We sent questionnaires to the sponsors of large-scale clinical trials. A 'sponsor' was defined as 'an individual, company, institution or organization that takes responsibility for the initiation, management and/or financing of a clinical trial' [13–15]. We targeted our questionnaires to the chairperson, or, when the sponsor was an institution or organization, to the individual with the most responsibility according to the official disclosure of the organization.
Between August 2007 and December 2007 we sent questionnaires to 90 large-scale clinical trial programs that commenced before February 2007. One-hundred and seventeen trials were found by 1) PubMed, 2) Ichushi (Japana Centra Revuo Medicina, URL: http://login.jamas.or.jp/), 3) websites of related medical societies, 4) University Hospital Medical Information Network (UMIN) Clinical Trials Registry (URL: http://www.umin.ac.jp/ctr/index-j.htm), and 5) clinicaltrials.gov (URL: http://www.clinicaltrials.gov/) on 25 February, 2007. Of 117 trials, we found addresses of sponsors for 90 trials, and we sent questionnaires to all sponsors whose addresses could be identified. In addition, 35 more trials were found using the same data sources as mentioned above on 25 July, 2009. We found addresses for sponsors of 29 of these 35 trials, and, sent questionnaires to all sponsors whose addresses could be identified between July 2009 and August 2009.
We defined 'large-scale clinical trials' as trials where the number of participants was 300 or more. If a trial was ongoing beyond the cut-off date and its planned number of participants was 300 or more, this trial was also regarded as a 'large-scale clinical trial'. We sought to include all large-scale clinical trials that examined cardiovascular and metabolism disease (i.e. cardiovascular, cerebrovascular events, and/or death) and used true endpoints as their primary endpoints. A true endpoint was an endpoint involving cardiovascular events, such as myocardial infarction, chronic heart failure, ischemic heart attack, and/or death. We sent questionnaires to the sponsors of all clinical trials that met the above criteria, except for sponsors that we did not have access to.
Our questionnaires consisted of categorical choices (Additional file 1 Figure S1). Sponsors were asked to return the questionnaires within 2 weeks, but all responses were included in our analyses, regardless of when they were returned. We analyzed all responses after checking the accuracy and appropriateness of the data, without specifying the trial or each sponsor.
We presented mean, standard deviation, median and IQR for the continuous variables, absolute numbers and percentages for binary and categorical variables. The statistical tests were used for the comparisons between the groups by funding source. P-values were calculated using Kruskal-Wallis test, and p < 0.05 was considered to be statistically significant. Data were statistically analyzed using software (R version 2.13.1).
We did not obtain an ethical approval for this study, because our study was not a medical research involving human subjects to understand the causes, development and effects of diseases and improve preventive, diagnostic and therapeutic interventions.