Background: Several studies have examined reporting of subgroup analyses in randomized trials. None have, however, addressed study characteristics associated with reporting versus not reporting of subgroup analyses. Objectives: To investigate, in randomized trials, the extent to which investigators reported subgroup analyses and claimed subgroup effects, and the study characteristics associated with reporting of subgroup analyses. Methods: We searched MEDLINE for randomized trials published in 2007 in 108 Core Clinical Journals defined by the National Library of Medicine covering all clinical and public health fields. Using standardized, piloted forms with detailed instructions, two reviewers independently screened and abstracted data. We used univariable and multivariable regression to examine the association of five pre-specified study characteristics with reporting versus not reporting of subgroup analyses. Results: Of 3662 citations identified, we randomly selected 700 reports. Our interim analysis (complete analysis will be available for the Colloquium) included 283 trials, of which 130 (45.9%) reported subgroup analyses, and 55 (42.3% of those reporting subgroup analyses) claimed subgroup effects (Table 1). Analyses showed that high impact journals and medical trials were independently associated with reporting of subgroup analyses (Table 2). A test of the interaction between funding source and significance of the primary outcome suggested that, among trials funded by private-for-profit organizations, study investigators reported subgroup analyses more frequently when the main effect was not statistically significant (OR 2.93, 95%CI 1.22–7.20), whereas no difference was present in trials without funding from private-for-profit organizations (OR 0.79, 95%CI 0.38–1.64, interaction p = 0.025). Conclusions: Subgroup analyses are frequently reported in randomized trials, and subgroup effects frequently claimed. Medical trials and those published in high impact journals were more likely to report subgroup analyses. Trials funded by private-for-profit organizations were more likely to report subgroup analyses if the result was not statistically significant for the primary outcome than if it was significant.