From the Erasmus Medical Center committee report regarding Poldermans D, Boersma E, Bax JJ, et al. The effect of bisoprolol on perioperative mortality and myocardial infarction in high-risk patients undergoing vascular surgery. N Engl J Med 1999;341:1789–94.

“With regard to the conduct of the DECREASE-1 study, the written documentation of the research process is largely lacking. Various explanations could be given for the discrepancies that have been noted between publications 440 and 453 [NEJM 1999], and the analysis file for publication 440. After the passage of 15 years, these explanations cannot be tested.

“There were wide differences in the memories of those involved regarding the way in which outcomes had been determined in the DECREASE-1 study. Similar to the first author of publication 453 (Dr Poldermans) [NEJM 1999], the last author claimed that these determinations were made in accordance with the stipulations defined in the protocol and reported in the publication. The members of the adverse event committee cannot confirm this.

“Regarding the decision to prematurely terminate the DECREASE-1 study, the Committee finds that this decision was not taken by the safety committee, as suggested in publication 453 [NEJM 1999], but by the three members of the executive board of the steering committee. Apart from this inaccuracy, memories of how closely the members of the safety committee were involved in the decision-making differ between those concerned.

“On the basis of these findings, the Committee is unable to confirm or dispel doubts about neither the care with which the DECREASE-1 study was conducted — and thus about the study’s integrity — nor about the reliability of its results.”

The full committee report is available at www.erasmusmc.nl/1172194/2014/4771610.