Transparent manuscript reporting practices in behavioral medicine research: An audit of publications in 2018 and 2008

Objectives: Concerns about rigor and replicability have led to reforms to increase science transparency. We aimed to document the use of transparent reporting practices in behavioral medicine journals in 2018 in order to inform future efforts to improve reporting practices. We also aimed to compare 2018 reporting practices to 2008. Methods: We examined a randomly selected portion of articles published in 2018 and 2008 by the four behavioral medicine journals with the highest impact factor. We excluded manuscripts that were reviews, presented qualitative data, or were purely descriptive. We coded whether articles were clear in their presentation of analyses as being primary or secondary; whether studies were registered/pre-registered; whether they used “exploratory” or a related term to describe analyses/aims; and whether they reported power analyses. Results: We identified and coded 162 manuscripts published in 2018 (87% observational and 12% experimental). Among 2018 studies, 16% were explicit in describing outcomes as primary or secondary, 51% appeared to be reports of secondary outcomes but did not use the term “secondary,” and 33% were unclear. Registration/pre-registration occurred in 14% of studies; 77.3% of registered/pre-registered studies did not report registration timing in relation to data collection, and 91% did not report which analyses were pre-registered. “Exploratory” or a related term was used to describe an aim or analysis in 31% of studies. Power analyses were reported in 8% of studies. Compared to studies from 2008 (n=120), studies published in 2008 were less likely to clearly report whether outcomes presented were primary or secondary and less likely to have been registered/pre-registered. Conclusions: Behavioral medicine stakeholders should consider strategies to increase clarity of reporting of key analysis details.