Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
JAMA. Author manuscript; available in PMC 2013 June 20.
Published in final edited form as:
PMCID: PMC3688082

A Model for Dissemination and Independent Analysis of Industry Data

Harlan M. Krumholz, MD, SM and Joseph S. Ross, MD, MHS


Each day, patients and their physicians make treatment decisions with access to only a fraction of the relevant clinical research data. Many clinical studies, including randomized clinical trials, are never published in the biomedical literature.1,2 Among those that are published, key information is often not presented, such as data on specific outcomes and safety endpoints.3,4 Moreover, patient-level data from clinical trials are rarely available, leaving investigators to conduct meta-analyses of summary-level data, an approach with limitations.5

Current clinical research standards lack sufficiently strong requirements for transparency and availability. There are no uniform international standards requiring that study protocols, statistical analysis plans, and study results be made available, nor that completed clinical trial data be posted for independent analysis. Even data submitted to the U.S. Food and Drug Administration are not made publicly available.

Selective Publication and Data Out of Public View

The availability of comprehensive clinical research data is particularly important when safety concerns are raised about a drug, device or other treatment strategy. This issue is germane to publicly and privately funded studies, but the public rarely has access to data from privately funded research. They often become available only through litigation.6 For example, litigation against Merck provided access to comprehensive clinical trial data that revealed the association of rofecoxib with an increased cardiovascular risk compared with placebo more than 3 years before rofecoxib was withdrawn from the market in 2004.7 At that time, little of the relevant data were available in the public domain.

The interests of industry and the public are not always aligned when concerns arise about the safety or effectiveness of a strategy or intervention that has already been marketed. The public has a compelling interest in having the entirety of the data available for independent analysis and validation. However, industry has legitimate reservations about releasing comprehensive clinical research data for independent analyses. These include concerns about inappropriate analyses or false positive findings in the course of data mining that may needlessly worry patients or serve as fodder for frivolous lawsuits and opportunities for competitors to gain advantage.

A Way Forward

An approach that serves the needs of society and respects the concerns of industry is needed. Such an approach must include the complete and voluntary release of patient-level clinical research data by industry for independent analysis by external investigators. Accordingly, we propose a model for independent data analysis of comprehensive clinical trial and post-market surveillance data that emphasizes independence, transparency, fairness, and reproducibility. It is designed to facilitate the release of data, ensure high-quality reviews of the evidence, and provide the public with confidence that industry is truly standing behind its data in the spirit of market competition based on science.

This model includes several critical components (Figure). First, an external coordinating organization, ideally an academic center with independence and legal resources, contracts with the pharmaceutical or medical device manufacturer, which agrees to provide access to all of its patient-level clinical research data, documentation and case report forms from clinical trials, post-marketing studies, and safety programs. These data will be de-identified. It is important that source materials, not just the databases, will be provided because they are essential to determining the quality of the studies. Second, the coordinating organization contracts with 2 qualified groups to conduct independent reviews of the data. This approach puts some distance between the company and the reviewers. The engagement of 2 highly reputable organizations provides for rigorous review by credible scientists, allows an opportunity to check the reproducibility of the findings, and further instills confidence that the effort is not being manipulated by the funding or coordinating organizations. The reviews include an assessment of the adequacy and quality of the data to assess the benefits and harms of the product. Third, the data are made available by the coordinating organization to other investigators who may wish to confirm the findings of the review organizations or conduct their own analyses. The dissemination of the data provides an assurance of complete transparency. A steering committee that consists of subject matter experts as well as patients, researchers and policymakers is selected by the coordinating organization. The participation of such individuals, who are outside of the coordinating organization, lends further credibility and transparency to the effort.

Model for Dissemination and Independent Analysis of Industry Data.

Why would industry participate in such a model? The reputations of a number of companies have been damaged by a series of incidents that undermined public confidence in them. Industry is in need of fair and impartial assessments of its data, as safety concerns may arise that are based on speculation or incomplete data. The establishment of a new model will support competition based on science and foster the pursuit of an agenda that promotes greater transparency. Moreover, there is a mounting recognition that it is untenable to withhold critical information about safety and effectiveness from investigators, patients and their doctors.

