Customize your JAMA Network experience by selecting one or more topics from the list below.
Cruz ML, Xu J, Kashoki M, Lurie P. Publication and Reporting of the Results of Postmarket Studies for Drugs Required by the US Food and Drug Administration, 2009 to 2013. JAMA Intern Med. 2017;177(8):1207–1210. doi:10.1001/jamainternmed.2017.1313
Advancing scientific knowledge depends on the timely dissemination of research results, including those from medical product research conducted after regulatory approval for marketing. For certain drugs, the US Food and Drug Administration (FDA) is authorized to require postmarket studies, including submission of status reports and final results. Sponsors of certain applicable clinical trials1 are also required to publish summary results at the ClinicalTrials.gov website (https://clinicaltrials.gov/). Amid increasing calls for transparency in research,2 we describe publication rates and characteristics of postmarket drug studies.
On April 19, 2016, we queried an internal FDA database to identify all reportable3 postmarket drug studies designated by the FDA between 2009 and 2013 as having been “fulfilled,” denoting that the agency had received a final study report and notified the sponsor that the requirement had been met. Between April and July 2016, we searched for corresponding publications, using PubMed, Embase, Google Scholar, and the ClinicalTrials.gov website.
We categorized the postmarket studies by application type, regulatory authority, study type, and drug class. We further categorized interventional clinical trials by clinical focus; outcomes for trials with a primary efficacy end point were characterized as favorable or unfavorable. We used public data sources to identify sponsor size and ownership. With descriptive statistics, we examined study, drug and sponsor characteristics, and the publication frequency of study results at either the ClinicalTrials.gov website or in the scientific literature. Using logistic regression, we evaluated associations between publication rates for interventional clinical trials and study characteristics. The FDA Research Involving Human Subjects Committee waived approval because this was not considered human subjects research.
As of July 2016, 183 of the 288 postmarket studies (63.5%) meeting inclusion criteria were published in either the scientific literature or on the ClinicalTrials.gov website (Table 1); more studies were published in journals (175 [60.8%]) than in the trial registry (87 [30.2%]). Interventional clinical trials (n=183) had a higher overall publication rate (87.4%) than all other study types combined (n=105 [21.9%]) (Fisher exact test; P < .001). Sponsors of interventional clinical trials were predominantly industry-affiliated, large, and publicly traded. Of 69 interventional clinical trials focused on efficacy, 82.6% were classified as having results favorable to the sponsor; studies with positive results (n = 57) were no more likely to be published than those with negative results (n = 12) (Fisher exact test; P = .17).
In multivariate analysis (Table 2), publication of interventional clinical trials on the ClinicalTrials.gov website was positively associated with safety and efficacy end points (odds ratio [OR], 9.68; 95% CI, 2.04-46.03), and original drug applications (OR, 2.70; 95% CI, 1.20-6.16); approval under Accelerated Approval (OR, 0.12; 95% CI, 0.03-0.50) was negatively associated. Although no significant predictors were found for publication in the scientific literature, all drug postmarket studies with nonindustry sponsors (n = 6) and those authorized under Accelerated Approval (n = 33) were published in journals.
In keeping with previous findings, our analysis demonstrates that postmarket study results are not consistently disseminated, either through journal publication or trial registries. The overall publication rate found here (63.5%) is lower than a previous finding that 86% of pivotal clinical trials supporting drug approvals are published,4 but higher than the 51% publication rate reported for all postmarket device studies.5 Limitations of our study included the relatively small sample size and lack of sponsor heterogeneity, which reduced our ability to identify factors predictive of publication. Although additional studies may still be intended for publication, this is unlikely because we allowed a minimum of 30 months between the date of study fulfillment and the assessment of publication.
Despite calls for data sharing and publication of all clinical trial results,2 publication rates for completed postmarket studies required by FDA remain relatively low. Although FDA publication of required postmarket studies would improve accessibility, this approach would likely require new regulations. Alternatively, increased sponsor commitment to submitting to journals, and to publishing all clinical trial results on trial registries, regardless of whether publication is legally required, may serve to promote dissemination of scientific knowledge.
Corresponding Author: Marisa L. Cruz, MD, 10903 New Hampshire Ave, Silver Spring, MD 20993 (email@example.com)
Accepted for Publication: March 14, 2017.
Published Online: May 15, 2017. doi:10.1001/jamainternmed.2017.1313
Author Contributions: Dr Cruz had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Cruz, Lurie.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Cruz.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Cruz, Xu.
Administrative, technical, or material support: Kashoki, Lurie.
Study supervision: Lurie.
Conflict of Interest Disclosures: None reported.
Disclaimer: The views expressed are those of the authors and do not necessarily represent those of the FDA.
Create a personal account or sign in to: