Publication bias against negative results from clinical trials: three of the seven deadly sins

Johns Hopkins University, Baltimore, Maryland, United States
Nature Clinical Practice Neurology (Impact Factor: 7.64). 12/2007; 3(11):590-1. DOI: 10.1038/ncpneuro0618
Source: PubMed
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The pharmaceutical industry is a textbook example of a science based sector characterized by high R&D cost, uncertain and spillovers for which patent protection assures appropriability thus providing incentives for innovation. Indeed, Mansfield (1986) found that, absent patent protection, 60 percent of pharmaceutical inventions would not have been developed and 65 percent would not have been commercialized. Since then, the market segment for patented pharmaceutical products has gained a greater relevance than in many other industries. Moreover, patents play a crucial signalling role for venture capital and big pharmaceutical companies' investment decisions and dealmaking in markets for technologies. Against this background, the relationship between intellectual property rights, innovation and public health is at the heart of a blossoming international debate (WHO 2006). A recent empirical test of the "tragedy of the anticommons" (Heller and Eisenberg 1998) shows that, although modestly, upstream patenting in biotechnology might hinder the diffusion of scientific knowledge (Murray and Stern 2007). This paper aims at contributing to the pharmaceutical patent debate by focusing on a relatively neglected function of patents, which was already characterized by Kenneth Arrow in his seminal contribution (Arrow 1962), that is information disclosure. In particular, we focus on the effect of patent-induced disclosure of information on the dynamics of both R&D and market competition. Most of the literature on patent disclosure has highlighted the trade-offs involved by the fact that patents tend to reveal "positive" information on a firm's technological advancements that might be prone to imitation by competing firms. This paper, by contrast, identifies the crucial role patents play in providing negative information1. The value of negative information is particularly important in light of the fact that it wouldn't be otherwise available to competing firms due to the strong publication bias against negative results of clinical trials (Zarin and Tse 2008, Johnson and Dickersin 2007, Chanet al. 2004). Indeed, while information on biological and therapeutic properties of a pharmaceutical product can be easily obtained when it reaches the market, in most cases firms do not publicly disclose the reasons behind their failures, making the associated patent a fundamental source of information available. In order to address this issue, the paper relies on the combination of two large datasets: the first includes all pharmaceutical and biotechnology patents granted by the USPTO since 1965, while the second comprises firm data at the level of specific R&D projects. Once integrated, the two datasets allow a comparison of the patterns of citations received by patents associated to successful projects, i.e. those that led to a marketable product and patents associated to failed projects, i.e. projects that were discontinued in clinical trials. Citations, in turn, are taken as a measure of knowledge utilization and spillovers. One of the most interesting findings of the analysis is that, although discontinued patents do receive a lower overall number of citations, a large share of them continues to receive citations by other companies even after the project has turned out to be a failure, which suggests that the information these patents provide is relevant to competing firms. To further corroborate our results, we present two case studies, based upon throughput analysis of patent documents, related scientific literature and discussions with industry and scientific experts. The first case concerns anti-inflammatory drug development (p38 MAPK inhibitors), whereas the second one study the development of a family of anti-cancer drugs (DNA topoisomerase inhibitors). Both of them show patent disclosure's role in shaping firms' technological trajectories through the possibility of reciprocal monitoring they open up in a context characterized by parallel research efforts.
    Economics of Innovation and New Technology 07/2008; DOI:10.1080/10438590802547183
  • [Show abstract] [Hide abstract]
    ABSTRACT: Illumination-sensitive failure in integrated circuits (IC) has generated considerable interest in the failure analysis community, and powerful techniques such as OBIC and LIVA have emerged (Cole et al, 1995; Wills et al, 1990). However, these techniques required the use of a new generation of SOM-based equipment which may be inaccessible to some failure analysis laboratories. This paper presents an alternative to the SOM approach to the study of a illumination-sensitive and transient I<sub>cc</sub> failure. By using emission microscopy, FIB isolation, and mechanical probing, coupled with an in-depth circuit analysis, the effort leads to a full understanding of the failure mechanism
    Physical and Failure Analysis of Integrated Circuits, 1999. Proceedings of the 1999 7th International Symposium on the; 02/1999
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Research articles reporting positive findings in the fields of orthopedic and general surgery appear to be represented at a considerably higher prevalence in the peer-reviewed literature, compared to published studies on negative or neutral data. This "publication bias" may alter the balance of the available evidence-based literature and may affect patient safety in surgery by depriving important information from unpublished negative studies. A comprehensive review of all published articles in a defined 7-year period was performed in 12 representative journals in the fields of orthopedic and general surgery. Every article published in all volumes of these journals between January 2000 and December 2006 was reviewed and rated by three investigators. Rating of articles was performed according to a uniform, standardized algorithm. All original articles were stratified into "positive", "negative" or "neutral", depending on the reported results. All non-original papers were excluded from analysis. A total of 30,197 publications were reviewed over a 7-year time-period. After excluding all non-original articles, a total of 16,397 original papers were included in the final analysis. Of these, 12,251 (74%) articles were found to report positive findings, 2,709 (17%) reported negative results, and 1,437 (9%) were neutral. A similar publication pattern was found among all years and all journals analyzed. Altogether, 91% of all original papers reported significant data (positive or negative), whereas only 9% were neutral studies that did not report any significant findings. There is a disproportionately high number of articles reporting positive results published in the surgical literature. A bias towards publishing positive data will systematically overestimate the clinical relevance of treatment effects by disregarding important information derived from unpublished negative studies. This "publication bias" remains an area of concern and may affect the quality of care of patients undergoing surgical procedures.
    Patient Safety in Surgery 02/2007; 1(1):4. DOI:10.1186/1754-9493-1-4