Reporting characteristics of journal infographics: a cross-sectional study.

BACKGROUND: Infographics have become an increasingly popular method to present research findings and increase the attention research receives. As many scientific journals now use infographics to boost the visibility and uptake of the research they publish, infographics have become an important tool for medical education. It is unknown whether such infographics convey the key characteristics that are needed to make useful interpretations of the data such as an adequate description of the study population, interventions, comparators and outcomes; methodological limitations; and numerical estimates of benefits and harms. This study described whether infographics published in peer-reviewed health and medical research journals contain key characteristics that are needed to make useful interpretations of clinical research. METHODS: In this cross-sectional study, we identified peer-reviewed journals listed in the top quintile of 35 unique fields of medicine and health research listed in the Journal Citation Reports database. Two researchers screened journals for the presence of infographics. We defined an infographic as a graphical visual representation of research findings. We extracted data from a sample of two of the most recent infographics from each journal. Outcomes were the proportion of infographics that reported key characteristics such as study population, interventions, comparators and outcomes, benefits, harms, effect estimates with measures of precision, between-group differences and conflicts of interest; acknowledged risk of bias, certainty of evidence and study limitations; and based their conclusions on the study's primary outcome. RESULTS: We included 129 infographics from 69 journals. Most infographics described the population (81%), intervention (96%), comparator (91%) and outcomes (94%), but fewer contained enough information on the population (26%), intervention (45%), comparator (20%) and outcomes (55%) for those components of the study to be understood without referring to the main paper. Risk of bias was acknowledged in only 2% of infographics, and none of the 69 studies that had declared a conflict of interest disclosed it in the infographics. CONCLUSIONS: Most infographics do not report sufficient information to allow readers to interpret study findings, including the study characteristics, results, and sources of bias. Our results can inform initiatives to improve the quality of the information presented in infographics.