The team – whose study was published in PLOS Medicine last week – compared descriptions of analysis methods and inclusion criteria in published randomized controlled trials (RCT) with those presented in internal Pfizer documents obtained through an unrelated court case and found discrepancies.
“Compared to the internal research reports, the trial publications did not always accurately reflect what was actually done in the trial. Therefore, the trial publication could not be considered to be an accurate and transparent record of the numbers of participants randomized and analyzed for efficacy.”
Study author Kay Dickersin, director of the Center for Clinical Trials at Johns Hopkins Bloomberg School of Public Health, told Outsourcing-pharma.com that: “Our conclusions in this paper were related to the fact that definitions of whom was included in each type of analysis performed were not standardized. Consequently use of the terminology is confusing and may even obfuscate.”
In light of the findings Prof. Dickersin and her colleagues called for revision of the CONSORT statement, which is a checklist of information that should be included in published trials to prevent bias in results interpretation.
The suggested that asking study sponsors and authors to detail the criteria used to define each type of analysis and state the number of participants excluded would help to increase transparency.
Data disclosure
Publication of the study comes amid renewed criticism of the pharmaceutical industry’s publishing habits and increasing public and regulatory pressure to make all trial data public, which Dickersin acknowledged.
“There is considerable demand right now for access to original study data, both to confirm published findings and to be able to incorporate the data into meta-analyses, especially when an outcome is not publicly reported.
“Increased transparency generally enables appropriate judgments about the validity and reliability of a study’s findings. If the published data are confirmed, the public would find this reassuring. If the data are not confirmed then the public needs to be part of the discussion about the accurate presentation of findings for each specific case as well as more generally.”
With this in mind we asked Dickersin if her findings are indicative of any wider trends.
“We do not know whether the same could be said for other trials, done by other companies, since it has been rare that one gets to see both internal and external data. Additional comparisons between the internal and external reports should be possible once the EMA data become available.
“Application of our findings is relevant to all investigations and investigators, not just industry studies, and this is why we are suggesting changes to CONSORT.”
Pfizer response
Pfizer rejected the idea that its publications do not accurately reflect its research prgrammes, telling Outsourcing-pharma.com that: “Results are reported by Pfizer in an objective, accurate, balanced, and complete manner and are reported regardless of the outcome of the study or the country in which the study was conducted.”
It added that: "Regardless of the geographical location in which the study was conducted, Pfizer posts Basic Results on Clinicaltrials.gov for Pfizer-sponsored interventional studies conducted in patients that evaluate the safety and/or efficacy of a Pfizer product.”