文章基本信息

标题：Community Coalitions as a System: Effects of Network Change on Adoption of Evidence-Based Substance Abuse Prevention
本地全文：下载
作者：Thomas W. Valente ; Chich Ping Chou ; Mary Ann Pentz 等
期刊名称：American journal of public health
印刷版ISSN：0090-0036
出版年度：2007
卷号：97
期号：5
页码：880-886
DOI：10.2105/AJPH.2005.063644
语种：English
出版社：American Public Health Association
摘要：Objectives. We examined the effect of community coalition network structure on the effectiveness of an intervention designed to accelerate the adoption of evidence-based substance abuse prevention programs. Methods. At baseline, 24 cities were matched and randomly assigned to 3 conditions (control, satellite TV training, and training plus technical assistance). We surveyed 415 community leaders at baseline and 406 at 18-month follow-up about their attitudes and practices toward substance abuse prevention programs. Network structure was measured by asking leaders whom in their coalition they turned to for advice about prevention programs. The outcome was a scale with 4 subscales: coalition function, planning, achievement of benchmarks, and progress in prevention activities. We used multiple linear regression and path analysis to test hypotheses. Results. Intervention had a significant effect on decreasing the density of coalition networks. The change in density subsequently increased adoption of evidence-based practices. Conclusions. Optimal community network structures for the adoption of public health programs are unknown, but it should not be assumed that increasing network density or centralization are appropriate goals. Lower-density networks may be more efficient for organizing evidence-based prevention programs in communities. Community coalitions are often formed to help communities mobilize resources and coordinate activities that improve the public’s health. 1 ^– 3 Conceivably, coalitions may contribute to all phases of health program delivery, from planning to implementation and sustainability. 4 ^, 5 Most important, however, may be the role of coalitions in assisting communities with identifying, planning, and subsequently adopting effective health programs. In this regard, community coalitions may be best served by the promotion of evidence-based programs—those that have been systematically evaluated and shown to be effective in changing health-related behavior. One area in which evidence-based standards and programs have been well articulated is drug abuse prevention. 6 ^, 7 Coalitions are particularly important to the delivery of drug abuse prevention programs because coalitions include constituents and prevention stakeholders from many perspectives. 8 By bringing together representatives from local government, law enforcement, education, media, parent groups, health agencies, and businesses, coalitions can provide a community forum for identifying, planning, and adopting prevention programs that would not otherwise be possible through the efforts of a single agency. Several features of coalitions affect their performance. 9 One factor is having a clearly articulated structure in which subcommittees make decisions and assign tasks. 10 ^, 11 Other factors include professional representation (whether representatives from various professions are in the coalition), the variety of key stakeholder roles represented, participation (i.e., the frequency with which members attend meetings), and membership tenure. 12 ^, 13 Notably missing in the study of coalition effectiveness is attention to the coalition’s communication network, i.e., who is connected to whom and how those connections affect outcomes. 14 ^– 16 Social network analysis has shown how social network properties affect the adoption of health-related behaviors, such as smoking 17 ^, 18 and contraceptive use. 19 ^, 20 Network analysis also is used to study inter-organizational relations, because these relations are believed to affect the delivery of health services 3 ^, 14 and are useful for creating community capacity. 3 ^, 16 By adopting a network perspective in the study of coalitions, we hope to expand the potential of social network analysis for measuring social capital. 21 ^– 23 The field of network theory and analysis is well established, but it has had little application to prevention. 24 ^– 26 Of the many different network indices, 2 may have the most potential for representing a coalition’s structure: density and centralization. For example, studies of the diffusion of innovations have shown that network density and network centralization are positively associated with faster diffusion of innovations. 26 Dense networks provide more pathways where communication about prevention programs can flow compared with sparse networks. Conversely, sparse networks may not provide enough pathways for information to be circulated throughout the coalition. Density also may facilitate diffusion, because dense networks may reflect a cohesive normative environment. 3 A network with many links is more likely to have members who share common values or beliefs. Thus, a dense network may reflect a homogenous coalition, and this homogeneity will facilitate information exchange and decisionmaking. 20 Additionally, centralized networks—those with ties directed at 1 or a few members—are expected to facilitate the adoption of evidence-based programs. Centralized networks have hubs that can disseminate information to many other members quickly. A centralized coalition has leaders who can enact decisions more readily, because they have positions of power and control. 27 Moreover, once central members in a centralized network adopt a program, they are able to locate the right coalition members to implement that program. On the basis of these findings, we expect the adoption of evidence-based practices to be greater among dense coalitions than among sparse ones, and adoption should be greater among centralized networks than among decentralized ones. However, other structural characteristics, such as whether a coalition operates as a single group or as multiple subcommittees, may mitigate these relationships. 11 In a highly structured coalition, denser networks (those with a high volume of connections) may not facilitate efficiency or progress. 28 First, lower density within a network may reflect more formal collaborations. 29 Second, although there is probably a minimum density level within a network needed for coalitions to adopt innovations, once this level is reached—particularly in structured coalitions—too much density may be a liability. Too much density within a network can create communities with too few connections to external information and resources, thus making them disadvantaged. 26 ^, 30 ^, 31 Finally, organizational studies have shown that too much density within a network can hurt performance. 32 ^, 33 Similarly, networks that are too centralized concentrate power, which may result in less shared decisionmaking and lower commitment to prevention programs among noncentral members. Centralized networks are referred to as hierarchical networks, and studies have shown that employees in hierarchical organizations feel less satisfied with their work. 34 ^– 36 Some researchers have advocated for decentralized or horizontal communication networks as being more appropriate for organizations that use electronic communication technology. 37 Therefore, although a centralized network is more efficient, 38 a decentralized one may be more empowering. Thus, the adoption of new programs may be facilitated in sparser or more decentralized networks. Steps Toward Effective Prevention (STEP) was a large prevention diffusion trial that included a community coalition intervention component. 4 ^, 6 We evaluated the effects of this intervention on changing the coalition’s network density and centralization. We also evaluated the mediating effects of network change (change in network density or centralization) on subsequent planning and adoption of evidence-based prevention programs. The intervention was designed to increase the efficiency of coalition networks in planning and implementing evidence-based prevention programs by creating an organized coalition where one had previously not existed and by creating more decentralized task work groups where there had been only a single group. At least 2 previous studies have shown that the achievement of benchmarks is significantly associated with the adoption of prevention plans. 11 ^, 39 We used network analysis methods to measure the coalition structures and explore how the dynamics of the coalition system affected the coalition’s ability to implement drug abuse prevention programs. On the basis of previous research, we hypothesized that the intervention would increase the efficiency of existing coalitions by decreasing the networks’ density and centralization, which in turn would positively affect the progress and adoption of prevention planning.