Background: Knowledge translation (KT) refers to methods for the implementation of knowledge beyond disseminating data. It intends to reduce the gap between the way that information is generated and its application to real-world scenarios in benefit of patients. Audiovisual materials have gained importance in KT in health and different formats have been proposed for this purpose.

Objectives: To assess the effectiveness of audiovisual materials as interventions for KT in health.

Methods: Systematic review of scientific literature. Randomised clinical trials and observational studies comparing the effectiveness of audiovisual formats for KT in general population were considered.
Learning, recalling and health outcomes were retrieved. Database search included Medline, Embase, Cochrane Central Register of Controlled Trials, LILACS, PsycARTICLES and WHO International Clinical Trials Registry. Also, a search of grey literature was conducted. Screening of references and data extraction was performed by two independent reviewers. Also the risk of bias (Cochrane Collaboration tool,) and the overall quality of evidence (the Grading Recommendations Assessment, Development and Evaluation Working Group- GRADE) were assessed.

Results: 17 studies were included. Two randomised trials compared 3D with 2D videos in oral health, 3 narrative formats in cancer and surgery, and 12 techniques in the design of educational multimedia material based on learning and psychological theories (animated agent, animation and narration, accent/voice, animation and text, music/sounds, redundant features, personification). Important heterogeneity was detected; it was related to diversity in populations, interventions and criteria for evaluating outcomes. Selection, performance and detection bias were identified as potential threats to validity (Figure 1). Quality of the evidence was rated as low.

Conclusions: Redundant elements should be avoided, although when presented, action-oriented keywords improve retention. Narrative presented in human voice with standard accent and using personalisation increases recall and comprehension, also 3D format showed better recall than 2D.