Sponsors have a marketing interest to represent their products in the best light. This approach conflicts with scientific standards that require the symmetric and comparable reporting of safety and efficacy data. Selective reporting of the results of clinical trials can misrepresent the risk-benefit profile of drugs. We summarize how the sponsor represented mortality findings associated with rofecoxib in clinical trials of patients with Alzheimer disease or cognitive impairment. We reviewed documents that became available during litigation related to rofecoxib involving Merck & Co, including internal company analyses and information provided by the sponsor to the FDA. We also evaluated information in 2 published articles that reported results of these trials. In one article (reporting results of protocol 091) published in 2004, 11 "non-drug related deaths" were reported (9 deaths among 346 rofecoxib patients and 2 deaths among 346 placebo patients). In another article (reporting results of protocol 078) published in 2005, 39 deaths were reported among patients taking study treatment or within 14 days of the last dose (24 among 725 rofecoxib patients and 15 among 732 placebo patients) and an additional 22 deaths in the off-drug period (17 in rofecoxib patients and 5 in placebo patients). However, these articles did not include analyses or statistical tests of the mortality data, and the 2 articles concluded that regarding safety, rofecoxib is "well tolerated." In contrast, in April 2001, the company's internal intention-to-treat analyses of pooled data from these 2 trials identified a significant increase in total mortality (hazard ratio [HR], 4.43; 95% CI, 1.26-15.53 for protocol 091, and HR, 2.55; 95% CI, 1.17-5.56 for protocol 078), with overall mortality of 34 deaths among 1069 rofecoxib patients and 12 deaths among 1078 placebo patients (HR, 2.99; 95% CI, 1.55-5.77). These mortality analyses were neither provided to the FDA nor made public in a timely fashion. The data submitted by the sponsor to the FDA in a Safety Update Report in July 2001 used on-treatment analysis methods and reported 29 deaths (2.7%) among 1067 rofecoxib patients and 17 deaths (1.6%) among 1075 placebo patients. This on-treatment approach to reporting minimized the appearance of any mortality risk. In December 2001, when the FDA raised safety questions about the submitted safety data, the sponsor did not bring these issues to an institutional review board for review and revealed that there was no data and safety monitoring board for the protocol 078 study. The findings from this case study suggest that additional protections for human research participants, including new approaches for the conduct, oversight, and reporting of industry-sponsored trials, are necessary.