First Case Study

On August 3, 2011, Medtronic Inc. (Minneapolis, MN) agreed to be the first company to commit to this model.8 Recently, safety concerns were raised about recombinant bone morphogenic protein-2 (rhBMP-2),8 an orthobiologic used to promote bone growth and sold by Medtronic in combination with an implant to promote bone growth. To address these concerns, Medtronic has agreed contractually to release all of its clinical research data that are relevant to the use of rhBMP-2 for spinal indications. Although the company is providing financial support for these activities, it will have no influence on the selection of the investigators tasked with systematically reviewing the data, the selection of the steering committee, or the process by which the data will be analyzed or disseminated. The systematic reviews and meta-analyses will be conducted and written independently, will receive constructive feedback prior to release by the coordinating organization, Steering Committee members and peer reviewers, and will subsequently be made publicly available via the coordinating organization’s website. Yale University is solely responsible for the reviews and authorizing the release of the data.

As part of this first case study, a process will be developed to release the data after the systematic reviews. We will convene experts and stakeholders (including patients) to allow for public participation. We will focus on developing appropriate procedures for the release of comprehensive clinical data, establishing standards for the mining of clinical trial data, and identifying other best practices for the evaluation of product data. Our expectation is that there will be a standardized application procedure to obtain access to the data. The hope is to develop a process that best serves the interests of society and protects, as possible, against misuse of the data.

The goal of this initiative is to usher in an era of transparency. This model will set an ethical standard for other companies and is applicable to other products. Going forward, the biomedical community should require the availability of all research data, particularly for products that have been marketed to the public. Toward this aim, Medtronic has taken the first important step.

Supplementary Material

Online Supplementary Figure


The authors wish to acknowledge the constructive comments provided on earlier drafts by Drs. Cary Gross and Richard Lehman and Ms. Beth Hodshon.

Funding: Dr. Krumholz is supported by grant U01-HL105270-02 (Center for Cardiovascular Outcomes Research at Yale University) from the National Heart, Lung, and Blood Institute. Dr. Ross is supported by the National Institute on Aging (K08 AG032886) and by the American Federation for Aging Research through the Paul B. Beeson Career Development Award Program. Drs. Krumholz and Ross are recipients of a research grant from Medtronic, Inc. through Yale University. The funding agencies had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review or approval of the manuscript.


Conflict of Interest: Dr. Krumholz chairs a cardiac scientific advisory board for UnitedHealth. Dr. Ross is a member of a scientific advisory board for FAIR Health, Inc.


1. Eyding D, Lelgemann M, Grouven U, et al. Reboxetine for acute treatment of major depression: systematic review and meta-analysis of published and unpublished placebo and selective serotonin reuptake inhibitor controlled trials. BMJ. 2010;341:c4737. [PubMed]
2. Ross JS, Mulvey GK, Hines EM, Nissen SE, Krumholz HM. Trial publication after registration in ClinicalTrials.Gov: a cross-sectional analysis. PLoS Med. 2009 Sep;6(9) e1000144. [PMC free article] [PubMed]
3. Marciniak TA. Memorandum: Cardiovascular events in RECORD, NDA 21-071/S035. [Accessed September 7, 2011];FDA briefing document, advisory committee meeting for NDA 21071 Avandia (rosiglitazone maleate) table. 2010 Jun 14;
4. Ross JS, Madigan D, Konstam MA, Egilman DS, Krumholz HM. Persistence of cardiovascular risk after rofecoxib discontinuation. Arch Intern Med. 2010;170(22):2035–2036. [PMC free article] [PubMed]
5. Lyman GH, Kuderer NM. The strengths and limitations of meta-analyses based on aggregate data. BMC Med Res Methodol. 2005;5:14. [PMC free article] [PubMed]
6. Kesselheim AS, Avorn J. The role of litigation in defining drug risks. JAMA. 2007;297(3):308–311. [PubMed]
7. Ross JS, Madigan D, Hill KP, Egilman DS, Wang Y, Krumholz HM. Pooled analysis of rofecoxib placebo-controlled clinical trial data: lessons for postmarket pharmaceutical safety surveillance. Arch Intern Med. 2009;169(21):1976–1985. [PMC free article] [PubMed]
8. Winslow R. Medtronic funds Yale safety study; Wall Street Journal. [Accessed September 7, 2011];2011