ObjectivesThe role of translational research in improving mental health care has been highlighted in federal policy; however, an examination of how and to what extent it has been articulated at this level has not been undertaken. The aim of this scoping review was to characterise translational research concepts in federal mental health policy.MethodsAustralian Government websites were searched for federal policy documents that made recommendations for mental health services in primary care and/or community settings. Thirty eligible documents were identified, corresponding with 25 policies. Data extraction was informed by a conceptual model of translational research involving: (1) barriers and enablers and (2) recommendations and priorities codes; each had evidence generation and evidence translation subcodes. Coded text excerpts were further categorised into topics based on content.ResultsIn total, 1951 references were coded, about three-quarters of which were 'recommendations and priorities'. More were related to evidence generation (total = 1163, 59.6%) than evidence translation (total = 788, 40.3%). Most were generic without specific recommendations for how translational research should be supported. Specific recommendations for evidence generation included the use of routine databases, lived experience involvement (e.g. co-design) and strategic responsibilities (e.g. funding, policy). Specific recommendations for evidence translation mostly referred to lived experience, quality improvement and strategic responsibilities.ConclusionWhile the value of translational research is broadly acknowledged, recommendations and priorities in federal policy should emphasise evidence translation with greater specificity about how translational research should be supported. This may further influence state policy and drive improvements in practice to improve mental health care.