state agencies to promote the program. In a national study, Monahan and Scheirer (1988) found that when state agencies devoted more staff to the program and located a moderate proportion of their staff in regional offices (rather than in a central office) there was likely to be a larger proportion of school districts implementing the program. Other programs, such as the Heart Partners program of the American Heart Association (Roberts-Gray et al., 1998), have used the concept of linking agents to diffuse preventive interventions. Studies of these approaches attempt to identify the organizational policies, procedures, and priorities that permit the linking agent to successfully reach a large proportion of the organizations that might implement the health behavior program. However, the research in this area does not allow general conclusions or guidelines to be drawn.
Interorganizational networks are commonly used in community-wide health initiatives. Such networks might be composed of similar organizations that coordinate service delivery (often called consortia) or organizations from different sectors that bring their respective resources and expertise to bear on a complex health problem (often called coalitions). Multihospital systems or linkages among managed-care organizations and local health departments for treating sexually transmitted diseases (Rutherford, 1998) are examples of consortia. The interorganizational networks used in Project ASSIST and COMMIT, major NCI initiatives to reduce the prevalence of smoking, are examples of coalitions (U.S. Department of Health and Human Services, 1990).
Stage theory has been applied to the formation and performance of interorganizational networks (Alter and Hage, 1992; Goodman and Wandersman, 1994). Various authors have posited somewhat different stages of development, but they all include: initial actions, to form the coalition; the formalization of the mission, structure, and processes of the coalition; planning, development, and implementation of programmatic activities; and accomplishment of the coalition’s health goals. Stage theory suggests that different strategies are likely to facilitate success at different stages of development (Lewin, 1951; Schein, 1987). The complexity, formalization, staffing patterns, communication and decision-making patterns, and leadership styles of the interorganizational network will affect its ability to progress toward its goals (Alter and Hage, 1992; Butterfoss et al., 1993; Kegler et al., 1998a,b).
In 1993, Butterfoss and colleagues reviewed the literature on community coalitions and found “relatively little empirical evidence” (p. 315) to bring to bear on the assessment of their effectiveness. Although the use of