Background Creation of free website registries of clinical trial databases answers the absolute necessity of transparency in human research. Sponsors make a moral commitment to put results on the website not later than one year after the study's primary completion date. Voluntary participants, the public and investigators now have access to the results. Up to now, a lot of scientific teams have worked on publication bias of clinical trial results. A few have estimated registration of results on trial registries available for free consultation by the public.

Purpose Our objective was to quantify the rate of result registration for clinical trials on an international registry attested by the FDA.

Materials and methods The authors consulted the United States registry Clinicaltrial.gov, which listed 111,427 trials for 174 different countries, on 3 August 2011.

Only Phase III and IV studies were selected. The recruitment completion date had passed and the primary completion date had passed by more than one year for the clinical trials. Data were extracted from the registry: sponsor name, sponsor category (institutional or industrial) and whether the results had been put on the website or not.

Results After selection, 12,895 trials were studied. Of them, 42% had an institutional sponsor and 58% had an industrial sponsor. Only 779 trials had posted results (6.04%). Result registration rates of institutional sponsors were 1.48% and 9.32% for industrial sponsors (p < 0. 001).

Conclusions Protocols of clinical trials are easily available for public consultation on website registries. However, our study shows poor result registration. It is a moral responsibility for sponsors to post the results of clinical trials. This commitment, set out in 1964 in the declaration of Helsinki, has been confirmed by the FDA. The low rate of result registration and negative publication bias does not improve the transparency of clinical trial results.