervical cancer could be prompted to review its performance in screening women from those ZIP codes and to contact women who had not been screened. To define its footprint, a practice would provide the health department with a set of addresses from administrative data, and the health department would produce a report of the most urgent morbidity in the areas reached by the practice.