3. There are alternative strategies for addressing multiple audiences, as already described; a particular program will need to choose its approach depending on what resources it has available for creating multiple campaigns and on the observed variation across populations with regard to behaviors and promising message strategies.

4. Research with consumers is an essential aspect of all health communication interventions. Campaigns need to be committed to systematic formative and statistically projectable monitoring research among different potential segments of the population. Such research is needed to understand the target audiences within their cultural context as a basis for designing effective communication strategies. The research should prove more productive if it is driven by theory, as described in Chapter 2. For example, behavior change theory will suggest what the potential causes of behavior are, and drive the search for appropriate message strategies.

In the late 1980s, most women in the United States were not getting regular mammograms. Over the past two decades, mammography screening rates have increased significantly for women age 40 and over across all races. Although it is difficult to ascribe causal relationships, one can make the case for associating increases in screening rates with national campaigns initiated at this time. The synergy of these often opportunistic national activities as well as a multitude of local interventions track with increases over time. Government agencies, nonprofits, activists, and corporations have played a role. Although there was considerable focus on communications to women, broad, multilevel strategies addressed research, screening guidelines, access to mammography services, insurance coverage, and changes in regulation, legislation, and judicial actions (e.g., malpractice suits). The multilevel efforts targeting individual, system, and environmental changes demon-