Publication of Clinical Studies Supporting FDA Premarket Approval for High-Risk Cardiovascular Devices Between 2011 and 2013: A Cross-sectional Study

Selective publication of clinical studies supporting US Food and Drug Administration (FDA) approval of novel drugs and devices prevents patients and physicians from making informed decisions about these products.^1,2 Among 149 novel therapeutics approved by the FDA between January 2005 and December 2011, a total of 326 of 380 (85.8%) supporting pivotal studies were published.³ In contrast, less than half of all studies supporting FDA Premarket Approval (PMA) of novel, high-risk cardiovascular devices between January 2000 and December 2010 were published, and more than one-fourth of these presented results in a manner discrepant with FDA reviews.⁴ It remains unknown whether contemporary practices of disseminating medical device research have improved in the wake of the 2007 FDA Amendment Act, which expanded the registration and reporting requirements on clinicaltrials.gov to explicitly include medical devices. We therefore examined the publication and reporting of studies supporting novel, high-risk cardiovascular devices approved by the FDA between January 2011 and December 2013, which account for approximately half of all FDA PMAs,⁵ along with potential predictors of publication, including study, device, and company characteristics.

A search was done of the publicly accessible FDA PMA database (http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpma/pma.cfm) to identify all novel, high-risk cardiovascular devices approved between January 2011 and December 2013. Next, clinical studies supporting the approval were identified within the FDA documents and corresponding publications as of December 2015 and were searched for in a manner described previously,⁴ which included contacting the companies marketing these devices. A search was also made for all studies on clinicaltrials.gov. We categorized studies by type (pivotal or nonpivotal), device (FDA designated as implantable or nonimplantable, life-sustaining or non–life-sustaining, and priority or standard review), and company (public or private) characteristics. We compared the reporting of primary outcome results and study interpretation in publications and FDA reviews, categorizing publications as “concordant” if primary outcome results were an exact numerical match or differed by less than 5%. Analyses were performed using Microsoft Excel, version 14.1.0 (Microsoft Corporation).

Because our examination of trial publications did not involve human subjects, ethics committee review was not required by the Yale University Human Research Protection Program.

Between January 2011 and December 2013, the FDA approved 35 novel, high-risk cardiovascular devices via the PMA pathway. We identified 70 supporting studies (mean, 2 per device), of which 56 were published in peer-reviewed literature (56 of 70 [80%]) (Table 1); 5 additional studies reported results on clinicaltrials.gov only. Pivotal studies (34 of 38 [89%]) were more likely to be published than nonpivotal studies (22 of 32 [69%]; P = .03). Studies of devices designated as life-sustaining (33 of 35 [94%]) were more likely to be published than devices designated as non–life-sustaining (23 of 35 [66%]; P = .004). Studies funded by public companies (47 of 54 [87%]) were more likely to be published than those funded by private companies (9 of 16 [56%]; P = .006). Among published studies, 3 primary outcome results were discordant, but only 1 discordance changed the interpretation of the study’s findings; the remaining published results were concordant with results reported within FDA reviews (Table 2).

Among studies supporting FDA approval of novel, high-risk cardiovascular devices between January 2011 and December 2013, 8 of 10 reported their results in the peer-reviewed biomedical literature. Although this rate is consistent with the publication rate among pivotal trials supporting new drug approvals,³ it is substantially higher than the rate of 49% observed for trials supporting FDA-approved, high-risk cardiovascular devices between January 2000 and December 2010.⁴ Furthermore, nearly all publications in our study presented results in a manner concordant with FDA reviews, which was an improvement from prior study of medical device research results reporting.⁴

Improved publication rates, particularly among public companies, are probably a consequence of the explicit inclusion of medical devices in the 2007 FDA Amendment Act expansion of clinicaltrials.gov registration and reporting requirements. However, recent failures of high-profile cardiovascular devices may also have led patients and physicians to more frequently ask about the evidence supporting newly approved devices, prompting publication by companies seeking clinical adoption of their products.⁶ Although we found that results of the vast majority of clinical studies supporting recent FDA approval of high-risk cardiovascular devices are now being made available to the clinical and research communities, efforts to improve the accessibility of results through online FDA resources and within medical device labels may still be needed to ensure dissemination of historical studies.

Corresponding Author: Joseph S. Ross, MD, MHS, Robert Wood Johnson Foundation Clinical Scholars Program, Section of General Internal Medicine, Department of Internal Medicine, Yale University School of Medicine, PO Box 208093, New Haven, CT 06520 (joseph.ross@yale.edu).

Published Online: February 22, 2016. doi:10.1001/jamainternmed.2015.8590.

Author Contributions: Drs Phillips and Ross had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: All authors.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Phillips, Rathi.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Phillips, Rathi.

Study supervision: Ross.

Conflict of Interest Disclosures: Dr Ross reported receiving support through Yale University from Medtronic, Inc and Johnson & Johnson to develop methods of clinical trial data sharing, from the Centers of Medicare & Medicaid Services to develop and maintain performance measures that are used for public reporting, from the Blue Cross Blue Shield Association to better understand medical technology evidence generation, and from the US Food and Drug Administration to develop methods for postmarket surveillance of medical devices. No other disclosures were reported.