The most common and enduring explanation for the way research is used (or abused or not used) in policy is the ‘two communities’ theory. According to this theory, the problematic relationship between research and policy is caused by the different ‘cultures’ inhabited by policy makers and researchers. The most common and enduring types of strategies that are put forward to increase research use in policy involve bridging or linking these ‘two communities’. This study challenges this way of thinking about the relationship between research and policy. Four case studies of national public health policy in Australia—breast cancer screening, prostate cancer screening, needle and syringe programs in the community, and needle and syringe programs in prisons—are used to present the context, events, processes, research, and actors involved in policy making. Three theories are deployed to explore the relationship between research and policy in each of the cases individually and across the cases as a whole. These theories bring different determinants and dynamics of the relationship to light and each is at least partially successful in increasing our understanding of the relationship between research and policy. The Advocacy Coalition Framework (ACF) understands the relationship in terms of a power struggle between competing coalitions that use research as a political resource in the policy process. The Policy Making Organisation Framework (PMOF) understands the relationship in terms of institutional and political factors that determine the way data is selected or rejected from the policy process. The Governmentality Framework (GF) understands the relationship in terms of the Foucauldian construct of power/knowledge that is created through discourse, ‘regimes of truth’ and ‘regimes of practices’ found in public health policy and research. This study has found that in three of the four case studies, public health policy was strongly influenced by research, the exception being NSP in prisons. In all cases, however, it is not possible to construct a robust and coherent account of the policy process or the policy outcome without considering the multifaceted role of research. When these theories are explored at a more fundamental level they support the argument that when research influences policy it is transformed into knowledge-for-policy by being invested with meaning and power. This process of transformation occurs through social and political action that mobilises ideal structures (such as harm minimisation and the World Health Organisation’s principles for evaluating screening programs) and material structures (such as medical journals and government advisory bodies) to resolve meta-policy problems (such as how to define complex public health problems in a way that makes them amenable to empirical research and practical action). This study provides good evidence that the notion of ‘research transfer’ between ‘two communities’ is a flawed way of understanding the research–policy relationship. Rethinking the relationship between research and policy involves building an enhanced theoretical repertoire for understanding this complex social interaction. This step is essential to the success of future efforts to make public health policy that is effective, just and emancipatory. This study makes a contribution to this task.