Quality of pharmaceutical industry press releases based on original research

Abstract

Background: Press releases are a popular vehicle to disseminate health information to the lay media. While the quality of press releases issued by scientific conferences and medical journals has been questioned, no efforts to assess pharmaceutical industry press releases have been made. Therefore, we sought to systematically examine pharmaceutical company press releases about original research for measures of quality.

Methodology/principal findings: Press releases issued by the ten top selling, international pharmaceutical companies in the year 2005 were selected for evaluation. A total of 1028 electronic press releases were issued and 235 were based on original research. More than half (59%) reported results presented at a scientific meeting. Twenty-one percent of releases were not explicit about the source of original data. While harms or adverse events were commonly cited (76%), study limitations were rarely noted (6%). Almost one-third (29%) of releases did not quantify study results. Studies presented in abstract form were subsequently published within at least 20 months in 53% of cases.

Conclusions: Pharmaceutical company press releases frequently report basic study details. However, readers should be cautioned by the preliminary nature of the data and lack of identified limitations. Methods to improve the reporting and interpretation of drug company press releases are desirable to prevent misleading media coverage.

Figure 1. Percentage of scientific meeting abstracts published within 20 months.

“High-impact journal” was classified as a journal within the top 10-impact factor rating in the relevant specialty area (e.g. cardiovascular, oncology), general medicine category or research and experimental medicine category. Low-impact journals were not in any of the 3 top lists.