The quality of reporting of trial abstracts is suboptimal: Survey of major general medical journals

OBJECTIVE: To evaluate the quality of reporting of abstracts describing randomized controlled trials (RCTs) published in four major general medical journals. STUDY DESIGN AND SETTING: Systematic survey of published RCT abstracts, with two reviewers independently extracting data. We searched MEDLINE and identified 227 RCT abstracts published in the New England Journal of Medicine (NEJM), Journal of the American Medical Association (JAMA), British Medical Journal (BMJ), and The Lancet in the year 2006. RESULTS: Most abstracts identified the study as a randomized trial (98.7%), reported the objectives (92.5%), described the population (90.3%), detailed the intervention (81.5%), and defined the primary outcome (71.3%). Methodological quality was poorly reported: one (0.4%) described allocation concealment; 21 (9.3%) clearly specified blinding; 51 (22.5%) described intention-to-treat analysis; and 32 (14.1%) outlined losses to follow-up. Most of the abstracts reported the effect size and the confidence intervals (62.3%), but just half of them reported side effects or harms. CONCLUSION: The quality of reporting of RCT abstracts published in main general medical journals is suboptimal. Space limitations notwithstanding, with the recent recommendations from the CONSORT for Abstracts, it is expected that the transparency of abstract reporting can and should improve.